Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.1109/ACCESS.2023.3331012

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press Ltd.  

This paper proposes a method to improve the accuracy of 3D measurement of a stereo camera by marking a measured object using a line laser. Stereo cameras are commonly used for 3D measurement, but the accuracy of 3D measurement is affected by the amount of texture. Therefore, a new measurement system combining a stereo camera and a line laser is developed. The accuracy of 3D measurement with a stereo camera is improved by using a line laser to mark arbitrary points on the measured object and measuring the marked points, regardless of the amount of texture on the measured object. Because the laser is only used to mark points on the measurement target, calibration is not required with the stereo camera. Experimental evaluation showed that our proposed method can obtain millimeters.

DOI： 10.20965/jrm.2023.p1374

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/IROS55552.2023.10341968

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

In this paper, we propose a novel algorithm for estimating road surface shapes and object heights using a fisheye stereo camera. Environmental recognition is an important task for advanced driver-assistance systems. However, previous studies have only achieved narrow measurement ranges owing to sensor restrictions. Moreover, the previous approaches cannot be used in environments where the slope changes because they assume inflexible constraints on the road surfaces. We use a fisheye stereo camera capable of measuring wide and dense 3D information and design a novel algorithm by focusing on the degree of division in a disparity image to overcome these defects. Experiments show that our method can detect an object in various environments, including those with inclined road surfaces.

DOI： 10.20965/jrm.2023.p1354

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publisher：The Japan Society for Precision Engineering  

This study proposes a measurement method for the interior of large structures for infrastructure maintenance and inspection. The 3D point cloud of a room is obtained by a light section method using sphere camera that can capture a 360-degree view of the surrounding area and a 360-degree ring laser. Specifically, the camera's orientation is determined by detecting line components in real space using the spherical camera images and comparing the distribution of these line components with the three directions in real space, based on the Manhattan world assumption. Using the camera's orientation, the 3D point cloud of the entire room is generated by integrating the point clouds obtained from multiple light sections.

DOI： 10.11522/pscjspe.2023a.0_614

CiNii Research

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/RO-MAN57019.2023.10309557

researchmap

Other Link： https://dblp.uni-trier.de/db/conf/ro-man/ro-man2023.html#YokotaMPU23

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE  

DOI： 10.1117/12.3000424

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.14865/ahi.5.1.1

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Springer Science and Business Media LLC  

Abstract

In recent years, the number of pipes that have exceeded their service life has increased. For this reason, earthworm-type robots equipped with cameras have been developed to perform regularly inspections of sewer pipes. However, inspection methods have not yet been established. This paper proposes a method for anomaly detection from images in pipes using Generative Adversarial Network (GAN). A model that combines f-AnoGAN and Lightweight GAN is used to detect anomalies by taking the difference between input images and generated images. Since the GANs are only trained with non-defective images, they are able to convert an image containing defects into one without them. Subtraction images is used to estimate the location of anomalies. Experiments were conducted using actual images of cast iron pipes to confirm the effectiveness of the proposed method. It was also validated using sewer-ml, a public dataset.

DOI： 10.1186/s40648-023-00246-y

researchmap

Other Link： https://link.springer.com/article/10.1186/s40648-023-00246-y/fulltext.html

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Robotics Society of Japan  

DOI： 10.7210/jrsj.41.187

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/sii55687.2023.10039474

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/sii55687.2023.10039364

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/sii55687.2023.10039109

researchmap

Authorship：Last author   Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：Japan Society for Precision Engineering  

DOI： 10.2493/jjspe.89.90

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method to accurately estimate the position of a UAV on satellite images by using the region information of buildings extracted by semantic segmentation for UAV pose estimation. The localization of UAV using satellite images is based on the similarity between images features using intensity information. However, the estimation accuracy of intensity information alone is degraded by changes in the seasons and moving objects. Therefore, in addition to the intensity information, the proposed meathod uses information on building areas extracted by semantic segmentation from overhead images to improve the accuracy of position estimation.

DOI： 10.1299/jsmermd.2023.2p2-f22

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Recently, in addition to the use of remote controllers and voice recognition, gesture-based methods have been studied as a way to operate home appliances. In this research, we construct an intuitive system in which home appliances are selected and operated via arm pointing. The conventional method uses 3D information of home appliances prepared in advance, but our proposed system can handle home appliances from only image without requiring their positions in advance. We conducted an evaluation experiment of the proposed system and obtained a high gesture recognition rate. In addition, a usability study also indicated good results for real world use.

DOI： 10.1299/jsmermd.2023.2p1-d03

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This study proposes a gesture-based home appliance operation system that can support multiple users. In recent years, increasing complexity of home appliances has led to interest in gesture-based operation. Many systems have been developed to operate home appliances by recognizing gestures from camera images. However, conventional systems require an operator to memorize a large number of gestures and have problems of misoperation. To solve these problems, the proposed system introduces an operator specific command space and a start gesture. The effectiveness of the proposed system was verified by experiments, with an average recognition rate of around 95 %. This ensures that the system can be used by multiple operators.

DOI： 10.1299/jsmermd.2023.2p2-d03

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose an image feedback control for an underactuated hand to grasp brittle/deformable food without damaging it. Automation using robots in food factories are being promoted to compensate labor shortages and improve productivity. Since food products vary in shape and hardness and are difficult to be grasped by conventional robots, robot hands that conform to the shape of the object have been developed. In this research, we aimed to control an underactuated hand by using a monocular camera, without sensor or marker. In our method, we detect the deformation of the object using optical flow, adjust appropriate grasping force using area information of robot hand, and detect the grasping state using the relative displacement of the object and the hand. Through the experiments, twelve types of real foods were successfully grasped without dropping and without damaging expect two foods.

DOI： 10.1299/jsmermd.2023.2p2-d23

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method for obstacle detection and height estimation using a fish-eye stereo camera. Disparity and intensity information is used. The method using only disparity information may incorrectly detect road surface as obstacles. Therefore, we obtain the intensity of the road surface by using the obstacle information obtained from the disparity and compare the intensity throughout the entire image to detect obstacles. Experiments show that the proposed method using the disparity and intensity information can detect only obstacles without detecting the road surface as obstacles. Experiments also show an improvement in the accuracy of height estimation.

DOI： 10.1299/jsmermd.2023.2a1-d07

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose an improved distance measurement method using blur for 3D measurement. Stereo disparity is often used for 3D measurement, but when a target object is too close to the sensor, it cannot be obtained properly. To solve this problem, we develop a compact range image sensor capable of close-range measurement focusing on the amount of blur using a multi-slit laser and a camera.

DOI： 10.1299/jsmermd.2023.1a2-g27

CiNii Research

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Archery is a sport in which it is important to repeat the same shooting form and to improve its repeatability. For this reason, it is essential to understand and master the basic shooting form. However, there is a shortage of instructors, and in university club activities, the burden of teaching is placed on the senior students. A possible solution to this problem is the use of a practice support system based on sports video analysis, but existing systems have a fixed positional relationship between the camera and the person to be measured, and it is necessary to create the same environment each time the system is used. Therefore, this study proposes a sports practice support system that uses 3D skeletal point information extracted from video images to evaluate the learner’s shooting style in terms of points, eliminating the need to fix the positional relationship between the camera and the learner and allowing easy recognition of points that need improvement.

DOI： 10.1299/jsmermd.2023.1a2-f07

CiNii Research

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Authorship：Last author   Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Taylor & Francis  

DOI： 10.1080/01691864.2022.2084346

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.2493/jjspe.88.276

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII52469.2022.9708745

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII52469.2022.9708600

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII52469.2022.9708838

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/SII52469.2022.9708610

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.2493/jjspe.87.959

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/UR52253.2021.9494671

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE  

DOI： 10.1117/12.2589196

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：The Japan Society of Mechanical Engineers  

In recent years, the number of pipes that have exceeded their service life has been increasing. Thus, regular inspection of the pipes is required. In particular, a flexible pipe made of vinyl chloride shows its unique degradation, such as oblatening. In this paper, we propose a method of estimating the degree of oblatening of a pipe from its internal image captured by an earthworm robot. Covariance matrix of the image is used to obtain the degree of oblatening.

DOI： 10.1299/jsmermd.2021.2p2-h14

CiNii Research

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/UR52253.2021.9494671

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.1299/transjsme.20-00310

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/ieeeconf49454.2021.9382647

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/REM49740.2020.9313886

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.2493/jjspe.86.969

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/REM49740.2020.9313929

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

DOI： 10.1109/REM49740.2020.9313074

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/UR49135.2020.9144873

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII46433.2020.9025978

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII46433.2020.9026287

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1109/SII46433.2020.9025910

researchmap

Copyright © 2019. The Authors. Published by Elsevier Ltd. All rights reserved. In this paper, we propose a method to generate a 3D map in the narrow pipeline from inside images captured from a camera mounted on earthworm robot. Pipes that have reached the end of their useful life cause leaks and pits in the road; thus, it is necessary to inspect the inside of the pipe. In order to inspect the inside of the pipe, images in the pipeline are converted to development images and motion estimation of the camera is performed using the optical flow. Finally, the development images are connected to generate a 3D map.

DOI： 10.1016/j.ifacol.2019.11.053

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.1299/transjsme.19-00087

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

DOI： 10.1007/978-981-32-9441-7_103

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

DOI： 10.1299/transjsme.19-00069

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Taylor & Francis  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Planetary exploration rovers face severe energy and safety restrictions, which have a strong connection with terrain slopes. During a steep slope traverse, a rover consumes more power and is exposed to higher risks of getting stuck or of overturning. It is essential for a rover to autonomously recognize and avoid steep slopes for efficient and safe operations. Existing techniques (e.g. stereo vision) do not completely address challenges in planetary exploration, such as low-textured terrain appearance and computational resource limitations. This paper presents a novel slope estimation method using a monocular infrared camera. The proposed method estimates slope normals based on surface temperatures on two different slopes. The surface energy model is employed to correlate thermal properties to geometrical properties of the terrain. The idea behind this approach is that the solar radiation, which is a major energy source for terrains, can differ by time, slope angles and directions. The difference in energy input generates the gap of surface temperatures between target and reference surfaces, which can be remotely detected with an infrared camera. The proposed method avoids the problem of terrain appearance as it only uses temperature measurements, and is also computationally efficient thanks to efficient preprocessing. The algorithm is validated through simulations and outdoor experiments. The results show the effectiveness of the proposed scheme to estimate terrain slope normals solely from temperature measurements.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society for Precision Engineering  

This paper proposes a stereo camera using two fisheye cameras. The measurement range of the stereo camera is increased by using fisheye cameras. Fisheye images are deformed by using an equirectangular projection to simplify the stereo correspondence search in the images. Stereo images are rectified to correct the influence of external parameters between a left camera and a right camera. Rectification parameters are estimated from feature points obtained from an image of an arbitrary environment. Errors due to deviation of the position and rotation of the left and right cameras are removed by the stereo rectification. The corresponding points are calculated by using stereo matching in the deformed image, and the distance is measured from the disparity. The performance of the fisheye stereo camera using equirectangular images is evaluated by simulations and experiments.

