Search

In this paper, we introduce a method of moving insects tracking. Our method is robust image processing to change of illumination under orientation code matching. Further, based on extended Kalman filter, it can track moving insects which are doing the random motion. Through experiments, this paper shows the effectiveness of proposed method.

The purpose of this study is to develop a method to control power-assist systems with an energy regeneration system that achieves the desired load-reduction ratio even if torque saturation occurs. In this article, we use the ratio of the assist-force impulse to the operator-force impulse as an indicator to evaluate the control method. The proposed method predicts the needs of the operator and delivers the desired load-reduction ratio by anticipating deficient impulses via the reliability of predicted power requirements.

This paper develops a vision-based displacement measurement system for remote monitoring of vibration of large-size structures such as bridges and buildings. The system consists of one or multiple video cameras and a notebook computer. With a telescopic lens, the camera placed at a stationary point away from a structure captures images of an object on the structure. The structural displacement is computed in real time through processing the captured images. A robust object search algorithm developed in this paper enables accurate measurement of the displacement by tracking existing features on the structure without requiring a conventional target panel to be installed on the structure. A sub-pixel technique is also proposed to further reduce measurement errors cost-effectively. The efficacy of the vision system in remote measurement of dynamic displacements was demonstrated through a shaking table test and a field experiment on a long-span bridge.

This paper describes a real-time stereo vision sensor based on the Field Programmable Gate Array (FPGA). It was developed using an image matching algorithm called Orientation Code Matching (OCM). Our implementation, using OCM, provides not only high performance, but also flexibility and a greater robustness, all of which are necessary to embed vision sensors in mobile robots and vehicles. The prototype sensor we implemented runs stereo matching with images of 752 by 480 pixels via image sensing chips, at a frame rate of 60 fps. The maximum disparity, using the current design, is 127 pixels at an operating frequency of 53MHz on the FPGA. By changing the parameters of the hardware description, the architecture of our design on the FPGA becomes flexible and expandable to the required size of input images and pixel disparity.

In this paper, we report the development of a power-assisted valve for fire engines. In Japan, the number of firefighters is decreasing year by year; therefore, it is important to reduce the amount of labor involved in firefighting. Although the development of an automatic system is possible in strictly technical terms, firefighting regulations in Japan require that the valve be operated by hand. In this study, we aim to develop a power-assisted system that can be controlled in accordance with the intent of the operator. First, we confirm a relationship between the impedance parameters and valve opening time. From this relationship, we propose a method for predicting the valve opening time based on the initial value of the operation torque. Evaluating this method, we found that the average operation torque and average tracking error are improved by a factor of two. Second, we reduce prediction error by using a real-time prediction method. While a prediction method based on initial torque can adjust parameters only once, the proposed real-time prediction method can update parameters continuously. As a result, the evaluation score of the method improved from 0.55 to 0.35, and average prediction error was improved from 295ms to 205ms.

We propose a robust algorithm for matching three-dimensional dot cloud data in an effort to detect defects during manufacturing processes. We apply our proposed method to inspect a complex three-dimensional die-cast product. Our approach recognizes the difference between two data sets as a defect after matching the data sets. Moreover, our method improves matching accuracy by detecting geometrical features such as edge points, and by using such property values as gradients. Fundamental experiments using real three-dimensional dot cloud data show that the method is effective as a defect inspection system.

Camera vibration in recorded video is a problem with both handheld and stationary devices that vibrate due to wind and other external forces. To facilitate the development of a robust stereo system that is immune to such camera vibration, this study aims to study establishes an error model for a stereo measurement system by considering camera vibration. In stereo measurements, we approximate the distribution of measurement error in images as a uniform distribution. Given this approximation, parameters of the distribution of measurement error are mean value μ and standard deviation σ, whereas amplitude A and frequency F are parameters of camera vibration. To verify the relationship between these parameters, we performed experiments using a vibration testing system. Results indicated that the vibration did not affect the mean value μ. We also found a positive correlation between standard deviation σ and amplitude A as well as a negative correlation between standard deviation σ and frequency F. Based on these findings, we estimate the parameters of measurement error by the camera vibration parameters and establish our proposed error model.

