We have a lot of kinds of social infrastructures including lifelines, public facilities, transportation systems, environment systems, and communication systems. They are spending a large labor and cost in maintenance, management, and the operation of them. Along with rapid advancement of technologies, they are applying various computer vision systems to decrease these labor and cost to many scenes of social infrastructures. However, we are not recognizing that so much. This paper introduces some applications related to the social infrastructure systems achieved for the first time by applying our computer vision techniques.

View full abstract

As well known, human sensing is useful for our life. But, the development of the human sensing system is not easy, because the shape and the movement of human are complicated and Indistinct; the environment of obtaining image is not always constant and so on. We have developed some methods of identifying individuals using facial image, counting passing people, observing of human behave and so on. In this paper, human sensing and its applications are introduced. In addition, examples of application are shown.

View full abstract

In this paper, the authors describe a license plate (LP) detection method which can exclude a restriction on LP position, three dimensional rotation and size. In the method, LP reference images, which replaced characters on the plate image by rectangles, are prepared beforehand. Using a frontal LP image, they are calculated by perspective projection at regular direction intervals and are calculated by a processing of expanding and shrinking at regular size intervals. Using an input image of vehicle passing scene and one reference image, a correlation map as a distribution of matching degree is obtained by matched filter. By matched filter with every reference image, the correlation maps corresponding to the reference images are obtained. Detecting the highest value in all correlation maps, the matched reference image and the coordinate as LP position are decided. Furthermore, a shape of LP in the input image is detected by fitting of LP frame which is calculated using the matched reference image. In the paper, the proposal method is explained and an example of the experimental results is shown.

View full abstract

In the past, all methods which understand deformation / fracture (D/F) characteristics have been limited on the surface indirectly. D/F characteristics are impacted by nano-scale structural features like air bubbles (pores); therefore, they need to be analyzed including inside. In this paper, we propose a system which automatically obtains the corresponding relations between pre- and post-D/F pores. Our system enabled analyzing three dimensional, local, high-accuracy D/F characteristics.

View full abstract

This paper presents a new algorithm for feature generation, which is derived based on geometrical interpretation of the fisher linear discriminant analysis (FLDA). This algorithm (Simple-FLDA) is an approximation algorithm that calculates eigenvectors sequentially by an easy iterative calculation by expressing the maximization of variance between classes and minimization of variance in each class without the use of matrix calculation. We carry out computer simulations about recognition of wrist motion patterns by EMG measured from wrist and personal authentications that use face images to verify the effectiveness of this technique. The result was compared with the result of principal component analysis (Simple-PCA).

View full abstract

We propose a novel surveillance system that uses hybrid camera network. The system contains a panorama camera and PTZ cameras to take the wide range images and face images at high enough resolution for identification tasks. The robust human detection methods include robust background subtraction method (RRC), skin color segmentation and face tracking method. First, the system detects persons from a panorama image, and then the detailed face images are obtained with the PTZ cameras. The PTZ cameras can track the faces using the four directional features and the relaxation matching. In addition, the system has automatic camera position calibration feature. Thus, the user can use the system without any troublesome settings.

View full abstract

According to the Gestalt principals, in this paper, we model features for measuring the attraction degree between couples of image components, and grouping areas in trademark images are recognized. This investigation would be used for content-based image retrieval from the view of mirroring human perception for images. The features of proximity, shape similarity, closure, and good continuation are extracted from every combination of two components in an image. After that, according to results of the judgments, a grouping pattern for the query is fixed. Besides, changing combination of the features, the proposed method can output plural grouping patterns. In the experiments, we have evaluated the proposed method on 74 test images comparing between outputs by the proposed method and grouping patterns for the test images obtained from results of questionnaires by 104 participants.

View full abstract

In this paper, high-speed size and orientation invariant lips detection of a talking person in an active scene using template matching and genetic algorithms is proposed. As part of the objectives, we also try to acquire numerical parameters to represent the lips. The information is very important for many applications, where high performance is required, such as audio-visual speech recognition, speaker identification systems, robot perception and personal mobile devices interfaces. The difficulty in lips detection is mainly due to deformations and geometric changes of the lips during speech and the active scene by free camera motion. In order to enhance the performance in speed and accuracy, initially, the performance is improved on a single still image, that is, the base of video processing. Our proposed system is based on template matching using genetic algorithms (GA). Only one template is prepared per experiment. The template is the closed mouth of a subject, because the application is for personal devices. In our previous study, the main problem was trade-off between search accuracy and search speed. To overcome this problem, we use two methods: scaling window and dynamic search domain control (SD-Control). We therefore focus on the population size of the GA, because it has a direct effect on search accuracy and speed. The effectiveness of the proposed system is demonstrated by performing computer simulations. We achieved a lips detection accuracy of 91.33% at an average processing time of 33.70 milliseconds per frame.