DOI： 10.2493/jjspe.83.1095

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：The Japan Society for Precision Engineering  

DOI： 10.11522/pscjspe.2017S.0_55

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：The Society of Instrument and Control Engineers  

In this paper, we propose a method for detecting moving objects using a moving stereo camera. First, the camera motion parameters are estimated by using optical flow with a stereo camera. Second, the optical flow occurring in the background is removed. Finally, moving objects are detected individually by labeling the remaining optical flow. The proposed method is evaluated through experiments using two pedestrians in an indoor environment.

DOI： 10.9746/jcmsi.10.344

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Springer  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

<p>This paper presents a generation method of trajectories for score-based inverse reinforcement learning. While most inverse reinforcement learning methods require demonstrations of the expert, score-based inverse reinforcement learning can estimate the expert's reward function from scores of arbitrary trajectories. This study brings active learning into score-based inverse reinforcement learning to estimate a reward function from few trajectories. An agent generates informative trajectories for reward estimation and pose them as queries to the expert. The proposed method generates the trajectories using expectation of discounted accrued features which is calculated by dynamic programming. The informativeness of the trajectories is evaluated by criteria for queries. Simulation results in a cart-pole domain demonstrate that the proposed method efficiently estimates the reward function from few trajectories.</p>

DOI： 10.1299/jsmermd.2017.2P2-E05

researchmap

Language：Japanese   Publisher：公益社団法人 精密工学会  

ロボットハンドが物体を認識・把持するために，ロボットハンドに取り付け可能なマルチスリットレーザプロジェクタを用いた小型距離画像センサの開発を目的とする．従来センサの計測範囲は100～300mmである．本研究では，レーザ光のぼけを計測しガウス分布に近似する手法で50～100mmでの距離情報の取得を可能とした．また，実験で距離を計測し，誤差評価を行った．

DOI： 10.11522/pscjspe.2017S.0_53

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPIE-INT SOC OPTICAL ENGINEERING  

This paper introduces a simultaneous estimation method of the extrinsic parameters of multiple fish-eye cameras using simple calibration markers. Precise extrinsic parameters of cameras mounted on a car are important, for example, to provide a seamless overhead view image to the driver. Calibration markers are set in the area that are observable from adjacent two cameras. Extrinsic parameters of each camera are estimated individually and then combined and refined using a geometric constraint. Cube markers are chosen as the calibration markers. The method is evaluated by simulation and experiments using a real car. It is shown that extrinsic parameters are obtained by the proposed method and suggested that the errors of intrinsic parameters affect the estimation of extrinsic parameters.

DOI： 10.1117/12.2266999

Web of Science

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

<p>In this paper, a three-dimensional (3D) mapping method is proposed using a range image sensor with a multi-slit laser projector. Motion parameters of measurement points are directly estimated using range images and optical flow of color images. A 3D map is constructed by registration of range images between frames using the relationship of the range images and estimated motion parameters. In experiments, a 3D map is generated by the constructed sensor using the proposed method. The computational time of estimating the motion parameters and the accuracy of the map are verified.</p>

DOI： 10.1299/jsmermd.2017.2A2-N10

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a target-tracking system for a mobile robot equipped with a stereo camera. Mobile service robots with the ability to track a specific person in dynamic environments have been required. In such environments, varying illumination and the presence of multiple people are challenges to carrying out target tracking. Color and location information is used for the targets features, that are useful for distinguishing a target from the other people. However, color information is not resilient to illumination changes. On the other hand, location information might be infeasible when non-target people present in the environments. Therefore, it is necessary to combine each information according to the situations, in which either information is infeasible to use as the feature. In order to make the system robust to varying illumination and presence of non-target people, a parameter of illumination changes is introduced in this paper. The parameter is defined using automatically adjusted white balance. The color and location features are weighted based on white-balance changes to determine a target. The effectiveness of the proposed system is verified through target-tracking experiments in outdoor environments. It is demonstrated that the proposed method can successfully recognize a target in the environments where the lighting condition changes extremely and non-target people present.

DOI： 10.1299/transjsme.15-00368

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Engineers  

In this paper, a high-speed three-dimensional map generation method using direct estimation of the motion parameters of the sensor is proposed. The method aligns range images between frames after estimating the motion parameters of the sensor using two kinds of information. One is the relationship of the range images and the motion parameters of a small motion. The other is the relationship of the intensity gradient information of the color images and the motion parameters. The method can create three-dimensional maps quickly because the method does not need to obtain the correspondences of features. Both the processing time required for the motion parameter estimation and the accuracy of the map are verified by experiments using a developed fast RGB-D sensor. It is also shown that the method is applicable to the Kinect sensor.

DOI： 10.1299/transjsme.15-00367

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publisher：公益社団法人 精密工学会  

マルチスリット光を照射するレーザプロジェクタとCMOSカメラを組み合わせた超小型距離画像センサを構築し，視差とスリット光像のぼけ計測との２つの手法を用いることで，より広い測定範囲で利用可能なロボットハンドの指先などに搭載するシステムを提案している．

DOI： 10.11522/pscjspe.2016S.0_163

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper proposes making a stereo camera using two fisheye cameras. The measurement range of the stereo camera is increased by using fisheye cameras. Fisheye images are deformed by using an equirectangular projection to simplify the stereo correspondence search in the images. Perspective transformation is applied to images so that feature points of far objects coincide in the images of the two cameras in order to remove mismatches due to the errors of intrinsic and extrinsic parameters. The corresponding points are calculated by using stereo matching in the deformed image, and the distance is measured from the disparity. The accuracy of the constructed stereo camera's distance measuring is evaluated by experiments. The performance of the constructed stereo camera is evaluated by indoor and outdoor experiments.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：ASSOC COMPUTING MACHINERY  

The detection of moving objects has become an important field of research for mobile robots. It is difficult to detect moving objects from a moving camera because moving objects and the background can appear to move. This paper proposes a method for detecting moving objects using a moving stereo camera. First, the camera motion parameters are estimated by using optical flow with a stereo camera. Second, the optical flow occurring in the background is removed. Finally, moving objects are detected individually by labeling the remaining optical flow. The proposed method has been evaluated through experiments using two pedestrians in an indoor environment.

DOI： 10.1145/3004010.3004016

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, a method that directly estimates motion parameters of a range image sensor using range images and optical flow of color images is proposed. Linear equations for motion parameters are introduced by utilizing optical flow. A three-dimensional (3D) map is constructed by registration of range images between frames using the estimated motion parameters. Experiments to construct a 3D map using the proposed method are given. In the experiments, the accuracy of the constructed map and the time of estimating the motion parameters are evaluated. It is shown that the proposed method can construct an accurate 3D map for several scenes and estimation time of motion parameters is less than 30 ms.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：The Japan Society of Mechanical Engineers  

We propose a simple method to estimate fish-eye camera's intrinsic parameters without calibration targets. Our method takes advantage of trajectories of feature points in the scene. The trajectories of feature points are obtained by a rotation movement of the camera in a horizontal plane. We therefore can utilize rich feature points for calibration, and furthermore, specific calibration targets are not required. The validity of proposed method is verified by simulation experiments using artificial data.

DOI： 10.1299/jsmeicam.2015.6.133

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：The Japan Society of Mechanical Engineers  

In this paper, development of a miniature size range image sensor is presented. The sensor is designed for a robot hand by using an endoscopy camera and a multi-slit laser projector, which can obtain range images. The improvements over the previous sensor are size, accessibility and measurement range in close distance. The dimension of the sensor is 16×40×53 mm and its weight is about 28 g. By using a universal webcam chip, the computer can capture the image data without the driver. The sensor can measure the distance from 50-150 mm. The effectiveness of the sensor is verified by experiments.

DOI： 10.1299/jsmeicam.2015.6.92

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

There is a need for precise extrinsic parameters of vehicle mounted cameras in order to provide a good overhead view image to the driver. In this paper, we propose a simultaneous estimation of the extrinsic parameters of multiple fish-eye cameras using simple calibration targets. The proposed method estimates the parameters directly from the fish-eye images, giving more accurate results in comparison with a previous method that uses perspective projection converted images. Moreover, this method allows parameter estimation by simply arranging calibration targets anywhere around the vehicle. We conducted two experiments to quantitatively evaluate the estimation accuracy of the proposed method. In each experiment, two cases have been considered. In one case (the known case), targets' position were known beforehand. In the other case (the unknown case), targets' positions were unknown beforehand. In each experiment, the estimation error for the unknown case was 4-5 times bigger than the known case. In a simulation experiment, we compared the estimation accuracy of different target shapes, in order to identify the optimal shape of the target. The cubic shape turned out to be the most appropriate.

DOI： 10.2493/jjspe.81.164

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

In this paper, a system for measuring flow of pedestrians using a stereo camera is proposed. The system does not need adjustment after installation and its installation environment is not restricted. The proposed method consists of two stages, i.e., estimation process and measurement process. Estimation process automatically estimates parameters that are typically adjusted manually after installation by using images and distance information obtained from a stereo camera. Measurement process measures flow of pedestrians using one of the two methods, i.e., particle filter and a method using KLT and Voronoi. The selection is performed depending on the congestion level of the pedestrians. The proposed method is evaluated through experiments for four scenes that are typical in real operation.

DOI： 10.2493/jjspe.81.149

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Tracking a specific person in dynamic environments is a fundamental task of mobile service robots. Image information is essential to identify a target person, however, the information is not reliable under varying illumination. In this paper, we propose a target-tracking system using a stereo camera. Color and location information is used for the target's feature, which is useful to distinguish a target from the other people. An evaluation value to identify a target is defined as weighted sum of the color and location features. The weight to the features is derived from a parameter of illumination changes. The parameter of illumination changes provides the system with capability of robust tracking even under varying illumination. We confirmed robustness of the proposed system through target-tracking experiments in outdoor environment where the lighting condition changes extremely.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper reports that uncertainty of detected edges on a fish-eye image depends on the direction to observe the edges. In fish-eye cameras, dimensions of observation space corresponding to a pixel greatly change with the pixel location. This dimension is defined as "spatial uncertainty" in this paper and formulated with typical projection models of fish-eye cameras. Experimental results showed that the uncertainty of edge detection on fish-eye images increased with increase of the spatial uncertainty.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper deals with the background maintenance problem and proposes a novel pixel-wise solution. The proposed backgroud maintenance algorithm is histogram-based. The algorithm has the following main features: fast background initialization, high accuracy in describing the real background and fast reaction to sudden changes. The basic idea of our algorithm is that the pixels are updated only if a statistic measure on the intensity variations of each pixels is greater to an adaptive threshold, thus reducing the I/O channel occupation. Experimental results on dynamic scenes taken from a fixed camera show that the proposed algorithm produces background images with an improved quality with respect to classical pixel-wise algorithms.

DOI： 10.1109/CISIS.2015.23

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Apprenticeship learning based on inconsistent demonstrations is presented in this paper. We address a problem where given demonstrations are not directly applicable to reward function estimation due to the non-stationarity of an environment or the difference between the dynamics of a robot and a demonstrator. A basic idea of the proposed method is to use a subset of the trajectories sampled from the baseline policy as training data for inverse reinforcement learning. All consistent sample trajectories and inconsistent demonstrations are abstracted by an affine transformation invariant feature. Using the feature, the importance of each sample trajectory is estimated. Rating the sample trajectories based on importance, the training data for inverse reinforcement learning are identified. The validity of our approach is verified through simulation in two scenarios: inconsistency caused by variation of an environment and performance of a robot.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

This paper describes a real-time implementation of a recently proposed background maintenance algorithm and reports the relative performances. Experimental results on dynamic scenes taken from a fixed camera show that the proposed parallel algorithm produces background images with an improved quality with respect to classical pixel-wise algorithms, obtaining a speedup of more than 35 times compared to CPU implementation. It is worth noting that we used both the GeForce 9 series (actually a 9800 GPU) available from the year 2008 and the GeForce 200 series (actually a 295 GPU) available from the year 2009. Finally, we show that this parallel implementation allows us to use it in real-time moving object detection application.