This paper describes semi-active assist control in the form of a smart suit for horse trainers. We have developed a soft and flexible power assist device named ”smart suit” for reducing user's physical fatigue. The smart suit is a semi-active and flexible power assist devices. The semi-active assist mechanism consisting of elastic materials and a small DC motor is our original technology. The assist force is gained by the elastic force of elastic materials, and it is controlled by adjusting the length of the elastic materials by using DC motor. In this study, we applied the smart suit to horse trainers for reducing their physical fatigue in horse training to prevent their injuries. Moreover we aim developing an assist force control method based on the energy control for considering not only the power assist but the skill assit. In this paper, we confirm the energy control has the effectiveness of the power assist in the situation without the skill assist.

Automating robot laser welding requires that laser irradiation direction and laser focus positioning influencing welding strength be controlled precisely along the seam line. Since the edge of the welded object can be transformed in processing by such as cutting, bending and grinding, direct measurement is necessary to detect the seam line. A robust image detection method of seam lines is proposed for hand-eye laser welding system equipped with an optical microscope, a CMOS camera and two slit lasers for measurement. The seam line is detected indirectly to utilize the difference between local and global distribution of image of measurement lasers. Results of experiments demonstrated the effectiveness of our proposal.

In this study, we propose a multi-portal human interface (M-HI) as an innovative human interface for power assist systems. M-HI allows users to apply force anywhere in the power assist system. This paper describes the concept, the design, the algorithm and possible applications of M-HI. In addition, this paper demonstrates the effectiveness of M-HI with respect to expanding the maneuverable workspace of power assist systems containing multi-DOF manipulators.

Due to the vastness of the vineyards in Hokkaido, it is very hard for its workers to eradicate weeds. In order to solve this problem, we aim to develop a weeding robot with a multi-link manipulator, using a visual sensing technique of stereo camera system for recognizing the trunk of grape. However this technique lacks accuracy because the stereo camera system generally cannot avoid the error in measurement. Considering this, we devised a method which defines the grape-trunk area. Additionally, in order to compensate the visual sensing technique, we used a tactual sensing technique using the multi-link manipulator, which can recognize trunk of grape by touching it. In this paper, we developed a robust sensing system using both visual and tactual recognition for the weeding robot.

In this paper, we propose a velocity sensor for agricultural vehicles by sensing ground motion with robust image matching algorithm called Orientation Code Matching (OCM). The sensor performs velocity measurement with high accuracy and robustness without error by slipping in the fields. The algorithm of OCM is implemented on a Field Programmable Gate Array (FPGA) to perform real-time image matching and to reduce the size and power consumption. Because of low Power consumption, the sensor can be operate within the hermetically-sealed chassis so that it has high resistance to the environment. The effectiveness of the sensor has been verified by agricultural vehicles installed with the sensors in real agricultural fields.

This paper describes a novel and robust method of estimating optical flow for underwater image sequences. This method can give out reliable optical flows output, even if the quality of image is very poor. In order to estimate optical flow of current templtate, first, we divide the template into some sub-templates, and then copute the similarity profiles of each sub-template. These similarity profiles can be used to extract two voting roles: positive voting role and negative voting role. The positive voting role can be used to increase correct optical flows, and the negative voting role can be used to reduce incorrect optical flows. We use two voting qualification variable (TP, TQ) to control the positive and negative voting processing, and use a SNR value to evaluate the each voting result. The estimated optical flow is reliability when SNR value converge to infinity only by useing small TP and TQ.

We successfully develop a defect inspection method based on a robust method for matching the distance between points in three dimensions. The three-dimensional distance data of an object is measured by means of a laser range finder. The data is compared with the measured data of a high-quality item. Then, we examine the differences between two sets of data in order to detect defects in the target object. The three-dimensional distance data is matched with high robustness by using the proposed method. Furthermore, we attach labels to sets of points corresponding to a detected defect. By performing an experiment with real data, we show that a high-quality object and a defect object can be distinguished on the basis of the features of each label.

Nighttime images of a scene from a surveillance camera have lower contrast and higher noise than their corresponding daytime images of the same scene due to low illumination. Denighting is an image enhancement method for improving nighttime images, so that they are closer to those that would have been taken during daytime. The method exploits the fact that background images of the same scene have been captured all day long with a much higher quality. We present several results of the enhancement of low quality nighttime images using denighting.

