基于梯度和全零块检测的快速帧内模式选择算法

Fast intra mode decision algorithm based on gradient and all-zero block detection

针对移动视频业务的市场需求和移动终端设备的性能局限,提出了一种基于方向梯度特征和全零块检测的H.264快速帧内预测模式选择算法。为降低预处理计算量,对像素进行降幂采样;进行主要预测方向上的纹理梯度检测,排除可能性较小的预测模式;对图像边界块和纹理平滑块进行优化处理;进行4×4全零块检测,略过不可能进一步降低率失真代价的候选模式。实验结果表明,在基本保持编码性能的前提下,该算法平均可降低51.69%的编码时间。

According to the market demand for mobile video services and the performance limited by the mobile terminal,a novel fast intra mode decision algorithm based on gradient and all-zero block detection is proposed. In order to reduce the calculation, it extracts the pixels according to the reduced power sampling. Then the texture gradient is calculated in the main prediction directions, thus excluding the prediction modes with larger gradients; and it simplifies the prediction modes for image boundary blocks and smooth texture blocks; finally, it carries out the 4×4 all-zero blocks detection, skips the candidate modes which can＇t reduce the RDO cost further. Experimental results show that, the proposed algorithm can save about 51.69% of encoding time, while in keeping with the R-D performance.