DOI： 10.1007/978-3-319-23237-9_19

Web of Science

researchmap

Language：Japanese   Publisher：The Japan Society for Precision Engineering  

DOI： 10.11522/pscjspe.2015A.0_271

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Springer Verlag  

In this paper we present a system for detecting shadows in dynamic indoor and outdoor environment. The algorithm we propose fuses together color and stereo disparity information. Some considerations on the nature of the shadow improves the algorithm’s ability to candidate the pixels as shadow or foreground. The candidate of both color and disparity information are then weighted by analyzing the effectiveness in the scene. The techniques employed allows separate computation and multithreading operations.

DOI： 10.1007/978-3-319-25252-0_38

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：Knowledge Systems Institute Graduate School  

In this paper a novel moving shadows detecting algorithm is proposed. The algorithm can be used in indoor and outdoor environments. The algorithm we propose fuses together color and stereo disparity information using the Dempster-Shafer combination rule. Some considerations on the nature of the shadow improves the algorithm's ability to candidate the pixels as shadow or foreground. The candidate of both color and disparity information are then weighted by analyzing the effectiveness in the scene.

DOI： 10.18293/DMS2015-030

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本ロボット学会  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes a method for automatic tracking of multiple humans in various scenes using a stereo camera. Surveillance cameras are increasing these days for security. Typically, images of a camera are stored and are surveyed after an incident occurred. It is required to use the cameras online, and real-time automatic human tracking is required. The proposed method detects candidate regions of humans using "Subtraction Stereo", which restricts stereo matching to foreground regions extracted by subtraction and obtains distance information for the regions. Human tracking is carried out for the extracted regions by using Particle Filter. Particle Filter consists of four steps, i.e., prediction, calculation of likelihood, data association, and resampling. Three features: distance, color, and direction of motion are used in the Particle Filter of the proposed method to achieve robust human tracking. When humans with similar cloth colors pass each other, occlusion occurs and human tracking often fails with distance and color only. The proposed method achieves the robustness to the occlusion by explicitly considering the direction of human motion. The proposed method is evaluated through experiments using a stereo camera that simulates a surveillance camera in real scenes. Tracking accuracy of more than 90% is achieved in three different scenes, which shows the effectiveness of the proposed method for human tracking.

DOI： 10.1299/transjsme.2014dr0052

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE COMPUTER SOC  

This paper proposes a human detection method that combines range image segmentation and human detection based on image local features. The method uses a stereo vision system called Subtraction Stereo, which extracts a range image of foreground regions. An extracted range image is segmented for each object by Mean Shift Clustering. Human detection based on local features is applied to each segment of foreground regions to detect humans. In this process, regions to scan a detection window for extracting local features are restricted. In addition, the size of the detection window is obtained using the distance information of a range image and camera parameters. Therefore, processing time and false detection can be reduced. Joint HOG features are used as the image local features. When applying the Joint HOG based human detection, occlusion of multiple humans is considered in construction of a classifier and in integration of detection windows, which improves the detection performance for the occluded humans. The proposed method is evaluated by experiments comparing with the method using Joint HOG features only. 11fps fast human detection is achieved.

DOI： 10.1109/ICPR.2014.734

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents a target tracking system for a mobile robot in both indoor and outdoor environments. A stereo camera, which is robust to sunlight and illumination changes, is used. Human regions in images are detected from 3D information. Using color information, a target region is discriminated from the detected human regions. Hue and saturation are chosen as features robust to illumination changes. Finally, a mobile robot is controlled based on the 3D information of the detected target region. The effectiveness of the proposed system is verified through human following experiments in both indoor and outdoor environments.

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society for Precision Engineering  

This paper proposes a method of detecting affected segments of glaucoma from optical coherence tomography (OCT) images. Thickness of nerve fiber layers and its asymmetry, difference, and variance are used as features. OCT images are segmented and the four features are obtained at each segment. Normal and glaucoma classes are constructed at each segment using training data. Detection of affected segments of glaucoma is carried out using four pattern classification methods : classification using Mahalanobis distance, maximum likelihood, nearest neighbor method, and support vector machine. The proposed method is evaluated by experiments to compare the detection results by the method and ones by a doctor.

DOI： 10.2493/jjspe.79.1124

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/10025751504?from=CiNii

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Engineers  

In this paper, a compact range image sensor for measuring objects in a short distance is constructed using a multi-slit laser projector. When a robot executes a task such as grasping an object or avoiding an obstacle, three-dimensional (3D) information of the target is usually necessary. A humanoid generally carries a vision sensor at its head and recognizes, grasps or avoids objects using the vision sensor. Measurement with a sensor attached to a robot's body is often suffered from occlusion by its hand just before grasping an object. The constructed sensor is compact enough to be attached to a robot's hand and thus occlusion can be avoided. The compactness of the sensor is achieved by using a small CMOS camera and a laser projector and setting the baseline length between them as short as possible. The short baseline length also enables measurement in a short distance. Some experiments verify that the constructed sensor can obtain 3D information of objects appropriately in a short distance.

DOI： 10.1299/kikaic.79.2451

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society for Precision Engineering  

In this paper, a method to detect slight finger-waving from images is proposed. The proposed method converts input images to low-resolution intensity images. To obtain the amplitude spectrum, the method applies windowed-Fourier transform to time series of each pixel in the low-resolution images. The difference value of the amplitude spectrum between 2Hz and the peak of 3 to 6Hz is accumulated as an evaluation value. If the evalutaion value becomes above the threshold value, the pixel is detected as a point with finger-waving movement. The proposed method can detect even slight finger-waving. A gesture interface system using the proposed method with multiple cameras is constructed. The system obtains 3D position of the finger-waving. The position is related to some command and the system controls the equipment.

DOI： 10.2493/jjspe.79.565

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/10020299346?from=CiNii

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ELSEVIER SCIENCE BV  

Environments significantly influence the sensation of pedestrians, while sensing and navigation technologies can help people improve their trip comfort. In this paper, we present an integrated framework, named NaviComf, which constructs pedestrian navigation systems to improve comfort in time varying environments taking into account the heterogeneous environmental factors. With NaviComf we aim to systematically provide solutions to the four key issues: (1) how to organize the huge amount of sensor data, (2) how to forecast future environmental information, (3) how to incorporate the heterogeneous environmental factors, and (4) how to select optimal paths in time varying environments. We have gathered sensor data of air temperature, relative humidity, and pedestrian congestion in real environments. We have also implemented a prototype system on the basis of the framework using the sensor data. Results of simulations and evaluations show that NaviComf can efficiently provide more comfortable paths as compared with the traditional navigation method. (C) 2013 Elsevier B.V. All rights reserved.

DOI： 10.1016/j.pmcj.2013.01.002

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Popular foreground-background segmentation algorithms are based of background subtraction. In complex indoor environments, if an object in motion initially remains stationary for a certain period, it can be absorbed into the background, becoming invisible to the system. Aiming at solving this problem, this paper presents a flexible and robust foreground-background segmentation algorithm based on accurate moving objects classification. Our algorithm combines low level and high level information, i.e. the data belonging to single pixels and the result of accurate object classification respectively, to improve the background management. Accurate object classification is obtained by combining classification evidence from different object recognisers using the Dempster-Shafer rule. The proposed algorithm has been tested with a large amount of acquired images; moreover, real test cases are reported. Reported experimental results include object classification accuracies obtained with a proposed Basic Belief Assignments and measurements of the quality of the background image such as Recall-Precision and F-measure computed with different background management algorithms. The experimental results show the superiority of the proposed segmentation algorithm over popular algorithms.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

This paper proposes a method to get depth information from image sequences obtained from a moving fish-eye camera. The motion is assumed to be along the optical axis of the fish-eye camera. In this case, a point on a fish-eye image moves to radial direction. This is one of the epipolar constraints that are very effective to find corresponding points in an image sequences. Epipolar-plane image (EPI) is an image having epipolar constraint in a volume of accumulating an image sequence. The gradient of a curve on EPI has depth information of the corresponding measuring point. Measurement accuracy of the proposed method is examined over a wide-angle range.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

A fast human detection system using a stereo camera is constructed. 'Subtraction stereo', that can measure distance information of foreground regions, is used to restrict regions for human detection and to adapt the detection window size. Two methods are introduced for human detection. One is a method based on template matching using gradient images, and the other is a method using approximated Shape Context (aSC) descriptors focusing on human upper bodies. High human detection performance better than the standard HOG-based method with low calculation cost is achieved by the combination of the two methods. The effectiveness of the proposed system is verified experimentally. © 2013 IEEE.

DOI： 10.1109/ICIP.2013.6738642

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

An accurate method of recognizing finger-pointing direction using multiple cameras is proposed. Input images are segmented by color clustering prior to shape analysis. Some restrictions in previous methods, such as narrowness of operation area and predefined skin color, are removed. Finger-pointing recognition in eight directions is achieved with an average of 90% recall and 93% precision.

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

In this paper, we report the results of experimental evaluation for methods of pedestrian traffic measurement at the festival held in Tatebayashi, Gunma prefecture. To measure pedestrian traffics, which include the direction and the number of pedestrian, we have already proposed two types of measuring methods based on subtraction stereo. One is the tracking method using Kalman filter, which is used in uncrowded scenes. The other is the method using KLT and Voronoi diagram, used in highly dense scenes. Finally, we show the accuracy of the proposed methods used in city environments. © 2013 The Japan Society of Mechanical Engineers.

DOI： 10.1299/kikaic.79.1036

Scopus

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society for Precision Engineering  

In this paper, we propose a fast and stable human detection based on "Subtraction Stereo" which can measure distance information of foreground regions. Scanning an input image by detection windows is controlled about their window sizes and number using the distance information obtained from subtraction stereo. This control can skip a large number of detection windows and leads to reduce the computational time and false detection for fast and stable human detection. Additionally, we construct a human detection method with little incorrect detection using multiple detectors. The effectiveness of our method is verified by experiments.

DOI： 10.2493/jjspe.78.494

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese   Publisher：パイオニア(株)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本ロボット学会  

In this paper, a compact high-speed range image sensor that can acquire a range image mapped with a color texture is constructed. The sensor consists of only commercially available products: a monochrome CCD camera, a color CCD camera, a laser projector, and two optical filters. The laser projector projects 19×19 multi-spot lights. Using the laser projector and the monochrome CCD camera with the optical filters, range images with 19×19, totally 361 points are measured. As fast as 100[Hz] measurement of a range image mapped with a color texture is achieved. Additionally, a three-dimensional (3D) mapping method using the constructed sensor is introduced. The performance of the constructed sensor and the 3D mapping method are evaluated by experiments. It is shown that the constructed sensing system can acquire 100[Hz] real-time textured range images and that it can be applied to 3D mapping by using range and color information.

