| Fast Fractal Image Encoding Using an Improved Search Scheme |
| |
| 作者姓名： | 周一鸣 张超 张曾科 |
| |
| 作者单位： | Department of Automation Tsinghua University，Department of Automation，Tsinghua University，Department of Automation，Tsinghua University，Beijing 100084 China，Beijing 100084 China，Beijing 100084 China |
| |
| 基金项目： | 国家家重点实验室自然科学基金 |
| |
| 摘 要： | As fractal image encoding algorithms can yield high-resolution reconstructed images at very high compression ratio, and therefore, have a great potential for improving the efficiency of image storage and image transmission. However, the baseline fractal encoding algorithm requires a great deal of time to complete the best matching search between the range and domain blocks, which greatly limits practical applications of the algorithm. In order to solve this problem, a necessary condition of the best matching search based on an image feature is proposed in this paper. The proposed method can reduce the search space significantly and excludes the most inappropriate domain blocks for each range block before carrying out the best matching search. Experimental results show that the proposed algorithm can produce good quality reconstructed images and requires much less time than the baseline encoding algorithm. Specifically, the new algorithm can speed up encoding by about 85 times with a loss of just 3 dB in the peak signal to noise ratio (PSNR), and yields compression ratios close to 34.
|
| 关 键 词： | 图象编码 图像处理 分形图像 图像特征 |
| 收稿时间： | 15 March 2006 |
| 修稿时间： | 2006-03-15 |
Fast Fractal Image Encoding Using an Improved Search Scheme |
| |
| ZHOU Yiming,ZHANG Chao,ZHANG Zengke.Fast Fractal Image Encoding Using an Improved Search Scheme[J].Tsinghua Science and Technology,2007,12(5):602-606. |
| |
| Authors: | ZHOU Yiming ZHANG Chao ZHANG Zengke |
| |
| Institution: | Department of Automation, Tsinghua University, Beijing 100084, China |
| |
| Abstract: | As fractal image encoding algorithms can yield high-resolution reconstructed images at very high compression ratio, and therefore, have a great potential for improving the efficiency of image storage and image transmission. However, the baseline fractal encoding algorithm requires a great deal of time to complete the best matching search between the range and domain blocks, which greatly limits practical applications of the algorithm. In order to solve this problem, a necessary condition of the best matching search based on an image feature is proposed in this paper. The proposed method can reduce the search space significantly and excludes the most inappropriate domain blocks for each range block before carrying out the best matching search. Experimental results show that the proposed algorithm can produce good quality reconstructed images and requires much less time than the baseline encoding algorithm. Specifically, the new algorithm can speed up encoding by about 85 times with a loss of just 3 dB in the peak signal to noise ratio (PSNR), and yields compression ratios close to 34. |
| |
| Keywords: | fractal image encoding image feature matching search |
| 本文献已被 CNKI 维普 万方数据 ScienceDirect 等数据库收录！ |