View full abstract

In large-scale System-LSI/SoC, how to standardize the communication method between diversified cores becomes one of the important problems, and various study reports have been done.
We propose a new method to solve the problem, where the communication between the cores is limited only to using message + parameters format. As the communication mechanism between IP cores and a common bus, we introduced the architecture, ACU, common to every core with various I/O functions, and it can be realized with less than 5k gates scale.
As the paradigm of Globally-Asynchronous Locally-Synchronous(GALS) system is attracting attention, this method is believed to be a good suggestion as a communication mechanism between local IP cores.

View full abstract

EEG is characterized by the unique and individual characteristics. Little research has been done to take into account the individual characteristics when analyzing EEG signals. Often the EEG has frequency components which can describe most of the significant characteristics. Then there is the difference of importance between the analyzed frequency components of the EEG. We think that the importance difference shows the individual characteristics. In this paper, we propose a new EEG extraction method of characteristic vector by a latency structure model in individual characteristics (LSMIC). The LSMIC is the latency structure model, which has personal error as the individual characteristics, based on normal distribution. The real-coded genetic algorithms (RGA) are used for specifying the personal error that is unknown parameter. Moreover we propose an objective estimation method that plots the EEG characteristic vector on a visualization space. Finally, the performance of the proposed method is evaluated using a realistic simulation and applied to a real EEG data. The result of our experiment shows the effectiveness of the proposed method.

View full abstract

In this paper we present a novel model of noise removal for the human body vibration signals in wireless sensor networks, which assumes which assumes that the noise appears not continuously or periodically but sparsely and it is the summary of some transient signals at discrete time. Furthermore, on the base of the model a new algorithm is developed which takes advantages of wavelet transforms to remove the noise by the method of processing scale functions coefficients at each level. A wavelet block is designed to implement the function of the proposed method. The results of the simulation and experiments indicate that the new algorithm has removed the noise of the signals successfully in the case of both human's one discrete action and continuous activities.

View full abstract

This paper discusses the Channel Acquisition Slot Protocol (CASP), a new protocol for the Media Access Control (MAC) of wireless networks. The CASP is intended to provide a more efficient use of communication channels; this is achieved by mechanisms for eliminating contentions among network stations and by an implicit handling of acknowledge messages. To avoid collisions of messages, the protocol divides the time into short slots, each of which is assigned to a single station, and any station must wait its own slot to start its transmission. However once a transmission is started the slot can be extended and no other station can make use of the channel. The use of an implicit acknowledge mechanism removes the necessity of an acknowledge message exchange for every transmission. In various simulated scenarios the CASP shows a 25% better throughput efficiency compared with the one of the IEEE802.11 protocol.

View full abstract

As many people use digital maps in the ubiquitous network, facilitating fair use has become a great issue for vector map content. Watermark is expected to trace content and notify contact information to the end users. We clarify the technical requirement for the small size map including several hundred vectors. A line-division based digital watermarking system is proposed by inserting new points that divide a line segment, where the ratio of internal division indicates embedding information. The method is evaluated for surviving daily user's operations and tracing content.

View full abstract

To recover 3D-information using two images taken by a moving camera at different positions, we need two information, the location of cameras and relation of correspondences between images. When the camera motion is unknown, we must find all clues from only images. In this paper, we propose a method of depth estimation using the knowledge of camera rotation. We assume that the rotation of a camera is controllable and that approximate information of camera motion is available. The camera motion is described with six parameters. We set initial values of parameters from the information of camera motion. We calculate parameters that minimize the distance between epipolar lines and correspondent points. When an iteration is repeated, we update the values, which must keep a solution that is possible as real camera motion. In the simulation experiment, we confirmed that our method performed effectively when matching error was small, while there were many matching errors using knowledge only. We also examined in real environment and obtained good results. Epipolar lines recovered by our method passed through the correspondent points and lines recovered by knowledge only didn't. It proved that object shapes were generally reconstructed and the proposed method was useful and effective.

View full abstract

In semiconductor wafer manufacturing processes, defect candidates are usually extracted by an inspection system. The defect candidates are composed of true defects such as open circuits, contaminants, and bridge as well as non-defect patterns, called nuisances, which dominate true defects. The goal of this study is to classify the defects candidates into each true defect and nuisance using a small number of training data given by SEM inspection. It is shown that the accuracy of the clustering is considerably improved by introducing qualitative knowledge, which is a priori given by an inspector, in the clustering processes.

View full abstract

In this paper, we investigate a procedure for decreasing the number of false positive findings in a reported Computer Aided Diagnosis (CAD) system for the detection of Ground Glass Opacity (GGO) nodules in chest Computed Tomography (CT) images. The proposed procedure consists of two main stages. The first stage is the application of a Radial Basis Function (RBF) neural network on the CT images using a sliding sub-image window scanned over the region of interest (ROI). The RBF network was trained using both, GGO nodule and false positive samples in order to gain the ability to differentiate between both findings. The second stage involves the examination of coronal CT images to confirm the existence of GGO candidates in the transaxial CT images based on the fact that nodular candidates tend to appear similarly in both sections. The algorithm was applied on 2100 slice images containing 27 GGO nodules representing the majority of the typical findings found in the real clinical practice. It succeeded to achieve a detection sensitivity of 96.3% with False Positive (FP) rate of 0.147 FP/slice in case of using the RBF network alone and further improved to 0.06 FP/slice when applying the RBF network together with the coronal images examination, which proves the potential effectiveness of the proposed algorithm.