DOI： 10.7210/jrsj.30.62

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00388755172?from=CiNii

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press  

In car driving support systems and mobile robots, it is important to understand three-dimensional environment widely at once. In this paper, we use a fish-eye camera as a sensor to measure three-dimensional (3D) environments. This camera can take a wide-range and distortional image and can be easily mounted on cars. We propose a method for reconstructing 3D environment using fish-eye images based on Epipolar-Plane Image (EPI) analysis. This method enables easy and stable matching of feature points. The effectiveness of the proposed method is verified by experiments.

DOI： 10.20965/jrm.2012.p0677

Scopus

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

In this paper, we propose a method to synthesize subject-specific three dimensional finger models including skin and bones from geometric database and representative dimensions that can be measured by a caliper. The geometric database necessary for our method was developed by analyzing principal feature of the individual differences by using principal component analysis. The proposed method synthesizes subject-specific models by blending these principal features. The weight for each feature was determined by numerical optimization so that the synthesized subject-specific model reproduces the representative dimensions of the actual subject. We applied our method to fifty fingers. The average of the maximum error was 1.0mm, while the average error was 0.2mm. For more than 70% of the subjects, the bone as well as skin geometry were synthesized with acceptable error range from the representative dimensions of surface geometry. As for the remaining 30%, there was significant error in bone length. This method has distinct advantage over the conventional method in which time consuming manual segmentation for medical images is necessary, since we can generate subject-specific finger models in 5 to 10 minutes automatic computation. © 2012 The Japan Society of Mechanical Engineers.

DOI： 10.1299/kikaic.78.1202

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we present an integrated framework, named NaviComf, which constructs pedestrian navigation systems for comfort in varying environments by using multimodal sensing technologies. With NaviComf we aim to systematically provide solutions to three key problems: (1) how to build the environmental data warehouse (EDW) which works as an infrastructure providing comprehensive and predictive environmental information, (2) how to integrate heterogeneous environmental information from multi-modal sensors into an aggregate value which facilitates further processing, and (3) how to determine the optimal path plans in environments which are varying continuously. In NaviComf the multidimensional data model and data prediction method are applied to build the EDW. Then a novel multi-factor cost (MFC) model is proposed as the fundamental concept to integrate the multi-modal sensor data. Based on the former two solutions, the optimal path planning (PP) problem is solved in a time-dependent network by applying a dynamic programming method. In the evaluations of NaviComf, sensor data for temperature, humidity, and pedestrian traffic flow have been gathered in real environments and a prototype system has been implemented with the data. Evaluations are conducted by using the prototype system and the results show that NaviComf can efficiently navigate pedestrians through more comfortable paths as compared to the traditional navigation method.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press  

This paper proposes a method of recognizing movements of the mouth from images and implements the method in an intelligent room. The proposed method uses template matching and recognizes mouth movements for the purpose of indicating a target object in an intelligent room. First, the operator's face is detected. Then, the mouth region is extracted from the facial region using the result of template matching with a template image of the lips. Dynamic Programming (DP) matching is applied to a similarity measure that is obtained by template matching. The effectiveness of the proposed method is evaluated through experiments to recognize several names of common home appliances and operations.

DOI： 10.20965/jrm.2012.p0311

Scopus

researchmap

Language：Japanese   Publisher：公益社団法人 精密工学会  

近年，緑内障に対するコンピュータ支援診断として網膜神経線維層厚のカラーマップを作成することで医師を補助するという方法が確立されている．しかし線維層厚のカラーマップでは病状が初期の場合，変化が表れず，補助することが困難である．そこで本論文では網膜神経線維層厚の非対称性を考慮することで疾患領域を検出する．検出された領域を医師のよって疾患領域と特定された個所と比較することで本手法の有用性を検証する．

DOI： 10.11522/pscjspe.2012A.0.519.0

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, a compact range image sensor used for short-range measurements is constructed with a multi-slit laser projector. Three-dimensional (3D) information obtained with a sensor is important for a robot that grasps an object. Sensors attached to robots for measurement are often hindered by occlusion by the hand immediately before an object is grasped. The constructed sensor is compact enough to be attached to a robot's hand, and occlusion can thus be avoided Compactness of the sensor is achieved by using a small CMOS camera and a laser project or and setting the baseline length between them as short as possible. The short baseline length also enables measurement in a short distance. Some experiments verify that the constructed sensor obtains accurate 3D information of objects in a short distance.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：TRANSACTION PUBLISHERS  

Texture mapping on scanned objects, which is the method to map color images on a 3D geometric model measured by a range image sensor, is often used for constructing a realistic 3D model. Color images are affected by the illumination conditions. Therefore, discontinuities of seams occur when simply applying texture mapping. In this paper, we propose a method for correcting the discontinuities using a range intensity image. A range intensity image is a kind of intensity image that is related to the reflectance ratio of the measured points, simultaneously acquired with a range image using an active range sensor. The method estimates the color information that is not affected by the lighting environment using multiple color images and a range intensity image. As a result, the method is effective to construct a 3D model with seamless color images. The effectiveness of the correction method is illustrated by experiments with real-world objects.

DOI： 10.4028/www.scientific.net/KEM.523-524.362

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：SICE  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Electrical Engineers of Japan  

In this paper, the authors describe multi-dimensional sensing that can be adopted to realize security in everyday life. In this study, the applications of sensing based on the use of surveillance cameras for equipment mounted on cars, road transport, fields such as farming, and health maintenance have been investigated. Recently developed systems and methods involving multi-dimensional sensing technologies are introduced, and current issues and trends are described.

DOI： 10.1541/ieejias.131.418

CiNii Books

researchmap

Language：Japanese   Publisher：パイオニア(株)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper a real time system for cars, pedestrians and bicycle detection and classification is presented. The system aims at monitoring the traffic flow in urban zones and uses video data acquired with both mono and stereo cameras. All the algorithms have been developed in a pixel-wise fashion in order to be parallelized on a GPU device for real-time performances. We show that by a GPU implementation of the time-consuming parts of the proposed system, we perform detection and classification at about 25 frame per second to ensure prompt and effective reaction to the monitored events.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

In this paper, we propose a new method for 3D-2D registration based on SIFT and a range intensity image, which is a kind of intensity image simultaneously acquired with a range image using an active range sensor. A linear equation for the registration parameters is formulated, which is combined with displacement estimations for extrinsic and intrinsic parameters and the distortion of a camera's lens. This equation is solved to match a range intensity image and a color image using SIFT. The range intensity and color images differ, and the pairs of matched feature points usually contain a number of false matches. To reduce false matches, a range intensity image is combined with the background image of a color image. Then, a range intensity image is corrected for extracting good candidates. Moreover, to remove false matches while keeping correct matches, soft matching, in which false matches are weakly removed, is used. First, false matches are removed by using scale information from SIFT. Secondly, matching reliability is defined from the Bhattacharyya distance of the pair of matched feature points. Then RANSAC is applied. In this stage, its threshold is kept high. In our approach, the accuracy of registration is advanced. The effectiveness of the proposed method is illustrated by experiments with real-world objects. © 2011 Springer-Verlag.

DOI： 10.1007/978-3-642-24028-7_30

Scopus

researchmap

Language：Japanese   Publisher：一般社団法人 映像情報メディア学会  

近年,操作者のジェスチャを認識して家電製品などを操作するシステムが様々な場面で期待されている.本論文では,ジェスチャ操作システムの一つであるインテリジェントルームにおいて,操作する家電製品の選択を,距離画像センサを用いた3次元ジェスチャ認識によって行うことを提案している.操作者が操作したい家電製品に対して指を差すことで,操作する家電製品の選択を行っている.距離画像センサで取得した画像から指領域を抽出し,指の3次元的な角度を求めることで指差し方向を認識する.実験により,指差し方向の認識率を評価している.

DOI： 10.11485/itetr.35.33.0_75

researchmap

Language：Japanese   Publisher：The Institute of Image Information and Television Engineers  

A control system is proposed for home appliances using only slight hand-waving gesture. The system detects the gesture with four cameras and calculates its 3D-position. Some operations of TV (turn on/off, channel change, volume up/down, etc.) can be related the hand-waving positions. Experimental results show that the system is reliable.

DOI： 10.11485/iteac.2011.0_9-9-1

CiNii Books

researchmap

Publisher：The Japan Society for Precision Engineering  

DOI： 10.11522/pscjspe.2011A.0.924.0

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, we propose a fast and stable human detection based on "subtraction stereo" which can measure distance information of foreground regions. Scanning an input image by detection windows is controlled in their window sizes and number using the distance information obtained from subtraction stereo. This control can skip a large number of detection windows and leads to reduce the computational time and false detection for fast and stable human detection. Additionally, we propose two-step boosting as a new training way of classifier with whole and upper human body models. Experimental results show that the proposal is faster and less false detection than the method described in the reference [1].

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：AMER INST PHYSICS  

In this paper, we propose a fast and stable human detection based on subtraction stereo for generic environments. Scanning an input image by detection windows, the numbers and size is controlled by distance obtained from subtraction stereo. A preprocessing phase reduces the effect of noise and shadow, for a better segmentation. This control decrease the computation time, important in real-time applications, skipping a large number of detection windows. Experimental results show that the proposal is faster and less false detection.

DOI： 10.1063/1.3596655

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Institute of Electronics, Information and Communication Engineers  

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

In this paper, a practical method of stereo vision, "subtraction stereo" is proposed. A huge number of studies have been carried out for stereo vision until now, and several practical stereo vision systems have been reported. However, what is called the correspondence problem that stereo matching becomes difficult and not robust for weak textures or recurrent patterns is inevitable for stereo vision. Subtraction stereo realizes robust measurement of range images by detecting moving regions with each camera first and then applying stereo matching for the detected moving regions. Detection of moving regions is carried out with a subtraction process. Concept and fundamental algorithm of subtraction stereo are introduced. Then measurement of three-dimensional position, height and width of a target object using the subtraction stereo is discussed. The basic algorithm is implemented on a commercially available stereo camera and the effectiveness of the subtraction stereo is verified by several experiments using the stereo camera. Although objects are restricted to moving ones, subtraction stereo gives sufficient information robustly for many applications such as surveillance.

DOI： 10.2493/jjspe.76.123

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00353149456?from=CiNii

Language：English   Publishing type：Research paper (scientific journal)  

DOI： 10.20965/jaciii.2010.p0793

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This paper describes a small high-speed range image sensor which can acquire a range image mapped with a color texture captured by another camera. The constructed sensor consists of only commercially available products: a monochrome CCD camera, a color CCD camera, a laser projector, and two optical filters. The resolution of the range image is 19×19 for a total of 361 points, and the measurement speed is 100Hz. A three-dimensional (3D) mapping method using the constructed sensor is introduced. The measurement precision of the sensor and the accuracy of the constructed 3D map are evaluated with some experiments. In addition, measurement examples of the sensor are shown. Experiments verify that the constructed sensing system can acquire 100Hz real-time textured range images and that it can be applied to 3D mapping by using range and color information. © 2010 IEEE.

DOI： 10.1109/ISOT.2010.5687379

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

In this paper, we propose a method to recognize periodic gestures from images. The proposed method uses a amplitude spectrum and a phase spectrum that are obtained by applying Fast Fourier Transform (FFT) to a time series of intensity images. FFT is applied to each pixel of low-resolution images. The method consists of 2 steps. First, the method detects periodic motion regions from the amplitude spectrum. Secondly, the method uses the phase spectrum in the detected periodic motion region to classify the gestures. The proposed method is robust to lighting conditions and individual differences in skin color because it does not rely on color information. Several experiments are performed to demonstrate the effectiveness of the proposed method. © 2010 IEEE.

