IMAGE BLOCK CLASSIFICATION AND VARIABLE BLOCK SIZE SEGMENTATION USINGA MODEL-FITTING CRITERION

A new variable block size segmentation for image compression is propos ed. The decision whether or not the given image block is homogeneous i s based on a model-fitting criterion. More specifically, calculating t he maximum log-likelihoods for all predetermined block patterns with r espect to the given image data, we apply a modified Akaike information criteria (AIC) to select a best match. Then we can classify a given i mage block into one of texture, monotone, and various edges according to the characteristics of the selected pattern. Having classified nono verlapping small square blocks, we can cluster homogeneous blocks to h ave a variable block size segmentation. Since the gray-level distribut ion in the block (i.e., the maximum log-likelihood) is considered in t he model-fitting criterion, the proposed algorithm can differentiate e dges from textures. Moreover, edge blocks can be further classified as having vertical, horizontal, or diagonal edges. Also, since the conte xtual information among neighboring blocks is considered to eliminate isolated blocks and to connect broken edges, we can have larger homoge neous blocks to guarantee a more efficient coding. (C) 1997 Society of Photo-Optical Instrumentation Engineers.