View full abstract

Compression artifacts are the results of an aggressive data compression scheme applied to an image that discards some data which is determined by an algorithm to be of lesser importance to the overall contents but which is nonetheless discernible and objectionable to the user. In this paper we present a post-processing algorithm that focuses on restoring the clarity of important image features, such as edges, and removing the compression ringing artifacts around edges at compression ratios of 100:1 and greater. At the decoder, the algorithm extracts the locations of edges from reconstructed image and classifies the pixels surrounding edges for removing compression ringing artifacts, then applies a linear procedure to restore the clarity of the edges. With this algorithm, edges in the images which important for recognition are well restored, and the compression ringing artifacts are removed at very low bit rates.

View full abstract

There are a lot of multi-inputs multi-outputs (MIMO) systems in the chemical plants of the manufacturing, and they have the multiple time delays which are different length in each their inputs and outputs pair. This paper proposes a new technique for the MIMO system with the multiple time delays. Generalized minimum variance control was used for the control design method. The following states of time delays are held in one sampling interval by selecting sampling period. The states between the sampling period can be expressed by using the modified z transform. Thus, the discrete-time model of controlled object is obtained, and MIMO for GMVC is designed by using this model.

View full abstract

Automation is an important issue in industry, particularly in inspection of underground facilities. This paper describes an intelligent system for automatically detecting faulty areas in a sewer pipe system based on images. The proposed system can detect various types of faults and be implemented in a real time system. The present paper describes system architecture and focuses on two modules of image preprocessing and detection of faulty areas. The proposed approach demonstrates high performance in detection and reduction of time and cost.

View full abstract

Message layer security is necessary to ensure the end-to-end security of web services. To provide the confidentiality against the intermediaries along the message path, XML Encryption is used to partially encrypt the message. Because the data structure is changed by the partial encryption, the encrypted message is no longer valid with respect to the original schema definition. So, the problems occur regarding the processing of the schema validation and the data binding by the intermediary. In this paper, we discuss the two possible methods to solve these problems. The first method is to transform the original schema definition. The second is to transform the received message. We examined these methods by applying them to demonstration experiment of web services.

View full abstract

A method to extract time series data about a specific matter related to the business from press releases of a company is proposed. In press releases of a company, a numerical data about a specific matter often co-occurs with specific words. From this view point, when keywords are input, our system extracts a set of numerical data and time stamp that co-occurs with them. Our system extracts numerical data close to keywords based on the word-distance, and time stamps are extracted in the same manner. The expression of extracted data is unified for visualization. Through applied results to the press releases of several kinds of documents, it is confirmed that the accuracy of the extracted numerical data by this proposed method is good enough to support business analysis.

View full abstract

Impedance discontinuity at bend in microstrip line is caused by the geometrical variation at the bend, and it deteriorates the signal integrity and radiates the undesired emission. Further, the phase delay time at the bend is reduced because of the shortcut of current path on the bend. In this paper, several cutting shapes of the bend were investigated in order to improve both characteristics of the S21 and the phase delay time at the bend. As a result, the optimum cutting shape improving the both characteristics was determined.

View full abstract

We propose an optically controlled photothermal switch using a liquid crystal phase grating (LCPG) fabricated by an atomic force microscope nanostroking method. In our experiment, the optically controlled photothermal switching of LCPG was worked at 0.5 Hz with the waveform rise time of less than 150 ms. These results show that this device can be effective as various diffraction optical components.

View full abstract

In case we intend to make use of Hurst parameter H in real time, fast derivation of H becomes a practical requirement. We propose an algorithm to reduce the calculation cost of the cumulative range in an R/S analysis. By installing our algorithm we could reduce the H derivation time less than one second at reasonable data sizes, which provides real time applicability.

View full abstract

This paper proposes a method of speech extraction of a target speaker from one mixed speech signal of some speakers. The proposed method uses a dictionary consisted of feature vectors of a speech signal. The feature vectors are spectra of a segment of the speech signal which contains all phonemes. The dictionary is made in advance from a clear speech signal of the target speaker, and is employed in extracting the target speech signal from one mixed speech signal. The effectiveness of the proposed method has been verified by computer simulations.

View full abstract

To detect the state of a driver while driving, the detection system should be applicable also to a driver who wears glasses for practical use of the system. This paper reports a method of detecting blinks of a driver who wears glasses for estimating driver's level of consciousness.

View full abstract