DOI： 10.1109/ISOT.2010.5687352

Scopus

researchmap

Language：Japanese   Publisher：一般社団法人 映像情報メディア学会  

現在,物体や場所を指示するためのインタフェースとして指差し動作の研究が盛んに行われている.本研究では,肌色抽出された顔領域と手領域に対して,指差し動作前後の移動量に注目し,抽出が乱れた場合でも顔領域と手領域の識別を可能にすることで,インテリジェントルームの指差し方向認識の認識率向上を実現している.

DOI： 10.11485/itetr.34.34.0_67

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Texture mapping is useful for constructing a three-dimensional (3D) model because a realistic 3D model can be obtained efficiently and quickly. This paper proposes a system to construct a 3D model using range intensity images. A range intensity image, which is also called a reflectance image, refers to the intensity image that is acquired simultaneously with the range image captured using an active range sensor. Such an image has an important property in which illumination conditions, such as geometrical arrangement and power of illumination, can be controlled at the capture time, which allows the estimation of the reflectance properties of the object. Several methods using range intensity images are improved and combined to construct an effective system
the registration of range images and color images is realized, an omnidirectional geometric model is constructed by registering and integrating multiple range images with range intensity images, and the influence of the illumination environment that occurs in color images is removed. In addition, a method to estimate the illumination color is introduced to compensate for the color of illumination light. Experiments show the effectiveness of the constructed system for obtaining a realistic 3D model. © 2010 IEEE.

DOI： 10.1109/CRV.2010.48

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

Detection of obstacles on a plane is important for a mobile robot that moves in a living space, especially for a humanoid that falls down even with a small obstacle. In this paper, a range image sensor for detecting small obstacles on a plane is constructed using a multi-slit laser projector. The sensor consists of a commercially available laser projector and a CCD camera. It measures a relative disparity map (RDMap) whose measurement errors are not affected by the distance. From the obtained RDMap, a plane is estimated using RANSAC and regions out of the plane are detected as obstacles. Experiments show that planes can be obtained with small errors in RDMaps with the constructed sensor, and that a humanoid with the constructed sensor can detect small obstacles such as a moving ping-pong ball and a LAN cable on a plane by the proposed methods while walking.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

In this paper we present a method for the detection of moving objects for unknown and generic environments under cast shadow and periodic movements of non relevant objects (like waving leaves), using a combination of non-parametric thresholding algorithms and local cast shadow analysis with stereo camera information. Good detection rates were achieved in several environments under different lighting conditions, and objects could be detected independently of scene illumination, shadow, and periodic changes. © 2010 IEEE.

DOI： 10.1109/ICPR.2010.89

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

We propose a method for multi-object segmentation in a projection plane. Our algorithm requires a stereo camera system called Subtraction Stereo, which extracts foreground information with a fixed stereo camera. The main contribution of this paper is how the image sequences that include partial occlusion of the foreground objects can be accurately segmented using mean shift clustering in real-time processing. The proposed method is suitable for inside a medium-sized environment, such as a room. Finally, we try to segment the sequences that include occlusion and show the accuracy of the proposed method. © 2010 IEEE.

DOI： 10.1109/ICPR.2010.806

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：富士技術出版  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：ACADEMIC PRESS INC ELSEVIER SCIENCE  

Most active optical range sensors record, simultaneously with the range image, the amount of light reflected at each measured surface location: this information forms what is called a range intensity image, also known as a reflectance image. This paper proposes a method that uses this type of image for the correction of the color information of a textured 3D model. This color information is usually obtained from color images acquired using a digital camera. The lighting condition for the color images are usually not controlled, thus this color information may not be accurate. On the other hand the illumination condition for the range intensity image is known since it is obtained from a controlled lighting and observation configuration, as required for the purpose of active optical range measurement. The paper describes a method for combining the two sources of information, towards the goal of compensating for a reference range intensity image is first obtained by considering factors such as sensor properties, or distance and relative surface orientation of the measured surface. The color image of the corresponding surface portion is then corrected using this reference range intensity image. A B-spline interpolation technique is applied to reduce the noise of range intensity images. Finally, a method for the estimation of the illumination color is applied to compensate for the light source color. Experiments show the effectiveness of the correction method using range intensity images. (C) 2009 Elsevier Inc. All rights reserved.

DOI： 10.1016/j.cviu.2009.03.015

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本機械学会  

We analyzed individual differences of finger geometry in order to synthesize finite element models with geometric variations. Volume registration technique was employed to quantify the individual differences from 3-dimensional MRI images of 50 male subjects. The registration method computes a displacement field that transforms geometry of the template subject into that of the target subject, thus the displacement field is a good measure of the geometric difference between two different subjects. The result of principal component analysis for the computed displacement fields has clarified the principal features of finger geometry corresponding to dirmension, thickness, thickness of finger pad, ratio of bone length and direction of bending. We can synthesize possible variations of finger geometry by transforming a template FE model according to a displacement field computed by weighted sum of these principal features.

DOI： 10.1299/kikaic.75.2765

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：電子情報通信学会  

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：電気学会論文誌  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Information Processing Society of Japan  

Although human activities in the World Wide Web are increasing rapidly due to the advent of many online services and applications, we still need to appraise how things such as a merchandise in a store or pictures in a museum receive attention in the real world. To measure people's attention in the physical world, we propose SPAL, a Sensor of Physical-world Attention using Laser scanning. It is challenging to use a laser scanner because it provides only front-side circumference of any detected objects in a measurement area. Unlike cameras, a laser scanner poses no privacy problem because it does not recognize and record an individual. SPAL includes many important factors when calculating people's attention, i.e., lingering time, direction of people, distance to a target object. To obtain such information for calculation, we develop three processing modules to extract information from raw data measured by a laser scanner. We define two attention metrics and two measurement models to compute people's attention. To validate the proposed system, we implemented a prototype of SPAL and conducted experiments in the real-world environment. The results show that the proposed system is a good candidate for determining people's attention.

DOI： 10.2197/ipsjtrans.2.93

researchmap

Language：Japanese   Publisher：The Japan Society for Precision Engineering  

DOI： 10.2493/jjspe.75.207

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society for Precision Engineering  

DOI： 10.2493/jjspe.75.213

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00359716836?from=CiNii

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

researchmap

Language：Japanese   Publisher：一般社団法人 映像情報メディア学会  

本研究では,二足歩行ロボットに小型距離画像センサを搭載して,歩行中に障害物検出および回避行動生成を行う手法を提案する.小型距離画像センサから得られる相対視差画像を用いて歩行可能平面と障害物を検出する.実験により,生活空間における本手法の有効性を示す.

DOI： 10.11485/itetr.33.34.0_47

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This paper discusses a method to detect and measure pedestrians in urban environment using "subtraction stereo." Subtraction stereo is a stereo vision method that calculates distance only for moving regions to make the stereo matching robust and fast. Pedestrians are detected based on their three dimensional features, which are position, height, width, and so on, obtained from a range image by standard labeling. The number of pedestrians in groups is estimated from their area on a range image. In urban environments, the effectiveness of this method is verified by experiments using a stereo camera which is commercially available and implemented with subtraction stereo. ©2009 IEEE.

DOI： 10.1109/INSS.2009.5409909

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

In this paper, detection of pedestrian groups and counting of the number of pedestrians in each group using "subtraction stereo" are discussed. Subtraction stereo is a stereo vision method that focuses on the movement of objects to make a stereo camera robust and produces range images for moving regions. Pedestrian groups are detected with a standard labeling, and three dimensional (3D) features of pedestrian groups are measured from range images obtained by subtraction stereo. Then a method to count the number of pedestrians in a group is proposed. The basic algorithm of the subtraction stereo is implemented on a commercially available stereo camera, and the effectiveness of the method to count the number of pedestrians is verified by experiments using the stereo camera. © 2009 Springer-Verlag.

DOI： 10.1007/978-3-642-10520-3_51

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：SPRINGER-VERLAG BERLIN  

This paper reports the effect of pre-operation intended to familiarizing oneself with vicarious experiences of different-sized hands. To measure this effect, the index of degree of immersion (DOI) is proposed which represents whether observed behavior is appropriate for the presented hand size. The DOI is measured for various sizes of hands when changing type of pre-operation which is classified based on relationship between hands and objects. The experimental results show that the pre-operation is effective for familiarizing the presented sized hand. especially in larger sized hands, and that behavior of touching and controlling an object in position is important for effective pre-operation.

Web of Science

researchmap

Language：Japanese   Publisher：一般社団法人 映像情報メディア学会  

本稿では,汎用的な周期ジェスチャ認識手法の構築を目指し,周期運動領域の重心座標を利用した手の動作領域抽出手法を提案している.実験を行い,提案手法を用いることにより的確に手の動作領域を抽出可能であること,各ジェスチャの認識率が向上することを確認し,有効性を示している.

DOI： 10.11485/itetr.33.34.0_79

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

In this paper, a camera simulator that simulates camera properties and simple methods to obtain its parameters are proposed. The camera simulator simulates distortion, limb darkening, blur by defocus, blur by aberration, motion blur, variation of pixel sensitivity, random noise, gamma correction, and salt and pepper noise. As for methods to obtain camera parameters, parameters of blur by aberration and distortion are obtained by getting an image of a stripe pattern, and parameters of limb darkening, variation of pixel sensitivity, random noise and salt and pepper noise are obtained by getting a number of images using a camera covered by a diffuser. Fundamental experiments show that the constructed Camera Simulator and the proposed methods to obtain camera parameters are effective.

DOI： 10.2493/jjspe.74.870

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00356059202?from=CiNii

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Eengineers (C)  

Time optimal control method for quadruped walking robots are developed and installed into a practical robot system. Each leg is modeled as a two link manipulator whose time optimal control theory has already been established by Bobrow. However, in legged robot systems, each leg supports their body weight, and the reaction forces from its ground must be inside of their friction corn. Moreover, the ZMP(zero moment point) of the robot is constrained for stable walk. Therefore, time optimal control inputs must be designed considering these constraints. SONY ERS-7 is used as a quadruped walking robot and a fundamental experiment is done. From the experimental results, the effectiveness of the developed control algorithm is verified.

