基于小波变换的方块效应去除法

Method for Removal of Blocking Artifacts Based on Wavelet Transform

低比特率JPEG压缩图像会产生严重方块效应,即马赛克现象,后处理技术可以有效地减少该效应。针对压缩图像在小波域中有结构信息和块不连续性的不同表现,提出了一种基于小波变换的分析方法,对图像在小波域中的不同响应分别加以处理。该算法根据小波系数在不同范围内的不同特性对图像进行区域分类,进而对不同区域做出不同的处理。试验结果表明,该方法能有效地减少方块效应,提高解码端图像的视觉质量。

Low bit rate JPEG compressed images present serious blocking artifacts which can be reduced by post processing. In this paper, a method for removal of blocking artifacts based on wavelet transform is proposed. The analysis is based on the different responses of block boundaries （constrained by quantization）, and the structural information about an image in wavelet domain. Experimental results show that the proposed method can significantly reduce blocking artifacts and improve visual quality.