基于率失真理论的低复杂度码率控制算法

Low complexity rate control algorithm based on rate distortion theory

为降低计算复杂度和提高图像质量,提出了一种新的码率控制算法.基于率失真理论,推导出了基于平均绝对差(MAD)的码率(R-Q)模型.编码过程中根据图像的复杂度分配帧层的目标比特数,在宏块层利用宏块的复杂度和运动信息计算其权重,并使用拉格朗日方法最小化失真度,确定宏块的最优量化器.实验结果表明,与测试模型8(TMN8)算法相比,该算法不用计算方差,计算量得到明显降低.与测试模型7(TMN7)和TMN8的算法相比,该算法能将码率更精确地控制在目标码率附近,减少了跳帧,图像质量更加稳定且有所提高.

To decrease the computational complexity and improve the image quality, a novel rate control algorithm was proposed. Based on the rate distortion theory, rate-quantization （R-Q） model based on mean absolute difference （MAD） was derived. Target bits were allocated for each frame according to the image complexity during the encoding procedure. The distortion weight of each macroblock was calculated for macroblock layer based on the complexity and motion information. Distortion was minimized using Lagrange theory to find the optimized quantizer. Experiment results demonstrate that the proposed algorithm does not need to compute the variance and decreases computational complexity obviously compared with test model version 8 （TMN8） algorithm. The proposed algorithm achieves a bit rate closer to the target, reduces skipped frames, gets more stable and better image quality compared with test model version 7 （TMN7） and TMN8 algorithms.