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

We demonstrate "Subtraction Stereo", a simple and robust stereo vision algorithm that is suitable for surveillance applications such as detection of pedestrians.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper presents an MRI-compatible optical force sensor: its principle, structure and performance. Our force sensor employs an optical micrometry based on differential measure of light intensity. This technology enables highly accurate and sensitive two degrees-of-freedom displacement sensing by using a point source, a photo detector and an optical lens. The developed force sensor achieves MRI compatibility by placing all the electric and metallic components outside MRI, and connecting the light path between the optical components by using multi-core optical fibers. The sensor head component has parallel plate structure and is made of glass fiber reinforced poly-ether-ether-keton those help reducing axial interference and hysteresis characteristics of plastic resin. The accuracy of this force sensor is better than 1.6 %F.S. under the applied force ranging from 0 to 3 N.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, a high-speed range image sensor using a multi-spot laser projector is constructed. Several highspeed range image sensors have been developed recently. Their sampling rate is around the video rate (30Hz or so) and a faster sensor is required. The proposed sensor has achieved 200Hz measurement. It consists of a commercially available laser projector and a high-speed CCD camera. The number of pixels is 361 and the measurement range is 800-2000mm. Although the acquired range image is sparse, the proposed sensor is thought to be adequate for several applications such as robot vision because of its high-speed imaging and compactness. Some characteristics such as measurement errors are discussed, and the effectiveness of the proposed sensor is verified by experiments.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press Ltd.  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press Ltd.  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

In this paper, methods for detecting obstacles on a plane using a relative disparity map (RDMap) are proposed and discussed. The RDMap, which was formerly introduced by the author Umeda, is relative to a plane that is observed at first as the reference. It has an interesting feature that a plane in real 3D space also becomes a plane in the map, and has homogeneous characteristics compared to an ordinary range image. The proposed methods work even when the pose of the sensor changes significantly, which is the case in humanoid walking. First, a method to detect planar regions and obstacles by fitting a plane to the RDMap and a method to obtain the pose parameters from the RDMap are introduced. Fundamental experiments are then conducted to verify that a plane in real 3D space becomes a plane in the RDMap and that obstacles can be detected using the residual sum of squares for the fitted plane, and measurement errors in pose parameters are then evaluated. Finally, an experimental system with a humanoid and a small range image sensor is constructed, and it is demonstrated that the humanoid can detect obstacles on a plane by the proposed methods while walking.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

We are constructing an intelligent room in which an operator can control home appliances such as a TV set with intuitive gestures. In this paper, we discuss three dimensional (3D) measurement of the intelligent room using a distributed camera system. The room carries CCD cameras with pan, tilt, and zoom functions. An operator makes hand waving and the room detects it. 3D position of the waving hand is measured using the cameras, and then pan, tilt and zooming-up of the cameras are carried out with the 3D information. A hand region is extracted using the color information, and the operator's gestures are recognized. 3D Finger pointing to select an appliance is measured using the cameras. Experiments verify the effectiveness of the proposed 3D measurement methods.

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE COMPUTER SOC  

This paper proposes a method of correcting intensity of a color image, which is used for the texture of a 3D model, using a range intensity image. A range intensity image has an effective characteristic that it is obtained under controlled illumination. This enables correction of the intensity of the color image without estimating illumination for the color image. Experiments show the effectiveness of the proposed method and a 3D model with proper color information generated by the method.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE COMPUTER SOC  

In this paper, an image-based self-localization method is proposed for a mobile robot. Images are compressed for each column, and the average and standard deviation of the pixels in each column are used. Environmental and observational data, which are the compressed image data at the registration and observational stages, are matched, and the position of the robot is obtained. The entire environment can be represented continuously with a small amount of data. A simple and robust matching method based on a voting process is introduced. The methods are evaluated through several experiments with omnidirectional images.

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Robotics Society of Japan  

In this paper, a simulator of environment and measurement that considers camera characteristics is developed for multiple autonomous mobile robots. RoboCup 4-legged robot league is chosen as the concrete target. The simulator introduces a sever/client system, and realizes separation of each robot's information, introduction of each robot's difference and distribution of processes. OpenGL is utilized to produce virtual images. So as to make a simulator reflect real environment, sensing noises should be considered. The simulator considers the effects of blur, element noises, vignetting, lens distortion and delayed exposure for each line of the CMOS image sensor. Experiments show the similarity between actual camera images and simulated camera images. Consequently, it is possible to verify programs for multiple autonomous mobile robots on the simulator, and therefore the proposed simulator is a useful tool for effective development of algorithms.

DOI： 10.7210/jrsj.23.878

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00259243876?from=CiNii

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society for Precision Engineering  

So as to improve the road safety, not only the safety of the vehicles but also the nature of drivers has to be considered. This paper proposes a concept of automatic instruction system that instructs a driver to improve his or her safety of driving using image processing techniques. So to speak, the system substitutes for an instructor at a driving school. An example of the modules of the system is constructed : it warns the driver if he or she has forgotten to turn his/her head when the driving lane is changed. Changing lanes is recognized by extracting white lines. On the other hand, face turning is recognized by extracting face and nose regions. These two are done simultaneously and whether the driver follows the rule at changing lanes is judged. Experiments show the effectiveness of the constructed module. Additionally, interviews for subjects indicate the effectiveness of the concept of automatic instruction system.

DOI： 10.2493/jspe.71.1046

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00293649658?from=CiNii

Language：Japanese   Publishing type：Research paper (scientific journal)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Robotics Society of Japan  

Though Monte Carlo localization is a popular method for mobile robot localization, it requires a method for recovery of large estimation error in itself. In this paper, a recovery method, which is named an expansion resetting method, is newly proposed. The combination of the expansion resetting method and the sensor resetting method, which is a typical resetting method, is also proposed. We then compared our methods and others in a simulated RoboCup environment. Typical accidents for mobile robots were produced in the simulator during trials. We could verify that the expansion resetting method was effective for recovery from small irregular changes of a robot's pose, and that the combination method could deal with both large and small irregular changes.

DOI： 10.7210/jrsj.23.466

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00250700286?from=CiNii

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press Ltd.  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

Various methods have been proposed until now for range measurement or three dimensional shape reconstruction. However, most of them need a large-scale equipment or a special environment. This paper proposes a technique which obtains a range image easily under a general environment using only an off-the-shelf digital camera. Distance is calculated by obtaining the irradiance of scene lighted by the flash of a digital camera using the fact that the intensity of reflected light of the flash is inversely proportional to the square of the distance from the object. The irradiance is obtained by subtracting an image without the flash from an image with the flash. The image without the flash is used to obtain the reflectance ratio at each pixel. The intensity of reflected light of the flash is affected by the inclination of the object surface. A method to estimate the inclination at each pixel is proposed which uses the change of the irradiance in adjacent pixels. The inclination is formulated as the function of the rate of change, and thus the inclination can be calculated by the rate which is easily obtained from the image. Additionally, color information is simultaneously obtained because visible light is used. Assumptions in the method are that the object surface has no specular reflection and the flash is set at the same position as the center of the lens. Experiments show that a range image is roughly obtained by the proposed method, and furthermore, that proper distance is obtained for inclined surfaces.

DOI： 10.1117/12.647730

Scopus

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

A range intensity image, which is also called a reflectance image, refers to the intensity image that is acquired simultaneously with the range image captured using an active range sensor. This paper proposes a method that uses this image to correct the color information of a textured 3D model. The color information is usually obtained by texture mapping of color images acquired by a digital camera. The lighting condition for the color images are usually not controlled, thus the color information is not precise. On the other hand, the lighting condition for the range intensity image is controlled since it is obtained from a controlled and known lighting as required for the purpose of range measurement. The paper describes the method for combining the two sources of information
experiments show the effectiveness of the correction method. © 2005 IEEE.

DOI： 10.1109/3DIM.2005.29

Scopus

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press Ltd.  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

This paper proposes a method for the registration of range and color (or intensity) images, based on the range intensity image that is simultaneously acquired with a range image. The gradient constraint between the range intensity image and the color image is introduced, and a linear equation for the registration parameters is derived, which combines displacement estimations for extrinsic and intrinsic parameters. By using the equation, precise automatic registration without explicit detection of correspondences is achieved. Experimental results illustrate the proposed method.

DOI： 10.1109/ICPR.2004.1334457

Scopus

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Robotics Society of Japan  

This paper proposes a method of detecting waving hands from images as a key technology for man-machine interface. FFT is applied to time series of intensity images. The images are converted to low-resolution ones, and FFT is applied to each pixel of the low-resolution images. The proposed method is robust to lighting condition and individual difference of skin color, because it doesn't use color information at all. Experiments show the stability and robustness of the proposed method. Additionally, the method is applied to an omnidirectional vision images and an infrared camera, so as to show the wide applicability of the method.

DOI： 10.7210/jrsj.21.923

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00234109463?from=CiNii

Language：Japanese   Publishing type：Research paper (other academic)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Japan Society of Mechanical Engineers  

In the progress of semiconductor technology, the diameters of Si substrates are increasing and the flatness accuracy is being raised. However, the evaluation method has not been accomolished. The sequential-two-points method, which was originally developed to evaluate straightness error motion of the tool as well as error profile of machined parts in machine tool, is applied to evaluate surface form of the flatness on the substrates. Forms along diameters and periphery are measured as a step to construct form of flatness on polar coordinate. Resolution of the capacitor type sensors are calibrated with order of nm, so that the measurable accuracy, of the method is assured and the straightness error from of 0.3 μm along diameter have been obtained, on the other hand which can be easily deformed by boundary conditions of support.

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Horological Institute of Japan  

Micro vision scnsors are required for milli machines or micro machines. In this paper, a prototype of a compound-eye-type micro vision sensor that imitates insect's vision is constructed. The validity of the constructed sensor is verified by simulation and experiments to obtain images. Additionally, a method to detect optical flow with the sensor is proposed.

DOI： 10.20805/micromechatronics.45.2_54

CiNii Books

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：V D I-V D E - VERLAG GMBH  

This paper proposes methods of fusing range images, intensity images and tactile images. The methods are intended for applying to distributed sensing of multiple mobile robots. As the features for fusion, practically important and possible to robustly extract ones are selected: planar regions and cylindrical regions for range images, linear edges for intensity images and planar regions and linear edges for tactile images. The fusion methods to measure a 3D edge and a cylinder are constructed and formulated by utilizing least square approach. Additionally, fusion of these features measured by multiple mobile robots is discussed. The measurement of 3D edges and cylinders are verified by experiments.

Web of Science

researchmap

Language：English   Publishing type：Research paper (international conference proceedings)   Publisher：IEEE  

This paper focuses on a method of estimating contact force dynamically between environment and an object, which is grasped by the soft planar fingers with tactile sensors. The external contact force acting on the grasped object causes the tactile sensors to generate the displacement distributions and the for-cc distributions, We present a method of dynamic tactual image acquisitions for both distributions at the pseudo video rate. Experimental results show the measurement of the tactual image flows due to the deflection of an object and the magnitude and orientation of force vectors operating on the finger according to the external contact for-cc. The contact position of the external force acting on the grasped object is estimated within 15 % from the tactual image distributions.

Web of Science

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：The Horological Institute of Japan  

Wireless Location services have recently attracted much interest. To know mobile caller's location opens a new class of services and products. In this paper, individual, simple and widely applicable methods which utilize GPS unit only are investigated. We use the velocity data from the unit and propose new method for human navigation. Firstly, devices, the best position of the GPS antenna installation was examined. Secondly, the accuracy of the position measurement using the method on an ideal condition was investigated. The experiments show that the best position to wear the GPS antenna for using the method was the "head", the accuracy was about 10%. We also proposed the map matching method that revise position data measured by the proposed human navigation method.

DOI： 10.20805/micromechatronics.44.4_48

CiNii Books

researchmap

Language：Japanese   Publisher：The Robotics Society of Japan  

DOI： 10.7210/jrsj.18.204

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00363909682?from=CiNii

Language：Japanese   Publishing type：Research paper (scientific journal)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)  

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本時計学会  

In this paper, fundamental constitution of micro compound eye sensor, and a method to measure 3D position of an object by two compound eye sensors are proposed. Milli-machines and micromachines are becoming more and more important these days, and thus the improvement of micro vision sensor is required. This paper aims to improve an artificial compound eye sensor which imitates insect's vision system. A simple compound eye sensor which consists of photo sensors and a simple cover is proposed. The method to measure 3D position is verified by experiments using a macro-model of the compound eye sensor.