This paper proposes a novel focus measure based on self-matching methods. A unique pencil-shaped profile is identified by comparing the similarity between patterns extracted around their neighborhood in each scene. Based on this, a new criterion function, CPV, is defined to evaluate focused or defocused scenes. OCM is recommended due to its invariance with regards to contrasts. Experiments using a telecentric lens are implemented to demonstrate the efficiency of proposed measure. Comparing OCM-based focus measure with conventional focus measures shows that OCM-based CPV is robust against illuminations. Using this method, pan-focused images are composed and depth information is represented.

This study aims to establish an error model of the stereo measurement system considering camera vibration. At first, we verified the distribution of disparity error under the circumstance without the camera vibration and with the camera vibration. As the result, we found that we can approximate the distribution of disparity error by normal distribution under the circumstance without camera vibration and with camera vibration. And, the parameters of normal distribution are changed by the camera vibration. The parameters of the distribution of the measurement error are average μ and standard deviation σ. The parameters of the camera vibration are considered amplitude A and frequency F. In order to verify relationships during the parameters of the distribution of measurement error and the parameters of the camera vibration, we experimented using the vibration testing system. We imposed simple harmonic motion to the stereo camera. In this paper, we use stereo camera Bumblebee. As the result of experiment, the camera vibration didn't affect average μ. We found positively correlation between standard deviation σ and amplitude A. And, we found negatively correlation between standard deviation σ and frequency F. We estimate the parameters of measurement error by the parameter of the camera vibration using these relationships. So, we establish the error model of the stereo measurement system. Moreover, we define existing probability of object using the parameter of measurement error.

Due to the area of the vineyard in Hokkaido is extremely large, it is very difficult and hard to eradicate weeds by human being. In order to solve this problem, we developed a dynamic image measure technique, which can be applied to the weeding robots in vineyards. The outstanding of this technique is that it can discriminate the weed and the trunk correctly and efficiently. Meanwhile, we also attempt to measure the root of trunk accurately. And a new method to measure the blocked trunk of grapes in vineyards has also been developed in this paper.

Image processing method that detects a particular moving object from an image by a fixed camera and tracking is noticed in various fields and it is a very important subject. In this paper, we propose a moving object tracking method that can cope with change of the area accompany the random walk movement of the moving object oneself and change of the brightness arise from change of the environmental such as a masking or change of the illumination. Proposal method can be robust processing for change of the illumination based on Orientation Code Matching that is demonstrated that is robust for the masking or change of the illumination. And, using Motion Vector derived from a continuity of the random walk model motion, under the condition that there are similar walk models, it can discriminate the walk model and individually tracking. Through the some experiment, this paper inspects the effectiveness of our proposed method.

We have been developing ldquoSmart Suitrdquo as a soft and light-weight wearable power assist system. A prototype for preventing low-back injury in agricultural works and its semi-active assist mechanism have been developed in the previous study. The previous prototype succeeded to reduce about 14% of average muscle fatigues of body trunk in waist extension/flexion motion. In this paper, we describe a prototype of smart suit for supporting waist and knee joint, and its control method for preventing the displacement of the adjustable assist force mechanism in order to keep the assist efficiency.

To maintain the safety operation for the crane track, it is important to measure the high accurate three-dimensional position of a crane hook. However, it is not easy to obtain the position information of the swinging crane hook by using sensors. In this paper, we introduce the measurement method of three-dimensional position of the moving crane hook based on the image data obtained by one single camera. First, we treat the detection method of the area of the crane hook by using the optical flow, and then track the detected area based on orientation code matching. Through the experimental results, we show the usefulness of our proposed method.

This paper proposes a novel scheme for image focusing by introducing a new focus measure based on self-matching methods. A unique pencil-shaped profile is identified by comparing the similarity between all patterns extracted around the same position in each scene. Based on this profile, a new criterion function called Complementary Pencil Volume (hereafter CPV) is defined to evaluate focused or defocused scenes based on similarity rate of self-matching, which visually represents the volume of a pencil-shaped profile. Among matching methods, Orientation Code Matching (hereafter OCM) is recommended due to its invariance with regards to illumination and contrasts. Several experiments using a telecentric lens are implemented to demonstrate the efficiency of proposed measures. Outstandingly, comparing Orientation Code Matching-based (hereafter OCM-based) focus measure with conventional focus measures shows that OCM-based focus measure is robust against changes of illuminations and contrast. Using this method, depth is measured by comparing the focused and defocused region in the scenes both under high and low illumination conditions.

