摘要： 分形图像压缩应用了图像的自相似性,目前许多对分形编码特性的研究都是在图像域中进行的。该文却试图探索在频率域中分形编码的特性。首先,综述了图像域中分形编码的特性,然后导出在离散余弦变换(简称DCT)域中仿射变换的伸缩因子和偏移量的对应公式。应用DCT的能量压缩特性,在度量图像子块和母块之间相似性中通过使用少量的低频DCT系数提出一种快速分形编码算法。接着进一步提出一种可能的快速混合分形编码的算法,该算法将快速搜索方法、统计规范化、频域比较结合起来。

关键词: 分形, 图像压缩, DCT, 快速分形编码

Abstract: Fractal image compression applies the self-similarity property of image. Many researches have been done to study the properties of fractal coding in image domain. In this paper, however, we try to explore the features of fractal coding in frequency domain. We firstly overview the properties of fractal coding in image domain, then we derive the corresponding formula of scaling factor and offset of affine transform in DCT domain. Applying the energy encoding algorithm by using only a small number of low frequency DCT coefficients in measuring the similarity between range block and domain block. We further propose a possible fast hybrid fractal encoding algorithm which combines existing fast search methods, statistical normalization and frequency domain comparison.

中图分类号: 

• TP391.41