A novel image segmentation algorithm derived in a fuzzy entropy framework is presented. First, the fuzzy entropy function is computed based on fuzzy region width and the Shannon's function of the image. Then all of the local entropy maxima are located in order to find the optimal partition for image segmentation scene local entropy maxima corresponding to the uncertainties among various regions in the image. This algorithm is very effective for the images whose histograms have no clear peaks and valleys, or the number of the segmentation classes is unknown, or the probabilistic model of the image and the different segmentation classes are unknown. A large number of experiments have been carried out on different kinds of images. Good performances of the proposed algorithm have been achieved.

A review and a performance comparison of several often-used vector quantization (VQ) codebook generation algorithms are presented. The codebook generation algorithms discussed include the Linde-Buzo-Gray (LBG) binary-splitting algorithm, the pairwise nearest-neighbor algorithm, the simulated annealing algorithm, and the fuzzy c-means clustering analysis algorithm. A new directed-search binary-splitting method which reduces the complexity of the LBG algorithm, is presented. Also, a new initial codebook selection method which can obtain a good initial codebook is presented. By using this initial codebook selection algorithm, the overall LBG codebook generation time can be reduced by a factor of 1.5-2.

Three fast search routines to be used in the encoding phase of vector quantization (VQ) image compression systems are presented. These routines, which are based on geometric considerations, provide the same results as an exhaustive (or full) search. Examples show that the proposed algorithms need only 3-20% of the number of mathematical operations required by a full search and fewer than 50% of the operations required by recently proposed alternatives.