论文部分内容阅读
高效视频编码(HEVC)采用编码单元(CU)四叉树的分割结构,相比H.264/AVC显著地提升了编码效率,但却使编码复杂度急剧增加。为此,本文提出一种帧内快速编码算法。首先,根据视频图像纹理复杂度,提前判断是否进行最大编码单元(LCU)分割。然后,根据空域相邻CU的深度预测当前CU的深度范围,跳过不必要的计算;最后,根据预测模式被选为最优预测模式的统计特性,去掉可能性小的帧内预测模式。本文算法在HM14.0的基础上实现。仿真结果表明,本文算法在全I帧模式下与HM14.0相比,帧内编码时间平均减少38%,码率(BR)只增加1.41%,峰值信噪比(PSNR)只降低0.29dB,在保证编码性能和视频质量几乎不变的情况下,本文算法降低了编码的计算复杂度。
High-Efficiency Video Coding (HEVC) uses a coding unit (CU) quad-tree partition structure that significantly improves coding efficiency over H.264 / AVC but dramatically increases coding complexity. To this end, this paper presents a fast intra-frame coding algorithm. First of all, according to the complexity of the video image texture, it is determined in advance whether or not to perform a maximum coding unit (LCU) division. Then, the depth range of the current CU is predicted according to the depth of the adjacent CU in the spatial domain, and the unnecessary calculation is skipped. Finally, according to the statistic characteristic that the prediction mode is selected as the optimal prediction mode, the intra prediction mode with less possibility is removed. The algorithm in this paper is based on HM14.0. The simulation results show that the proposed algorithm reduces the intra-frame coding time by 38%, the bit rate (BR) only increases by 1.41% and the peak PSNR decreases by only 0.29dB compared with HM14.0 in full I frame. The algorithm in this paper reduces the computational complexity of coding while ensuring almost the same coding performance and video quality.