Address‐Predictive Vector Quantization of Images by Topology‐Preserving Codebook Ordering

Abstract

Address vector quantization is a recently proposed image coding technique based on vector quantization (VQ) which exploits the interblock dependence by jointly encoding groups of VQ addresses. In this paper, by further developing the idea of address encoding, a new image coding technique is proposed. It is based upon the design of an ordered VQ codebook such that co‐devectors which are similar to each other have close addresses. As a result, the performance in terms of bit‐rate is very close to that of Address VQ, but much lower computational complexity and memory storage are required.

Codebook organization to enhance maximum a posteriori detection of progressive transmission of vector quantized images over noisy channels

“…Equation 12, the decision equation, is quadratic in and therefore can be expressed as: P 1 (1 0 ) 2 + P 2 (1 0 ) P 3 (1 0 ) + P 4 2 ; (14) where the terms P 1 ; P 2 ; P 3 , and P 4 , whose meaning are clear from Equation 12, can be calculated from the training set for each specic 3 2 3 binary block r. Then for each r, min (r) satises Equation 14 with equality. Thus, if the channel error rate is greater than min (r), the central bit of block r would be ipped, and if it is below min (r), it would not be changed.…”

“…In [16], we organized a full search VQ so that images coded with it could be sent progressively as in TSVQ. Related work on ordered VQ codebooks has been studied by Poggi and Cammorota to reduce the bit rate of standard and progressive transmission VQ [2,14], by Nasrabadi and Feng to lower the bit rate in nite-state VQ [10] and in our earlier work [15].…”

Applications of the kohonen algorithm in vector quantization

“…Based on this idea, address-predictive vector quantization (APVQ) [t3. 141 realizes low-rate encoding at low computation and memory costs. It makes use of an ordered codebook, generated by means of the KA, and exploits the dependence among spatially close addresses through a simple classified-linear prediction.…”

<title>Image compression based on genealogical relation of the TSVQ indices</title>

“…Two groups ofmethods have been proposed to achieve this goal. The first group is VQ schemes with memory: predictive, finite state, adaptive VQ The second group, which we call it index compressed VQ, is based on basic VQ, and a compression scheme to exploit the correlation of codewords (or indices) 6,7,8.9.10,11 The results of researchers show that the second group of schemes is able to provide better than the first one at bit rates less that 0.3 bits/pixel 6,7,8,9,10,11 Based on the method utilized in inter-block dependency removal, the index compressed VQ schemes can be divided into two groups. The first group, such as Address VQ 6,7and Index Compressed VQ 8,9, uses a lossless, and the second one uses a lossy IOu scheme.…”