This paper proposes a method to search a target image area in an inclined gray-scaled objective image by the template matching. In the template registration phase, a template image is separated into plural block areas. The target image areas are searched in inclined images for learning by using only pixels in the selected partial block areas. The optimal block areas that are useful for the matching in inclined images are selected by using the genetic algorithm. Only block areas that are not so influenced by inclinations of objective images are selected however the total number of pixels that are used for the matching phase decreases. The experimental results show that the number of pixels in the selected block areas decreased to only 13.7% of the total number of pixels in the original template image. The recognition rate reached to 100% in the range of ± 8 degrees of inclined angles. And the over 96% recognition rates were recorded when the inclined angles were ± 10 degrees.

View full abstract

MPEG-4, a new audio-visual coding standard has strong compression tools, such as sprite coding. We have focused on sprite coding and proposed “Sprite Mode”. In sprite mode, video sequence is divided into foreground object and background object. Foreground object is compressed using MPEG-4 object coding, while background object is compressed using sprite coding. Sprite mode does not always adopt any video sequence such that images having still camera motion can be compressed in enough quality in low bit-rate. So, in this paper, we developed new MPEG-4 video encoder using sprite coding. In the proposed MPEG-4 encoder, multi-mode coding is newly proposed. Video shot suitable for conventional coding method is coded by normal mode (conventional method), while video shot suitable for sprite coding is coded by sprite mode. These two modes are automatic selected in any video sequences. We improved the processing flow using stuck system in order to consequently process video such as AVI format. We improved global motion detection algorithm, foreground object and background sprite generation algorithm. Then we proposed the simple rate-control for sprite mode. Then, we have some experiment on MPEG-4 video coding. The result shows effectiveness' of proposed encoder in low bit-rate coding.

View full abstract

Generally, image reconstruction by ultrasonics uses reflective data and observation position unidirectional information. This study examines the method to obtain a high quality reconstruction image from projection data of a few directions. To reduce the number of projection data, the model of the object expresses the wavelet function. As the applied wavelet function to be used has high localization property, the quantity required for reconstruction to calculate decreases. Thereby, high quality images are obtainable by computer simulation.

View full abstract

This paper proposes a new method of extracting moving objects from a video sequence. The proposed method is robust to noise and the intensity change of the observed image, which can be affected by the illumination or the function of a camera. It is basically the background subtraction of the background image which can be estimated by the global change of intensity in the observed image. The method is based on the assumption that pixels of the same intensity in the original background image will keep the same intensity even with a change in intensity of the whole image. A mapping table between the present image and the original background image is generated to estimate the present background image. After the background is subtracted for each pixel, the observed image is divided into some blocks. The mean value of the difference is calculated for each block. Then, the region of moving object is extracted based on the mean value. This blocking process results in the higher robustness for the segmentation. Since the background brightness, which is occluded by the moving object itself, is estimated from the observed image and the background image, the method is applicable to the intensity change of the observed image by varying the illumination under the automatic control function of camera. The performance of the real-time implementation of the proposed approach is evaluated in comparison with the previous approaches.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract