论文部分内容阅读
针对分形图像压缩算法编码时间过长的问题,提出采用相关信息特征作为最近邻搜索特征的快速分形编码算法.通过深入分析图像子块的结构特性,提出相关信息特征的定义,证明并分析了采用该特征进行最近邻搜索操作的合理性.与传统特征相比,相关信息特征能够更好地反映子块的结构特性,所以基于相关信息特征的最近邻搜索能够更准确地确定后续局部匹配的范围.实验表明,在编码时间相同的情况下,本文算法较其他三种同类算法能够得到更好的解码图像质量.
Aiming at the problem of too long coding time of fractal image compression algorithm, a fast fractal coding algorithm using relevant information features as nearest neighbor search features is proposed. By analyzing the structural characteristics of image sub-blocks, the definition of relevant information features is proposed, This feature is reasonable for the nearest neighbor search operation.Compared with the traditional features, the related information features can better reflect the structural characteristics of the sub-blocks, so the nearest neighbor search based on the related information features can determine the scope of the subsequent local matching more accurately Experiments show that the proposed algorithm can get better decoded image quality than the other three kinds of similar algorithms under the same encoding time.