Object recognition based on fractal coding using domain blocks dictionary

Abstract

This article investigated the method of fractal compression for obtaining image descriptors adapted for use on mobile platforms, optimizing the performance and the amount of stored fractal images. The main idea is to apply the fractal compression method based on iterated function systems to reduce the dimension of the original images. The proposed method is more preferable than the existing methods of fractal compression by the number of computational operations and by the volume of stored data for compressed images. Also, a new concept of fractal coding using domain blocks dictionary is proposed in this article.