摘要
针对AVS-M帧间编码模式的复杂性,在对帧间预测编码最佳模式选择统计结果分析的基础上,提出了一种模式分类的帧间预测模式快速选择算法.该算法将全部编码模式进行分类,对SKIP模式采用提前终止判决,而对其它模式通过判别宏块的一致性和比较宏块的空时相关性逐类型比较.实验表明,与全模式选择算法相比,该算法的计算次数减小了52.5%~85.8%,在保持编码效率和重建图像质量的同时提高了编码速度.
To the complexity of AVS-M interframe coding mode,This paper poses a interframe prediction mode quick selection algorithm of mode classification on the basis of the analysis to interframe prediction coding best mode selection statistical results.The algorithm classifies all coding modes,uses advanced ending decision to SKIP mode,but contrasting other modes in different types by macroblock consistency and comparing the space-time relevance of macroblock.Experiments have shown that,comparing with full mode selection algorithm,this algorithm reduces computational times by 52.5%~85.8%,while maintaining the coding efficiency,rebuilding the image quality,and increasing the encoding speed.
出处
《小型微型计算机系统》
CSCD
北大核心
2011年第5期1021-1024,共4页
Journal of Chinese Computer Systems
基金
湖南省科技厅科技发展计划项目(2009FJ4062)资助
湖南省教育厅高等学校科学研究项目(08C364)资助