This paper has proposed a novel and robust scheme of image focusing by introducing a new measure of focusing based on Orientation Code Matching.(hereinafter,OCM) A unique pencil-shape profile has been found through comparing the similarity between any patterns extracted at the same position within their own scenes. Based on this profile, a new evaluation function, named Complemental Pencil Volume, (hereinafter, CPV), is defined and calculated to represent local sharpness of images, either in or out of focus. Then through a designed local-scope searching algorithm, the maximum of CPV, which corresponds to the just focused image, is searching in this schema. The proposed method of focusing can be applied to cases of ill-condition with low contrast observations. Experiment results show that the OCM-based focusing is very robust to change of brightness and even more irregularities in the real imaging system, like dark condition and so on.

In a conventional power assist system, an operation direction and power of a manipulator were adjusted when operator looked at a state and position of an operation object. So, position of an object is acquired by a visual sensor, and not only follows operation of human, but sets up a work route between an operation object and a manipulator. And an end-effector is guided to an object so that an operator can operate manipulator smoothly to an object. In this paper, the work path creation using the potential that suits an operator is proposed. A near work path is set up according to an end-effector orbit that an operator draws, and by guiding to an object aims at the improvement in maneuverability of a power assist.

We propose a novel three-Dimensional measurement approach of flexible cables for factory automation appliations, such as cable handling, connecter insertion without conflicts with cables by using robotic arms. The approach is based on motion stereo with a vision sensor. Laser slit beams are irradiated and make landmalks on the cables to solve stereo correspondence problem efficiently. These landmark points and interpolated points having rich texture are tracked in a image sequence, and reconstructed as the cable shape. For stable feature point tracking, a robust texture matching method which is Orientation Code Matching and tracking stability analysis are applied. In our experiments, arch-like cables have been reconstructed with an uncertainty of 1.5 % by this method.

We propose a novel three-dimensional measurement approach of flexible cables for factory automation applications, such as cable handling, connecter insertion without conflicts with cables by using robotic arms. The approach is based on motion stereo with a vision sensor. To solve stereo correspondence problem efficiently, feature point extraction from cables and a robust texture matching based feature tracking is applied. Additionally, by using stability analysis of the feature point tracking, defective feature points can be rejected and stable 3-D shape reconstruction is achieved. In our experiments, shapes of arch-like cables have been measured to an accuracy of 5 mm. It is enough to find connecters without conflicts with the cables.

Many researches for modeling the shape of object have been treated. In this paper, we propose a new method of modeling reconstruction with high accuracy. Then, we introduce the silhouette method based on radial reach filter. From this method, we make the precise three-dimensional shape of object. Finally, the usefulness of our proposed method is shown through some of experiments.

This paper has proposed a novel and robust scheme of image focusing by introducing a new measure of focusing based on orientation code matching. (hereinafter, OCM) A unique pencil-shape profile has been found through comparing the similarity between any patterns extracted at the same position within their own scenes. Based on this profile, a new evaluation function, named complemental pencil volume, (hereinafter, CPV), is defined and calculated to represent local sharpness of images, either in or out of focus. Then through a designed local-scope searching algorithm, the maximum of CPV, which corresponds to the just focused image, is searching in this schema. The proposed method of focusing can be applied to cases of ill-condition with low contrast observations. Experiment results show that the OCM-based focusing is very robust to change of brightness and even more irregularities in the real imaging system, like dark condition and so on.

We propose a new method for simultaneous joint torque and angle recognition in dynamic motion by using ultrasound echo. Test Feature Classifier has been employed as a classifier. Ultrasound is emitted from the surface of skin and it is reflected inside of living body. Feature values for the recognition are extracted from the reflected wave. The feature value changes as the movement of the muscle relating to the target joint motion. Therefore the joint movement can be recognized based on the feature value. We carried out a experiment of the recognition of the movement of human elbow joint. The task was to classify the test data set into its correct torque class and angle class based on the feature values. The feature value was the rectified integration value on each 6 blocks cut out from the ultrasound echoes. The rate of correctly classified data was 0.74 in the result of the torque recognition. That of in the angle recognition was 0.82. Some data were classified into incorrect classes, however, these data were classified into classes near the correct class. This results shows that the recognition by the proposed method were successful.

