摘要
提出了一种基于视觉的码率分配策略.通过对人类视觉系统HVS的某些视觉特性进行分析,并对图像分段后获得的区域做出相应的视觉重要性估计,确定每个区域在主观意义上的视觉感兴趣值,将该值作为关键区域选择和码率调整的重要依据.实验结果表明,利用该方法所获得的图像恢复质量符合人眼的主观判断,而且算法复杂度较低,具有较高实用价值.
Presents a bit allocation scheme based on visual properties. Through the analysis of visual properties, corresponding visual importance estimates of segmented regions are made. As a result, the important regions of interest （ROI） will be ascertained and allocated in order to the quantified ROI values which reflect the importance of each region. Experimental results show that the reconstructed image quality of the video coder produced by the algorithm accord with human subjective conclusion well, and the algorithm have desirable application value and low computation cost.
出处
《北京理工大学学报》
EI
CAS
CSCD
北大核心
2009年第9期803-806,共4页
Transactions of Beijing Institute of Technology
关键词
视频编码
视觉特性
位分配
video coding
visual properties
bit allocation
