1 HEVC的率失真优化过程 - 中国图象图形学报.DOC

中图法分类号：TN919 文献标识码：A 文章编号： ： 融合视觉感知特性的HEVC率失真优化 费马燕，彭宗举，李持航，陈芬，郁梅，蒋刚毅 宁波大学信息科学与工程学院宁波 315211 摘 要：目的 高效视频编码(High Efficiency Video Coding, HEVC)采用率失真优化技术选择最佳的编码参数，实现编码比特率和视频图像失真之间的平衡。失真度量通常采用均方误差和绝对误差和，这些方法并没有考虑人眼的主观感受。为了提高视频编码的主观感知质量，本文提出了一个融合视觉感知特性的率失真优化算法，并应用于帧间率失真优化过程中。方法 首先定义了一个视觉感知因子，该因子考虑了人类视觉系统对视频图像的空域活动性区域、纹理区域、时域运动性区域和亮度的感知特性，然后以编码树单元为单位对拉格朗日乘子进行自适应调整，最后根据拉格朗日乘子与量化参数之间的关系，对量化参数做进一步的修正。结果 实验结果表明与HEVC参考软件相比，提出的算法明显提高了率失真性能，对于相同的结构相似度(SSIM)分值，提出的算法在随机访问和低延时配置下平均分别节省3.1%，4.9%的码率，最高能节省9.0%的码率。。结论 本文提出的率失真优化策略能够根据图像不同的视觉特性自适应的调整率失真优化过程中的拉格朗日乘子，在保持编码质量基本不变的情况下，节省了码率，提高了HEVC的编码性能。 关键词：高清视频编码视觉感知；率失真优化；拉格朗日乘子 Human erception features based rate distortion optimization for HEVC Fei Mayan，Peng Zongju，Li Chihang，Chen Fen，Yu Mei，Jiang Gangyi Faculty of Information Science and EngineeringNingbo University，Ningbo 315211 China Abstract: Objective In the high efficiency video coding (HEVC), rate distortion optimization (RDO) techniques play an important role because it is usually used to select optimal encoding parameters and implement the trade-off between bit-rate and distortion of reconstructed video. In HEVC, the distortion is usually evaluated by mean squared error (MSE) or sum of absolute difference (SAD) since they are well-understood quantities and convenient. However, they are not clearly reflected the human visual perception because the video signal is ultimately receive by the human eye. So it is essential to develop a perceptual Lagrangian multiplier by considering the human visual system characteristics. Method In this paper, a novel RDO algorithm is proposed based on human perception and is applied for inter prediction video coding. Firstly, a human perception factor is defined by the features of the human vision system about spatial motion region, texture region, temporal motion activity and illumination in video. The multiplication of spatial motion region, texture region, temporal motion activity and illumination factors are used to evaluate the huma