Matching scheme based on PIFS of compression for image retrieval

There exists the similarity among different images, while the degree of the similarity is different. For the images retrieval based on fractal compression, this paper purposes a novel matching scheme for image retrieval using partitioned iterated function (PIFS) codes. Because a compression code contains mapping information between similar regions in the same image, this mapping information can be treated as vectors. Also, it is important to generate the representative vectors using the mapping vectors. This representative vector can describes the features of the images, so the similarity between images is directly calculable from representative vectors. This similarity is applicable to image retrieval, and the scheme will be explained and demonstrated experimentally on its efficiency in this paper.