一种基于小波域的分形图像编码改进算法

提出了一种基于小波域的分形图像编码改进算法。该算法首先将图像分解到小波域,然后根据各子图像所包含能量的大小和所代表的方向等信息,采用不同大小、形状和类别的图像块及相似块,进行分形编码。在编码过程中按照零树结构在同方向不同分辨率的各个子带图像上确定要预测的图像块,同时在同方向低一级分辨率的子带图像上寻找与其最佳分形匹配的相似块,并由各级相似块构成预测树。实验证明,这种改进算法能够大大提高分形编码的速度,并取得较高的压缩比。

Fractal image coding improved algorithms based on wavelet domain

A new improved algorithm of fractal image coding based on wavelet domain is proposed in this paper.Through the algorithm,the image is decomposed into subimages by the wavelet transform.Then,according to the energy and orientation infor-mation,every subimage is encoded using fractal coding with range block and domain block of different size,shape and sort.During the fractal coding,we pick out the range blocks to be predicted based on zero-trees from subimages of the same orientation but different resolutions.For each range block,we look for its optimum fractal-matched domain block in the subimages of the same ori-entation but lower resolution.These domain blocks of all levels are used to construct the predictive trees.The experimental re-sults indicate that the improved algorithm of fractal image coding based on wavelet domain in the paper can enhance the speed of fractal coding greatly and achieve a higher compression rate.