Due to the area of the vineyards in Hokkaido is extremely large, it is very difficult and hard to eradicate weeds by human being. In order to solve this problem, we developed a new dynamic image measure technique, which can be applied to the weeding robots in vineyards. The outstanding of this technique is that it can discriminate the weed and the trunk correctly and efficiently. Meanwhile, we also attempt to measure the root of trunk accurately. Due to techniques so far cannot suffice today's applications; a new method to measure the blocked trunk of grape in vineyards has also been developed in this paper.

This study aims to develop an environment recognition method for a mobile robot using visual information in irregular ground. A mobile robot can autonomous movement, an environment recognition and object tracking by using visual information. But researches using visual information do not consider irregular ground. Visual information have errors in irregular ground. There are handling method by image processing technology, but that have limitations. So, We realize an environment recognition using visual information in irregular ground. We describe a relationship desparity errors between measurement errors of stereo method. Next, we measure distribution of disparity errors and propose correction method by disparity errors. We experiment moving on flat surface and irregular ground and verified effectiveness correction method by disparity errors.

This paper has mainly discussed about two problems, object focusing and depth measurement. First, we propose a novel and robust scheme of image focusing by introducing a new measure of focusing based on Orientation code matching. A new evaluation function, named Complemental pencil volume, CPV, is defined and calculated to represent local sharpness of images, either in or out of focus, by comparing the similarity between any patterns extracted at the same position within their own scenes. An identified and unique maximum or peek, which of ill-condition scenes with low contrast observations. Experiments show that the OCM-based focusing is very robust to change in brightness, and to even more irregularities in the real imaging system, like dark condition. Second, based on this robust focusing technique, we applied it to an image sequence of an object surface to measure the depth of profile. A simple plane object surface has been implemented to demonstrate the basic approach. The results showed the successful and precision depth measurement of this object.

Robot vision is an important robotics and mechatronics technology for realizing intelligent robot systems that work in the real world. Recent improvements in computer processing are enabling environment to be recognized and robot to be controlled based on dynamic high-speed, highly accurate image information. In industrial application, target objects are detected much more robustly and reliably through high-speed processing. In intelligent systems applications, security systems that detect human beings have recently been applied positively in computer vision. Another attractive application is recognizing actions and gestures by detecting human – an application that would enable human beings and robots to interact and cooperate more smoothly when robots observe and assist human partners. This key technology could be used for aiding the elderly and handicapped in practical environments such as hospital, home, and so on. This special issue covers topics on robot vision and motion control including dynamic image processing. These articles are certain to be both informative and interesting to robotics and mechatronics researchers. We thank the authors for submitting their work and for assisting during the review process. We also thank the reviewers for their dedicated time and effort.

This study aims to expand a measuring range of stereovision system. In the previous paper, the authors achieved a 3D motion capture system by using one camera with triangle markers, named as Mono-MoCap (MMC). MMC has two features. One is that MMC can measure 3D positions of subjects by using one camera on the basis of the perspective n point (PnP) problem. Another is that MMC does not need to recalibrate the camera parameters. In this paper, the authors apply MMC to binocular stereovision system for expanding its measurement range. MMC will solve the problem that stereovision system can not measure 3D positions of objects when even one camera does not capture the objects. In this study, 3D positions of three points where a geometrical relation each other is already-known will be measured by a stereovision. 3D position measurement by the stereovision is enabled by secondarily using MMC when a part of the object is hidden with one camera of the stereovision. Simulation and experimental results will show the effectiveness of the proposed method for expanding the measuring range of stereovision system.

Quality of local regions in the scene could be deteriorated by the effects of ill-conditioned lighting effects and reflection. We develop a method to improve the quality of local regions. The local region of the image is firstly selected by an algorithm that is based on similarity evaluation of the region. Then the histogram of brightness and saturation of the local region are expanded using the histogram equalization, both the brightness and saturation are improved. In the next step, based on the distance from the point selected by user, the improved data are combined with the original image. The local part of the image is naturally merged with the surrounding scenes.

