基于ROI的分形图像压缩编码

Fractal Image Compression Coding Based on ROI

分形图像压缩由于具有非常高的压缩比越来越受人们的关注,但是高压缩比的图像方块效应非常明显,这在很大程度上影响了解码图像的质量。为了解决这个问题,借鉴JPEG2000中提出的ROI概念,提出了将感兴趣区域(ROI)图像编码与分形图像压缩编码帽结合的图像编码方法,使得重构的图像中感兴趣区域的保真度高于背景区。该方法很好地解决了图像的压缩比和重构图像质量之间的矛盾。实验结果证明:此方法在获得较高压缩比的同时,压缩编码时间大为降低,解码图像质量也有较大的改善,且总体编码性能优于JPEG编码。

Fractal image compression is concerned more and more by people because of having high compression ratio, but blocking effects is very obvious under the condition of high compression ratio, which affects the quality of the decoded image to a large extent. To solve this problem, using for reference the concept of ROI brought forward in JPEG2000, a method was put forward that combined the image coding based on region of interest（ROI） with fractal image compression coding, which made the fidelity of ROI higher than that of background＇ s（BG）. The method solved preferably the conflict between image compression ratio and the quality of reconstructed image. The experiments prove that the method reduces the coding time greatly and that the quality of the decoded image is greatly improved, moreover, that the total coding performance is higher than JPEG coding, at the same time of high compression ratio.