摘要
Fractal image compression(FIC)technology is an interesting attempt at structure similarity-based image compression.It has been widely applied in many fields such as image encryption,image retrieval,image sharpening,and pattern recognition.However,overlong encoding time is the main difficulty for the application of FIC.In this paper,a new FIC speedup algorithm is proposed with two steps.Firstly,the simplified statistical variable expressions can speed up encoding twice more than the baseline fractal compression(BFC)without loss of image quality corresponding to BFC.Secondly,based on the fact that the affine self-similarity is equivalent to the absolute value of Pearson’s correlation coefficient,a new block classification strategy with flexible classification sets is proposed to speed up encoding further.The experiment results and theoretical analysis show that the proposed scheme achieves high performance in both image quality preservation and encoding efficiency.
Fractal image compression （FIC） technology is an interesting attempt at structure similarity-based image compression. It has been widely applied in many fields such as image encryption, image retrieval, image sharpening, and pattern recognition. However, overlong encod- ing time is the main difficulty for the application of FIC. In this paper, a new FIC speedup algorithm is proposed with two steps. Firstly, the simplified statistical variable expressions can speed up encoding twice more than the baseline fractal compression （BFC） without loss of image quality corresponding to BFC. Secondly, based on the fact that the affine self-similarity is equivalent to the absolute value of Pearson＇s correlation coefficient, a new block classification strategy with flexible classification sets is proposed to speed up encoding further. The experiment results and theoretical analysis show that the proposed scheme achieves high performance in both image quality preservation and encoding efficiency.
基金
supported by the National Basic Research Program of China(2010CB327902)
the National Natural Science Foundation of China(61231018)