Instead of tachometer-type velocity sensors, an effective method of estimating real-time velocities based on a robust image matching algorithm is proposed to measure real-time velocities of agrimotors or working machines, such as sprayers and harvesters, driven by them in real farm fields. It should be precise even they have any slipping, and stable and robust to many ill-conditions occurred in the real world farm fields, where the robust and fast algorithm for image matching, Orientation code matching is effectively utilized. A prototype system has been designed for real-time estimation of velocities of the agrimotors and then the effectiveness has been verified through many frames obtained from the real fields in various weather and ground conditions.

Recently, developing of image processing method which enables to track to moving objects on time series images taken by a fixed camera is one of important subjects in the field of machine vision. Here, we try to consider influences by change in brightness and change of region caused by moving objects, respectively. In this paper, we introduce a new tracking method which can be reduced the influences by those changes. First, we use Radial Reach Filter in order to detect the moving objects. In addition, the moving objects can be tracked by an image processing based on information obtained by applying RRF and block division. Further, we propose a method in the case that changes size of moving object by time progress. Finally, through experiments we show the validity of our proposed method.

We propose self-maintenance robot system as a method which realizes work for a long time without maintenance by the human workers. This system absorbs the change which occurs in robot's hardware by learning, and maintains working ability. We use a neural network which has a task common layer and a task independence layer to learning. In this report, we verify of this learning system by the computer simulation and the experiment in the real field

In this paper, we develop an observation device to measure a 3D position for a moving object by using a laser range finder and a CCD camera. Then, we propose new method for the object recognition and the tracking control, respectively. As for the recognition, we use a special mark which is called the cross mark. For the tracking control, we construct PID control with an extended Kalman filter to realize control system without delay. Through some experiments, we verify performance of observation device and show availability of our proposed method.

For ultrasonic muscle activity sensor (UMS), a signal processing method was proposed. The proposed method improved correlation coefficient between estimated joint torque and actual torque from 0.987 to 0.999 in an experiment. UMS is a brand-new noninvasive sensor for muscle activity. In this sensor, ultrasound is emitted and detected by transducers, though electrical change is detected in electromyogram (EMG) which is a common sensing method for muscle activity. In previous study, we measured only one ultrasound from transmitter on opposite side. In this study, we made a transmitter switching circuit and measured ultrasound coming from another transducer. Knee joint torque was estimated with neural networks in experiment. Correlation coefficient between actual torque and estimated torque was calculated as the index of the accuracy. New method (duplex UMS) showed higher correlation than existing method. The proposed method is effective to improve the accuracy of the torque estimation

We have designed the ultrasonic muscle activity sensor as a small sensor device, and the sensor can be built in the sensor suit for human machine systems. Measuring a change of the ultrasonic in muscles with this sensor, we have shown it's possible to estimate muscle force. In this study, we aim to design the algorithm that estimates the joint torque of knee with ultrasonic muscle activity sensor. Beside, in order to attain the generality of sensor suit, we also aim to examine the influence of constitutional differences, such as a physique index and a rate of body fat. Furthermore, we tried to obtain the basic data in order to the construction of system which can adjust individual difference.

Technique of image processing to detect moving objects on time series images is important subject in the fields of environmental monitoring and rescue system which used a mobile robot. In the past, it is difficult to detect the moving objects by the effects of changes in brightness which caused by the sunlight and measure positions of the moving objects by the change of region of the moving objects. Then, we proposed a detection and tracking method of the moving objects considering the renewal of a background based on Radial Reach Filter. However, the change of size of the moving objects on the image, that is, the movement in direction of an optical axis was not considered. Then, we propose a method which compares the features of pixels between the images in the change of size of moving object for calculation of motion vectors. Through the experiment using the real images, this paper shows validity of proposed method.

In recent years, many researches for reconstructing the shape of object which has the complicated structure such as a plant and a statue have been treated. In this paper, we propose new method of three-dimensional modeling with high accurate position data. In this paper, there are two main contributions. First, we introduce the improved silhouette method based on radial reach filter and opacity. Through this processing, we can make precise three-dimensional shape of object. Secondly, we compute three-dimensional co-ordinates value of object on image by using the stereo matching method. Finally, we combine three-dimensional shape obtained by the improved silhouette method and three-dimensional position data measured by stereo matching, and then can show three-dimensional shape with high accurate position data. The utility of our proposed method is proven through experiments

