[in Japanese]

View full abstract

This paper describes a method of digitally watermarking image signals based on the wavelet transform. Information data can be embedded in the lowest frequency components of the image signals by using a controlled quantization process. The data are then extracted by using both the quantization stepsize and the mean amplitude of the lowest frequency components without having access to the original image. Numerical experiments show that this method provides a high-quality watermark that is robust against common signal processing.

View full abstract

We propose a method to detect face regions by using the standard skin color which is taken from an image in a color histogram. Extraction of the face regions is based on an area in motion and the deviation from the standard skin color obtained by discriminant analysis. The standard skin color is updated from information obtained from the detected face region. To judge a face, we use templates composed of directional features and color features to detect eyes and mouth. In the face detection experiment, our algorithm could detect a face region by using motion of a subject's head. In addition, the updated standard skin color was correct. In the facial parts detection experiment, by updating the templates, our method worked better than the method that used only directional features. Consequently, our algorithm can extract the facial features even when a person tilts his or her head. With a SGI/O 2 workstation, it is possible to achieve a speed of approximately 0.25 sec/frame.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract

We are investigating the important physical factors needed for the reproduction of the high-order sensation in a display of images. Gamma is closely related to the reproduction of gradient and color. In this paper, we describe gamma in detail. We expand the traditional gamma defined only in the range in which a gamma value is constant to a non-linear gamma which depends on a signal level. We found that a non-linear gamma is not given by a differential coefficient of the characteristic curve but by a complicated function. And we describe the relationship between the definition of gamma in television and photography.

View full abstract

The induction field in vision can be an index for legible display of Japanese characters. The traditional display is based on types of characters, character size and the distance between characters. There is no method, however, that provides an index for the legible display of any types of characters. It was suggested recently that the interaction in the induction field by characters relates to the legibility of characters. It is thought that the strength of the induction field which surrounds each character may be an index for the legibility of characters. To confirm this theory psychological experiments were done. This paper describes these tests and it is confirmed that the strength of the induction field can be an index for the legibility of characters.

View full abstract

A color quantized image is represented by providing each pel with an index corresponding to a representative color of the pel. There must be a one-to-one correspondence between representative colors and their respective indices, and this correspondence can be designated freely. To encode efficiently a color quantized image using conventional predictive coding schemes such as JPEG lossless coding, this paper proposes a new method which assigns an almost optimal index to each representative color on the basis of evaluation parameters that indicate approximately the predictive efficiency or coding rates. Although the method renews repeatedly assignment of indices, the introduction of a two-dimensional histogram of the indices reduces remarkably the amount of computation required for the repetition. Simulation results show that the application of the method to JPEG lossless coding decreases the coding rates of 8-bit color quantized images by 0.39-0.80 bits/pel and that these coding rates are 0.78-1.81 bits/pel lower than those of GIF.

View full abstract

A high speed 2-D motion vector detection on CMOS sensor focal plane is proposed. Edge detection circuit composed of two crossed differential OTAs with time-multiplexed sampling is adopted for getting horizontal and vertical edges. 4-bits short-time digital memory is designed by transmission gate array, which can keep edge information by a smaller layout area. High speed block matching is designed by a Local Parallel and Global Collumn Parallel (LPGCP) processing architecture, which makes use of the parallel nature of images. The size of block for matching is reduced to 2×2 pixels and a search area of (±1, +1) pixels at a high frame rate, such as 1000 frames/sec. Experimental results of pixel circuit with PD, edge detection and memory are given.

View full abstract

[in Japanese]

View full abstract

[in Japanese]

View full abstract