利用人眼视觉系统理论实现DCT域快速分形编码
A Fast Fractal Image Coding Method Using the Characteristics of Human Vision System in DCT Domain
-
摘要: 考虑到人眼视觉系统的特点,提出了基于视觉敏感度的DCT域中的均匀块选择标准,该标准优于常用的块方差与信噪比度量基于此标准,选择出DCT域中的均匀部分,直接将其直流部分编码输出,无需分形匹配,降低了DCT域分形图像压缩的复杂度为进一步缩短编码时间,在DCT域内引入人眼视觉系统频率响应对图像块与相似块进行分类,在过滤冗余的匹配计算的同时保证了解码图像的视觉效果实验结果表明,在不影响压缩比的前提下,编码时间大幅度减少,且PSNR有所提高,视觉效果较好.Abstract: Considering the characteristics of human visual system, this paper presents a plain block selecting standard in DCT domain based on visual acuity. This standard excels the frequently used block variance and SNR measurements. Based on the standard, the computational complexity of the fractal image compression is reduced by selecting out the uniform parts in the DCT domain and encoding output the direct current directly without any fractal matching. For encoding time further reducing, the frequency response of human visual system is introduced in the DCT domain to classify the picture blocks and similar blocks and the visual effects of the decoded images are guaranteed in the matching computation of redundancy filtering. Experiment shows that provided with the same compression ratio, the encoding time is greatly reduced with improved PSNR and much better visual effect.
下载: