基于模式识别的质量可分级视频编码算法  

A novel quality scalable video coding algorithm based on pattern recognition

作　　者：艾新宇[1] 黎洪松[1] 

机构地区：[1]桂林电子科技大学信息与通信学院,广西桂林541004

出　　处：《桂林电子科技大学学报》2016年第1期19-22,共4页Journal of Guilin University of Electronic Technology

基　　金：国家自然科学基金(61261035);桂林电子科技大学研究生教育创新计划(GDYCSZ201451)

摘　　要：为了降低质量可分级视频编码算法的复杂度,提出一种基于模式识别的质量可分级视频编码算法。该算法将自组织神经网络用于可分级视频编码,利用较粗糙的特征模式库对图像编码生成基本质量层,通过精细的特征模式库对重建图像质量较差的部分区域编码生成质量增强层,从而实现质量可分级编码。仿真实验结果表明,该算法具有较好的质量可分级编码性能,在高压缩比情况下,其压缩性能优于传统的粗粒度质量可分级编码算法。In order to reduce the complexity of quality scalability, a novel quality scalable video coding algorithm based on pat- tern recognition is proposed. The self-organizing neural network is used for scalable video coding in the proposed scheme. A coarse pattern library is used for coding the base layer and two fine pattern libraries are used for recoding the area of the picture which has a bad reconstructed quality. Experimental results show that this algorithm has a better performance than the traditional coarse-grain quality scalable coding algorithm.

关 键 词：视频编码 模式识别 质量可分级 SOM 

分 类 号：TP37[自动化与计算机技术—计算机系统结构]