DOI： 10.20805/tokeieafj.162.0_1

CiNii Books

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：IEEE, American Society of Mechanical Engineers (ASME)  

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：TAYLOR & FRANCIS LTD  

This paper proposes a three-dimensional (3D) vision system based on range image processing for mechanical assembly and disassembly. In the field of assembly and, moreover, because of the increasing importance of recycling of industrial products, in the field of disassembly, a more flexible vision system is required. Conventional intensity image processing is not sufficient and so a 3D vision system is required. A prototype of a vision system with a high-speed range image sensor is constructed, and its robustness for assembly and disassembly is verified by experiments to detect a target region (dial of an audio amplifier) with blemishes. Shape recognition algorithms which are applicable to sparse range images of the 3D vision system are introduced. They utilize primitive surface features such as planar or cylindrical surfaces as features for model matching. Fundamental experiments show the effectiveness of the general shape recognition algorithms.

Web of Science

researchmap

Language：English   Publishing type：Research paper (scientific journal)   Publisher：Fuji Technology Press  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：精密工学会  

Visual functions for bin picking are realized fast by utilizing a sparse range image. The functions are composed of (1) detection of a necessary part from a lot of parts piled on each other, and (2) measurement of its three dimensional position and orientation. Framework of the vision system is discussed on the assumption that a shape model of the necessary part is given. Method of fast segmentation of a sparse range image is presented for the part's detection. Four kinds of shapes: polyhedra, cylinder, cone and sphere are chosen as important shapes for industrial bin-picking, and for each shape, examination if a segment of a sparse range image is the shape, and measurement of its three dimensional position and orientation are shown. Experiments with a real fast range sensor and a personal computer show that the presented vision system is fast: total process time is less than 1.5 s for multiple piled cylinders.

DOI： 10.2493/jjspe.60.858

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00026843698?from=CiNii

Language：Japanese   Publisher：The Robotics Society of Japan  

Measuring relative positions among robots are essential to avoid collision and to coordinate one another. This paper proposes a signboard system and an algorithm to measure the relative positions among mobile robots in realtime. The system consists of a CCD camera and signboards with LEDs, whose 2D images are obtained by a CCD camera loaded on a different robot. The algorithm first separates the image into each signboard, and second detects the position and the orientation of the target robot. Error analysis made it possible to measure the orientation of the target robot in 360 degree with the minimum errors. The experiments with 3 mobile robots indicates that one robot can measure the other two position with the accuracy of 5% in distance within 0.4 second, in the range of 0.5 to 3 meters. The measuring speed is fast enough to calcuate the collision avoidance. The experiments verified the system is very efficient.

DOI： 10.7210/jrsj.12.472

CiNii Books

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：日本ロボット学会  

researchmap

Language：Japanese   Publishing type：Research paper (scientific journal)   Publisher：情報処理学会  

researchmap

Total pages：400   Responsible for pages：223-228   Language：English   Book type：Scholarly book

researchmap

Total pages：1728   Responsible for pages：1615-1622   Language：English   Book type：Scholarly book

researchmap

Total pages：430   Responsible for pages：411-422   Language：English   Book type：Scholarly book

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)  

researchmap

Language：Japanese   Publisher：映像情報メディア学会  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人映像情報メディア学会  

本研究では,距離画像と同時にカラー画像のオプティカルフローを用いてセンサの運動パラメータを直接推定する手法を提案している.提案手法により得られた運動パラメータを用いて連続した2枚の距離画像の位置合わせを行うことにより三次元地図生成を行っている.また,三次元地図生成実験により,距離画像の位置合わせを行い,運動パラメータ推定にかかる処理時間を検証している.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Target tracking is one of the important functions for autonomous mobile robots. In the environments where a target-tracking system is used, there may be many people and objects. When the system is applied in such environments, occlusion is a challenge for achieving target tracking. In this paper, an occlusion handling method for target tracking with a mobile robot is proposed. By using a stereo camera, both color and location information is used for target detection. During target tracking, the state of occlusion is determined. According to the state, an appropriate tracking process is applied. The effectiveness of the proposed method is verified through target-tracking experiments. In the experiments, the proposed system is compared with the target-tracking system without the occlusion handling method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, the developing of an ultra-small size range image sensor is presented. The sensor is designed for robot hand by using an endoscopy camera and multi-slit laser projector, which can display 3D distance information. The improvements over the previous sensor are size, accessibility and measurement range in close distance. The dimension of our sensor is 16×40×53 mm and only weights about 28 g. By using a universal Webcam chip, the computer can capture the image data without installing the driver. Moreover, the sensor can measure the distance range from 50-150 mm. The effectiveness of the sensor is verified by experiment.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes a stereo camera using two fish-eye cameras. The measurement range of the stereo camera is increased by using fish-eye cameras. Fish-eye images are deformed by using an equirectangular projection to simplify the stereo correspondence search in the images. Affine transformation is applied to images so that feature points of far objects coincide in images of two cameras, in order to remove the mismatches due to the errores of intrinsic and extrinsic parameters. The corresponding points are calculated by using the stereo matching in the deformed image, and the distance is measured from the disparity. The distance measurement accuracy of the constructed stereo camera is evaluated by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method for estimating the pose of a human hand using normal histograms. The method is composed coast to fine of three steps. First, within the database which consist of hand model postures in a reduced degree of freedom space about human hand postures in a daily life, screening of postures is conducted using dispersion of point clouds. Second, the similarity between normal histograms generated from the input posture and the database posture is evaluated with Chi Square. Third, optimization for minimizing the evaluation function is conducted in the reduced parameter space. In the experiment, the hand model was used as input postures. The result of the experiment shows that the maximum estimation error was 21[%] using this method.

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (trade magazine, newspaper, online media)  

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人映像情報メディア学会  

カメラを用いた計測には,カメラのキャリブレーションが必要である.本研究では,特別なキャリブレーションターゲットを用いる必要のない,簡便なキャリブレーション手法を提案する.実環境に存在する平面テクスチャの3次元モデルとカメラから得られるカラー画像の位置合わせを行うことで,カメラの内部・外部パラメータを推定する.この際,2枚の画像の勾配拘束を利用する.実環境に存在する平面テクスチャを模したターゲットを用いて実験を行い,パラメータ推定を行うことができることを示す.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper presents a target tracking system for a mobile robot in both indoor and outdoor environments. A stereo camera, which is robust to sunlight and illumination changes, is used. Human regions in images are detected from 3D information. Using color information, a target region is discriminated from the detected human regions. Hue and saturation are chosen as features robust to illumination changes. Finally, a mobile robot is controlled based on the 3D information of the detected target region. The effectiveness of the proposed system is verified through human following experiments in both indoor and outdoor environments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method for estimating the pose of a human hand by orientation histograms. This method can estimate the pose of a human hand without placing any makers and using motion capture which is much expensive. First, some postures of a human hand model are generated using Dhaiba Hand. After getting the point cloud data of a human hand by using RGB-D sensor, histograms of normals are generated from a hand model and a point cloud. Those orientation histograms are compared with each other with Residual Sum of Squares. The result of the experiment shows that the hand posture could be estimated comparing with 3 postures of hand model using this method.

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (trade magazine, newspaper, online media)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：The Japan Society for Precision Engineering  

DOI： 10.2493/jjspe.79.891

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/10024833669?from=CiNii

Language：Japanese   Publisher：一般社団法人映像情報メディア学会  

近年,ジェスチャによって機器の操作を行うジェスチャインタフェースが注目されており,多くの研究成果が報告されている.本論文では,永易らによって提案された,操作者の位置,姿勢,向きに依らず手振りのみで家電機器の操作が可能なジェスチャインタフェースの改良を行っている.主成分分析に用いる領域を限定することにより,部屋に操作者以外の人物がいる場合でも家電機器の操作を可能とし,コマンド空間の認識率を向上させている.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, a tracking system based on Particle Filter is presented. Color and distance information are used. Distance information are obtained by "Subtraction Stereo", which restricts stereo matching to foreground regions extracted by subtraction. The effectiveness of the proposed system is verified by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method for expanding the measurement range of the range image sensor using multi-spot lights. This sensor can obtain range and color images at 200Hz. Measurement range of the original sensor is 850〜2500[mm]. It is necessary to solve the correspondence problem in order to expand the measurement range. The area of the spot image changes according to the distance. We use the area of the spot image to solve the correspondence problem. The effectiveness of the proposed method is verified by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper we conducted analysis of lifting task focusing on the trajectory of center of pressure (CoP) and posture of fingertip while human subjects lift up, hold and replace an object with their index finger and thumb. CoP was computed from the forces and torques measured by a 6-axis force sensor installed in the lifted device, while posture of the index finger and thumb was measured by a motion capture system and three optical markers attached to the fingernail and thumbnail. The result of the lifting task experiments under dierent friction and weight conditions showed coordination between the grip force and load force. The trajectories of the CoP during the lifting task had similar trend regardless of the experimental conditions, and were divided into two distinct phases. From the data of finger posture, the first phase was speculated due to change in the finger posture coming in contact with the device, while the second phase was speculated due to skin deformation in the direction of the load force.

CiNii Books

researchmap

Language：Japanese   Publisher：一般社団法人映像情報メディア学会  

自動車の運転支援システムやロボットにおいて,一度に広範囲を3次元計測することが重要である.本論文では,3次元計測を行うセンサとして魚眼カメラを用いる.魚眼カメラで取得した画像は大きく歪んでいるが,一度に広範囲を観測でき,自動車等への搭載が容易である.我々は,魚眼カメラを用いたEPI解析による3次元環境復元手法を提案している.本手法を用いることで,特徴点の対応を容易かつ安定に行える.本論文では,精度検証実験により本手法の有効性を検証する.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we proposed a fast and stable human detection based on "subtraction stereo" which can measure distance information of foreground regions. Scanning an input image by detection window is controlled in their window sizes and number using the distance information obtained from subtraction stereo. This control can skip a large number of detection windows and leads to reduce the computational time and false detection for fast and stable human detection. Additionally, we propose template matching using intensity gradients and aSC descriptor which is local features with upper human body models. Experimental results show that the proposal is faster and less false detection than method described in the reference.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

These days, home appliances in our living environment are becoming full of functions. On the other hand, the increase of their functions makes their operation complicated. Therefore, an intelligent room which recognizes gestures and support operators is required in various places. In this paper, we construct a gesture recognition system that considers an operator's posture using a range image sensor. The proposed system uses Support Vector Machine (SVM) to recognize gestures. Several experiments are performed to demonstrate the effectiveness of the proposed method.

