Fractal Image Encoding Based on Minimum Iterated Function System

A novel fractal encoding method based on minimum iterated function system was proposed in this study. First, we found that there existed a minimum domain block set (MDBS) which can complete the block-matching for all range blocks, and the definition of minimum iterated function system (MIFS) was introduced. The remaining domain blocks constitute the redundant domain block set (RDBS). Based on the characteristics of the range blocks coming from RDBS, the corresponding bock-matching operations can be optimized by removing the constraints on the scaling coefficients. Then, better block-matching effect can be achieved, and the decoded image quality can be improved as well. Finally, based on MIFS, the decoding process can be also accelerated effectively. Three fractal encoding methods were adopted to assess the performance of the proposed method. Experimental results show that compared the the previous methods, the proposed method can achieve better decoded image quality with slightly more computations and fewer computations in decoding process.