CAJ | 학술논문

在分析小波变换与分形编码特点的基础上，提出了一种新的基于遗传分形编码的嵌入式小波图像编码算法。根据图像进行小波变换分解为不同空间频带上的子图像这一特性，对其低频子带进行基于遗传算法的分形编码，利用遗传算法的快速全局搜索能力提高分形编码中值域块与定义域块的匹配搜索速度；对其高频子带部分进行阈值去噪处理后进行SPIHT编码，两部分同时进行，提高图像的压缩效率。实验结果表明：同传统的分形编码和SPIHT编码压缩方法相比，该算法在图像的重构质量方面有所提高，尤其是在中低码率下PSNR(峰值信噪比)得到了较大提高，并且算法执行时间明显减少。
재분석소파변환여분형편마특점적기출상，제출료일충신적기우유전분형편마적감입식소파도상편마산법。근거도상진행소파변환분해위불동공간빈대상적자도상저일특성，대기저빈자대진행기우유전산법적분형편마，이용유전산법적쾌속전국수색능력제고분형편마중치역괴여정의역괴적필배수색속도；대기고빈자대부분진행역치거조처리후진행SPIHT편마，량부분동시진행，제고도상적압축효솔。실험결과표명：동전통적분형편마화SPIHT편마압축방법상비，해산법재도상적중구질량방면유소제고，우기시재중저마솔하PSNR(봉치신조비)득도료교대제고，병차산법집행시간명현감소。
On the basis of analyzing the characteristics of wavelet and fractal coding,propose a new embedded wavelet image coding algo-rithm based on genetic fractal coding. According to the characteristics of image wavelet transform which is decomposed into different spa-tial frequency sub-images,the low frequency sub-band is carried out fractal coding based on genetic algorithm,using fast global search a-bility of genetic algorithms to improve the matching search speed of range block and domain block in fractal coding. The high frequency sub-band is conducted SPIHT coding after threshold de-noising processing. Performed simultaneously in two parts,improve the compres-sion efficiency of the image. Experimental results show that compared with the traditional fractal coding and SPIHT coding compression methods,the algorithm gets a better reconstruction quality of image,especially at the low bit rate the PSNR is improved and the running time of the algorithm is also reduced.