No abstract

No abstract

The JPEG statistical models in the lossy mode specify the procedures for converting the Discrete Cosign Transform (DCT) coeficients into binary strings and context modeling in a case where the binary arithmetic coder called QM-coder is employed as entropy coder. The JPEG lossy mode establishes two statistical models, one for prediction residuals of the D C coeficients and the other for the A C coeficients. I n this paper, we redesign these two models by taking account of their distribution. W e confirm that Laplacian distribution is appropriate f o r both the D C coeficients and the A C coeficients through Kolmogorov-Smirnov (KS) test; consequently, we propose the statistical models that fitLaplacian distribution. B y adopting the proposed statistical models in lieu of the conventional models, the number of the states decreases from 294 to 210 and the compression performance on several test images including super high definition images improves by 0.01 to 1.48%.

No abstract

Picture encoding methods based on the iterated function system (IFS) have been reported. However, they are designed for still pictures and few reports have been made for moving pictures. This paper proposes a coding method for moving pictures based on the recurrent iterated function system (RIFS). In RIFS, the function systems of contraction mappings in multiple spaces are composed. In this study, it is assumed that by using a sequence of moving pictures for multiple spaces, moving pictures can be coded. In this proposed method, several frames are treated as a group and the search in coding is adaptively switched, using the block (range cell) of each frame. The search is adaptively switched depending on whether the group of cells belongs to the still range or to the moving range. Using this technique, the degrees of spatial and time redundancy are reduced. This method is simulated to prove its validity. © 1999 Scripta Technica, Electron Comm Jpn Pt 1, 82(8): 32–41, 1999