一种HEVC样点自适应补偿改进方法

Improved Scheme for Sample Adaptive Offset in HEVC

新一代视频编码标准HEVC在环路滤波技术中采用样点自适应补偿算法解决变换量化引起的振铃效应,包括边带补偿和边缘补偿,但是边带补偿算法没有完全考虑边带分布情况,仅传递4个连续的边带。利用图像直方图充分分析了边带分布情况,白适应地提取了帧内预测模式纹理信息,针对不同大小的树形编码块提出了一种样点自适应补偿改进方法。实验结果表明,与HM10.0相比,该方法提高了编码性能,尤其在色度上比特率增益可达到-7.53%,图像边缘部分主观评价质量提升效果明显。

Sample adaptive offset（ SAO ）is adopted in the state-of-the-art video coding standard HEVC, including band offset （BO） and edge offset（EO） to attenuate ringing artifacts,which are more likely to appear after transform and quantization. Although BO spreads over 4 continuous bands, it is a fixed strategy for different blocks with the same size and does not consider a range narrower or wider than four bands. In this paper, band distribution of coding tree block（CTB） is analyzed by utilizing histogram, and adaptively bring a block＇s texture characteristics in intra prediction. Based on the analysis, an improved strategy is further developed for BO in SAO for HEVC. Experimental results indicate that significant performance and up to 7.53%, BD-rate reduction is achieved in chrome with marginal running time increase and the subjective visual quality is enhanced in the edge under common test conditions of HEVC reference software version 10. 0.