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (other)  

CiNii Books

researchmap

Language：Japanese   Publisher：日本工業出版  

CiNii Books

researchmap

Language：Japanese  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

These days, home appliances in our living environment are becoming full of functions. On the other hand, the increase of their functions makes their operation complicated. Therefore, an intelligent room which recognizes gestures and support operators is required in various places in recent years. In this paper, we propose a method to recognize mouth motion from images. The proposed method uses template matching and recognizes mouth motion for indicating a target object in an intelligent room. DP matching is applied to similarity measure obtained by template matching. Several experiments are performed to demonstrate the effectiveness of the proposed method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

We propose a method for personal identification based on hand-waving. The main contribution of this paper is how the people that wave a hand repeatedly can be identified using frequency features. We first construct a time series of hand low-resolution images. Next, frequency features are extracted from time series of hand low-resolution images applying FFT. Finally, our method computes the similarity of model data and test data using cone-restricted subspace method. In several experiments, we try to identify the person who waves a hand and show the accuracy of the proposed method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In car driving support systems and mobile robots, it is important to understand three-dimensional environment widely at once. In this paper, we use a fish-eye camera as a sensor which measures environment. This camera can take a wide-range and distortional image and can be easily mounted on cars. We propose a method for reconstructing three-dimensional environment from fish-eye images based on EPI analysis. Using this method, matching of feature points is easy and stable. The effectiveness of the proposed method is verified by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

An intelligent room which recognizes gestures and support operators is required in various places in recent years. In this paper, we propose a method to recognize mouth motion from images. The proposed method uses shape features and low resolution images of mouth region and recognizes mouth motion for indicating a target object in an intelligent room. DP matching is applied to low resolution images. Several experiments are performed to demonstrate the effectiveness of the proposed method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method for segmentation of multiple human in a projection plane. Our algorithm requires a stereo camera system called Subtraction Stereo, which extracts foreground information with a fixed stereo camera. The main contribution of this paper is how the image sequences that include partial occlusion of the foreground objects can be accurately segmented using mean shift clustering in real-time processing. The proposed method is suitable for inside a medium-sized environment, such as a room and an entrance. Finally, we try to segment the image sequences that include partial and full occlusion and show the accuracy of the proposed method.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a fast and stable human detection method using a stereo camera system called "Subtraction Stereo" with a Histograms of Oriented Gradients (HOG) feature. The proposed method performs scanning detection window using the distance of foreground regions from the subtraction image. Fast and stable human detection is achieved by the HOG feature of foreground regions with on Real AdaBoost classifier. And the experimentally demonstrated that our solution compared normal image with subtraction images. In addition, when we computed the HOG feature, we verified the size of an important cell.

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

DOI： 10.11522/pscjspe.2009A.0.185.0

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：日本ロボット学会  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper discusses a high-speed range image sensor which can obtain textured range images in real time. One of the important applications of this system is an outside sensing for mobile robots. The proposed system consists of a commercially available laser projector and two CCD cameras. The number of pixels of a range image is 361 and the measurement range is 800-2000mm. Experimental results show 40Hz sensing of textured range images. The effectiveness of the proposed system is verified by experiments. As a major conclusion, the constructed sensing system obtains real-time textured range images with commercially available devices.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper reports the influence of finger geometry on mechanics of contact. We synthesized nine representative FE models from geometric database of human finger, and simulated mechanical behavior of finger pad under the same boundary condition for this purpose. The results of the simulation revealed that there is strong correlation between thickness of a finger pad and the maximum von mises stress that arise in the boundary between finger nail and skin.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes skin color registration in an intelligent room. We apply the gesture recognition to the operation of the household electrical appliances. In order to recognize the gestures, the skin color information is used in this system. Existing method we had constructed took time to register the skin color. That means, it was difficult to operate household electrical appliances smoothly. Therefore we developed the new method that improved the skin color registration, it recognize the gestures more smoothly by any skin colors. The new method uses a result of waving hands recognition that identifies an operator. It was confirmed that the method could resister the skin color smoothly and it was valid for an intelligent room system.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In this paper, we propose a method to recognize periodic gestures from images as a key technology for human machine interface. The proposed method uses power spectrum and phase spectrum that are obtained by applying FFT to time series of intensity images. FFT is applied to each pixel of the low-resolution images. The method consists of 2 steps. First, the method detects periodic motion regions from power spectrum. Second, the system uses the phase spectrum in the detected periodic motion region to classify the gestures. The proposed method is robust to lighting condition and individual difference of skin color, because it does not use color information. To demonstrate the effectiveness of the proposed method, we have conducted several experiments and show that support vector machines are well suited for obtaining discriminate hyperplanes for recognition of periodic motion.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes an object detection method using characteristics of spectral reflectance in mid- infrared region. Retroreflective traffic signs are irradiated with infrared light and the reflected light from the traffic signs is taken with an infrared camera. Infrared lights with wavelength of around 1400nm and 1150nm are used, since solar lights do not contain these wavelengths and the detection of traffic signs can be realized without the effect of the sun. The effectiveness of the proposed method is verified by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

We have proposed a stereo camera system "subtraction stereo" that focuses on moving regions. This paper proposes a method to track multiple persons using a sequence of range images obtained by subtraction stereo. Tracking that is robust to occlusion, etc. is realized using Kalman filter. The effectiveness of the proposed method is verified by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In order to synthesize geometric models with individual variations, we created geometric database of human finger from MRI images. Volume registration technique was employed to quantify the individual differences from 3-dimensional MRI images of 50 male subjects. The result of principal component analysis for the measured individual differences (displacement fields) has clarified the principal features of finger geometry such as dimension, thickness and thickness of a finger pad. We can synthesize possible variations of finger geometry and subject specific finger geometry by transforming a template model according to a displacement field computed by weighted sum of these principal features.

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Other Link： http://id.nii.ac.jp/1001/00062371/

DOI： 10.11522/pscjspe.2009S.0.185.0

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：精密工学会  

researchmap

Language：English  

This study aims at developing a practical stereo camera that is suitable for applications such as surveillance, in which detection of anomalies or measurement of moving people are required. In such surveillance cases, targets to measure usually move. In this paper, "Subtraction stereo" is proposed that focuses on motion information to increase the robustness of the stereo matching. It realizes robust measurement of range images by detecting moving regions with each camera and then applying stereo matching for the detected moving regions. Measurement of three-dimensional position, height and width of a target object using the subtraction stereo is discussed. The basic algorithm is implemented on a commercially available stereo camera, and the effectiveness of the subtraction stereo is verified by several experiments using the stereo camera. © 2009 SPIE-IS&amp
T.

DOI： 10.1117/12.805718

Scopus

researchmap

Language：English  

In this paper, a simple and robust method for indoor self-localization is proposed. Infrared LEDs are used as landmarks. They are set at known positions and observed by a CCD camera on a mobile robot. As a consequence, self-localization is carried out. The acquisition of the position and orientation of the robot with two or more LEDs is formulated. The nonlinear least squares method is applied. Experiments with a constructed experimental system evaluate the proposed method. Error analysis is performed by simulations. The proposed method realizes a very simple indoor GPS using only several infrared LEDs and a CCD camera. ©2009 IEEE.

DOI： 10.1109/INSS.2009.5409955

Scopus

researchmap

Language：Japanese  

CiNii Books

researchmap

DOI： 10.11522/pscjspe.2008A.0.215.0

researchmap

Language：Japanese   Publisher：一般社団法人映像情報メディア学会  

本研究では,赤外領域における分光反射特性を利用して物体を検出する手法の提案を行っている.提案手法では,道路標識の反射率が赤外領域で他の物体に比べ比較的高いことを利用している.アクティブに赤外光を照射し,赤外カメラを用いて,太陽光の影響の小さい1150nm,1400nmの波長における画像を撮影し,2波長間の反射強度を用いて標識を検出する手法を提案している.

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：人工知能学会  

CiNii Books

researchmap

Other Link： http://id.nii.ac.jp/1004/00007015/

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes improvement of a method to recognize waving hands using FFT. We constructed an intelligent room that has gesture recognition as a human-machine interface. Recognition of waving hands from images is a key technology of the system to detect an operator in the room. The current method cannot recognize slow waving hands. Therefore we increase the number of sampling frames and add the features to solve this problem. The proposed method is verified by simulations and experiments, and it was confirmed that the method can recognize even slow waving hands.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

Time optimal control method for quadruped walking robots is developed and installed into a practical robot system. Swinging leg is modeled as a two link manipulator whose time optimal control theory has already been established by Bobrow. However, in legged robot systems, supporting legs support their body weight, and the reaction forces from its ground must be inside of their friction corn. Moreover, the ZMP(zero moment point) of the robot is constrained for stable walk. Therefore, time optimal control inputs must be designed considering these constraints. SONY ERS-7 is used as a quadruped walking robot and a fundamental experiment is done. From the experimental results, the effectiveness of the developed control algorithm is verified.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In recent years, active safety technologies of cars are becoming important and the cases that cameras are mounted in cars are increasing. Since the perception of the distance of a usual image is difficult for drivers, method to construct overhead view images and assist drivers are proposed. In this paper, we propose methods for estimating camera parameters of multiple cameras and construct overhead view images with small errors. Effectiveness of the methods is examined by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

In many computer vision applications such as surveillance, targets to detect and measure move. We have proposed a stereo camera system that focuses on moving regions. In the system, what is called the "correspondence problem" of the stereo cameras is solved by restricting the stereo matching process to moving regions. In this paper, the fundamental algorithm to realize robust measurement of stereo images by detecting moving regions with each camera and then applying stereo matching for the detected moving regions is verified, and detection of features of moving regions using the proposed stereo system is discussed and evaluated by experiments.

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper proposes a method for speed-up of indoor self-localization system. Infrared LEDs are used as landmarks. They are invisible and thus do not irritate a human. They are set at known positions, and a CCD camera on a mobile robot observes them and self-localization is carried out. A method to obtain position in two-dimensional plane and orientation with three or more LEDs is formulated. The nonlinear least squares method is applied. Experiments verify stable self-localization when the robot moves at a speed of 50 centimeters per second.

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：映像情報メディア学会  

CiNii Books

researchmap

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：日本工業出版  

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (other)  

researchmap

Publisher：自動制御連合講演会  

DOI： 10.11511/jacc.49.0.239.0

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

This paper deals with a target that a humanoid with a small range image sensor recognizes obstacles and avoid them autonomously while walking. Planar regions, on which the humanoid can walk, and obstacles are distinguished by using a relative disparity image obtained by a small range image sensor. In addition, an integrated system that can feed back obtained obstacle information to the humanoid is constructed. Experiments show that the humanoid can recognize obstacles and stop walking autonomously by the proposed system.

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)   Publisher：The Institute of Electrical Engineers of Japan  

This article has no abstract.

DOI： 10.1541/ieejjournal.125.692

CiNii Books

researchmap

Other Link： https://jlc.jst.go.jp/DN/JALC/00271346326?from=CiNii

Language：Japanese   Publisher：映像情報メディア学会  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

researchmap

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (other)  

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (other)  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

CiNii Books

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (trade magazine, newspaper, online media)  

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (other)  

researchmap

Language：Japanese   Publishing type：Article, review, commentary, editorial, etc. (scientific journal)  

researchmap

Language：Japanese   Publisher：The Japan Society of Mechanical Engineers  

A calibration method for three dimensional coordinate measurement machine(3DCMM) in terms of sequential-two-points method is proposed. Two sensors are attached to the probe of a machine and an Al plate machined by milling machine is placed on the table as the objective, then the probe is moved step wisely with interval of the distance between two sensors by NC indication of the machine along x axis, which makes it possible to identify the straightness error movement of the probe. By repeating the evaluation of the movement stepwisely again along y axis flatness error movement of the probe can be obtained, which is possible to calibrate the positioning accuracy of the probe in the plane. Statistical evaluation of the measuement is conducted, which leads to that reliability of the measurement is raised.

CiNii Books

researchmap

Language：Japanese