Abstract
The QM-coder [1] is an arithmetic code that is used in common in JPEG and JBIG as the binary information source coding. The most remarkable feature of arithmetic code is that dynamic adaptation can easily be realized. In other words, the code has high versatility. The same algorithm is effective for information sources with various statistical properties, and the code is expected to be applied to various kinds of information source coding [1–3]. However, the QM-coder uses a subtraction-type arithmetic code in order to simplify processing, with the augend operation executed only by constant substitution and subtraction. Thus, it is anticipated that the coding efficiency will be degraded. From such a viewpoint, this paper presents a proposal for the improvement of coding efficiency for subtraction-type arithmetic code [11, 12]. The proposed technique is to separate the augend into several states, and the constant to be assigned to the less probable symbol is varied according to the state. The probability estimation table in the proposed method is constructed by a theoretical investigation based on the table used in the QM-coder. As a result of the coding simulation for the binary image, it is seen that the proposed method can reduce the coding complexity by 0.686% to 1.301% compared to the QM-coder. © 1999 Scripta Technica, Electron Comm Jpn Pt 1, 83(1): 1–9, 2000
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have