摘要：

针对分形图像编码时间过长的问题，提出一种基于结构信息特征的快速分形图像编码算法。给出结构信息特征的定义，并以该特征作为特征量对码本进行分类和最近邻搜索。在由搜索结果给出的邻域内进行子块匹配操作，实现编码。实验表明，在相同编码时间的情况下，该算法解码图像质量高于同类快速分形图像编码算法。

关键词: 图像压缩, 分形图像编码, 结构信息特征

Abstract:

To solve the problem of long encoding time, a fast fractal encoding algorithm based on structural information feature is proposed. The definition of structural information feature is given, and every block’s structural information feature is computed as the feature which is used for classification and nearest neighbor searching to arbitrary range block among the local domain pool. Then, the matching procedure takes place in the neighborhood of the searching result. Experiments show that in the case of the same encoding time, better decoded image quality can be achieved than similar algorithms.

Key words: image compression, fractal image encoding, structural information feature

中图分类号: 

• TP391.41