<title>Image compression using a novel edge-based coding algorithm</title>

Genetic-algorithm-based image compression technique using pattern classification

“…The GA used in this study, search for the optimal thresholds by coding the index j of j q , the quantized values on the histogram as a chromosome, where [ ] 14 , 0 ∈ j , with j = 7 corresponding to the location of the block mean on the histogram. Assuming a multi-modal histogram with three distinct peaks, there are two threshold values, and for example if the initial thresholds correspond to the quntized values of 3 q and 8 q , the chromosome describing the bit string is "00111000", representing the indexes (3,8). In our work, an initial population of 10 chromosomes by randomly selecting the j q was created.…”

“…GAs are computationally simple and yet powerful in their search for improvement Recently, GAs have been used in the field of image compression such as block truncation coding of images [4], vector quantization [5], fractal image compression [6] and video coding [7]. In this paper, a genetic algorithm technique is used in conjunction with the block pattern coding technique, which we presented earlier in [8]. In the block pattern image compression algorithm, an image is block coded according to the type of individual blocks, following a classification stage.…”