A novel successive learning algorithm is proposed for efficiently handling sequentially provided training data based on Test Feature Classifier (TFC), which is non-parametric and effective even for small data. We have proposed a novel classifier TFC utilizing prime test features (PTF) which is combination feature subsets for getting excellent performance. TFC has characteristics as follows: non-parametric learning, no mis-classification of training data. And then, in some real-world problems, the effectiveness of TFC is confirmed through way applications. However, TFC has a problem that it must be reconstructed even when any sub-set of data is changed. In the successive learning, after recognition of a set of unknown objects, they are fed into the classifier in order to obtain a modified classifier. We propose an efficient algorithm for reconstruction of PTFs, which is formalized in cases of addition and deletion of training data. In the verification experiment, using the successive learning algorithm, we can save about 70% on the total computational cost in comparison with a batch learning. We applied the proposed successive TFC to dynamic recognition problems from which the characteristic of training data changes with progress of time, and examine the characteristic by the fundamental experiments. Support Vector Machine (SVM) which is well established in algorithm and on practical application, was compared with the proposed successive TFC. And successive TFC indicated high performance compared with SVM.

This paper presents a novel fast and high-accurate 3-D registration algorithm. The ICP (Iterative Closest Point) algorithm converges all the 3-D data points of two data sets to the best matching points with minimum evaluation values. This algorithm is broadly used, because it has good availability to many applications. But, it needs many computational costs and it is very sensible to error values. Because, it uses whole data points of two data sets and least mean square optimization. We had proposed the M-ICP algorithm, which is an extension of the ICP algorithm based on modified M-estimation for realization of robustness against outlying gross noise. The proposed algorithm named HM-ICP (Hierarchical M-ICP) is an extension of the M-ICP with selecting region for matching and hierarchical searching of selected regions. In this algorithm, we select regions using evaluation of variance for distance values in the target region and homogeneous topological mapping. Some fundamental experiments utilizing real data sets of 3-D measurement show effectiveness of the proposed method. We achieved more than 4-digits number reduction of computational costs and confirmed less than 0.1% error to the measurement distance.

A rotation-invariant template matching scheme using Orientation Code Difference Histogram (OCDH) is proposed. Orientation code features based on local distributions of pixel brightness are substantively robust against furious change in illumination plays a main role in designing the rotation-invariant matching algorithm. Since every difference between any pair of orientation codes is invariant in rotation of an image, we can elaborate a histogram feature by use of the differences, which can aggregate effective clues for searching rotated images through simple procedures. With gray scale images as targets, rotation angles of an image can be accurately estimated by the proposed method. It is fast and robust even in presence of some irregularities as brightness change by shading or highlighting. We propose a two-stage framework for realizing the rotation-invariant template matching based on OCDH. In the first stage, candidate positions are selected through evaluation of OCDH at every position, and then in the second stage, they are tested by use of a verification also based on orientation code features. The effectiveness of the proposed matching method has been shown through many kinds of experiments designed with real world images.

We propose an efficient template matching algorithm for binary image search. When we use template matching techniques, the computation cost depends on size of images. If we have large size images, we spend a lot of time for searching similar objects in scene image to template image. We design a scanning-type upper limit estimation that can be useful for neglect correlation calculation. For calculating the scanning-type upper limits, template and scene images are divided into two regions: R-region and P-region. In R-region, an upper limit of correlation coefficients can be derived as an interval estimation based on mathematical analysis of correlations of the object image and a pivot image. In P-region, another upper limit is formalized based on the number of white and black pixels in a template and the object image. By use of these upper limits, the scanning-type upper limit estimation of correlation coefficients can be formalized for the efficient matching algorithm. This upper limits estimation isn't over true values of correlation, so the accuracy of search by conventional search is the same as one by conventional search. The experiments with document images show the effectiveness and efficiency of the proposed matching algorithm. In these experiments, computation time by the proposed algorithm is between 5 and 20% compare of the conventional search.

In this paper, we propose a new method of object detection. In the past, there are various methods of object detection. Especially, the method of the background subtraction has the effectiveness. However, the methods based on brightness differences are easily influenced by change in lighting condition. In this paper, we use Radial Reach Filter (RRF). RRF is called as the effective method of the change in lighting conditions. However, RRF is not considered change that caused by moving objects on the background image. Then, we propose the new method of object detection that considered motion of the moving objects on the background image. And, we verify the effectiveness by the experiments using a time series image.