An adaptive orthogonal transform coding algorithm utilizing the classification technique is presented. Coding efficiency for natural images can be improved by adjusting coding parameters in accordance with the local property of images. “Classification” is used to categorize small areas of an image into classes based on their characteristics. High coding efficiency results from changing coding parameters adaptively according to the classification index. Classification methods using ac energy, binary pattern and vector quantization index are compared, and the advantage of the classification method with the vector quantization technique is shown. Also, an orthogonal transform coding algorithm with an adaptive variable length coding method is proposed, and its structure and characteristics are described. A coding parameter normalization method to avoid mismatches between input images and coding parameters is also described. Coding experiments show excellent performance of this algorithm, for example, the number of bits for obtaining 40‐dB SNR of monochrome “GIRL” is 0.65 bits/pel, which is 20 percent smaller than conventional methods. Copyright © 1989 Wiley Periodicals, Inc., A Wiley Company

A motion picture coding algorithm using motion-compensated interframe prediction and the adaptive discrete cosine transform (DCT) encoding technique is proposed. High coding efficiency is obtained by the adaptive DCT encoding technique in which encoding parameters are fitted to widely varying characteristics of the interframe differential signal. Segmented DCT subblocks of interframe prediction error are classified into categories based on their coefficient power distribution characteristics. The adaptation gain results from using a suitable variable word length code set designated by the above classification for encoding each quantization index of DCT coefficients. In addition, a new coding parameter control method is introduced based on the information rate estimation of the current frame. This classification promotes high stability because good estimation accuracy of bits consumption for each DCT subblock is obtained by utilizing the category indexes. Simulation results show that the proposed algorithm has enough coding efficiency to transmit videoconferencing motion pictures through a 384 kbit/s channel. Copyright © 1987 by The Institute of Electrical and Electronics Engineers, Inc.

New two‐dimensional nonrecursive digital filters are proposed for noisy blurred images. A two‐dimensional nonrecursive filter is developed to minimize the mean‐squared error of the restored image. Here, the point spread function of the blurring system and statistics of the original image and noise are assumed to be given. A simplified version of the above filter is also developed and tested for restoring images blurred by a linear blurring system. As a result of their nonrecursive nature the proposed filters are stable and are determined by solving linear systems of equations. The blur by the point spread function and that due to degradation by noise are considered in the design of the filters. Images can be restored while the accompanying noise is suppressed. The filters herein are shown to be more effective if combined with noise‐removal filters. Illustrative examples are included to demonstrate the excellent quality of images restored by the proposed filters. Copyright © 1985 Wiley Periodicals, Inc., A Wiley Company