Fast Fractal Coding Algorithm Based on Twice Pre-searching

2011 Fourth International Conference on Information and Computing Pub Date : 2011-04-25 DOI:10.1109/ICIC.2011.64

Ming Zhao, Ling Bai, Hui Tang, Shaofa Zhou

引用次数: 1

Abstract

A fast fractal coding algorithm is proposed in this paper. This algorithm is based on Jacquin's full search algorithm. When matching range-domain blocks, by utilizing the variance of the image block and the relativity between the image blocks, this algorithm performs twice pre-searching, thus avoiding vast matching computation and reducing the coding time greatly. Experimental results show that the coding time can be reduced approximately by 80% while the PSNR of the most reconstructed images is increased by this algorithm.

查看原文本刊更多论文

基于二次预搜索的快速分形编码算法

提出了一种快速分形编码算法。该算法基于Jacquin的全搜索算法。在匹配距离域块时，该算法利用图像块的方差和图像块之间的相关性进行两次预搜索，避免了大量的匹配计算，大大缩短了编码时间。实验结果表明，该算法可将编码时间缩短约80%，同时提高重构图像的PSNR。

本文章由计算机程序翻译，如有差异，请以英文原文为准。

求助全文

约1分钟内获得全文求助全文

来源期刊

2011 Fourth International Conference on Information and Computing

自引率

0.00%

发文量