基于块特性的去块滤波后处理算法

A Post-processing De-blocking Algorithm Based on Characteristics of Blocks

在基于块的离散余弦变换编码的图像压缩技术中,低比特率时其重构图像的块边界上会产生严重的方块效应,降低了图像主观质量。提出了一种基于块特性的自适应去块效应算法。该算法以8×8块为单位对图像划分为平坦区域、边缘区域、纹理区域,并针对不同区域选用相应的滤波算法。仿真结果表明,该算法能够有效地去除图像中的块效应,并保留了大量的图像细节。

The block-based DCT compression technology often cause blocking artifacts near the block boundaries at low bit rates, seriously harming the image＇ s subjective quality. An adaptive post-processing de-blocking algorithm based on the characteristics of blocks is proposed. The images taking units of 8 ×8 block divide into smooth area, edge area and texture area, uses the most appropriate filtering algorithms in the different regions. Simulation results show that the proposed algorithm can effectively remove the blocking artifacts and preserve the edge and texture information.