论文部分内容阅读
为提高分布式视频编码系统解码WZ帧的图像质量,提出了像素域基于广义高斯分布的WZ帧重构方案.该方案充分考虑了边信息与原始WZ帧间的相关性,以广义高斯分布作为虚拟相关信道模型,提高率失真性能;用广义高斯分布做相关噪声模型,对给定量化区间,计算边信息已知情况下WZ的条件期望作为WZ重构值;为了不过多地增加重构方案的复杂度,将广义高斯分布的形状参数固定为0.5,推导出重构WZ帧的闭式表示.实验结果表明,基于广义高斯分布的WZ帧重构方案能有效提高率失真性能和改善重构WZ帧的图像质量.
In order to improve the quality of WZ frames decoded by distributed video coding system, a WZ frame reconstruction scheme based on generalized Gaussian distribution in pixel domain is proposed, which fully considers the correlation between edge information and original WZ frames, and uses generalized Gaussian distribution as The correlation-based channel model is proposed to improve the rate-distortion performance. Using the generalized Gaussian distribution as the correlation noise model, the conditional expectation of WZ is computed as the WZ reconstruction value given the quantization interval and the known side information. In order not to increase the reconstruction scheme too much, , The shape parameter of generalized Gaussian distribution is fixed at 0.5, and the closed representation of the reconstructed WZ frame is deduced.The experimental results show that the WZ frame reconstruction scheme based on generalized Gaussian distribution can effectively improve the rate distortion performance and improve the reconstruction WZ frame image quality.