摘要
为了克服高效视频编码(High Efficiency Video Coding,HEVC)的R-λ模型速率控制算法现使用的比特分配策略缺乏对图像内容的复杂性的参考的问题,提高标准码率控制算法的码率控制精度和编码器率失真性能,提出一种基于新的R-λ模型的帧层码率控制算法。首先引入一种新的图像复杂度的表示方法,改进帧层比特分配,然后采用重新计算的目标比特参与率失真模型参数更新,以提高参数精度,改善编码器率失真性能。实验结果显示,在恒定比特率的情况下,与参考算法相比,该算法的实际输出码率更接近目标码率,较标准算法相比峰值信噪比平均提高0. 681 7 d B。在低延时的测试环境下,算法的BD-Rate值为-3. 2%,率失真性能较好;对于具有更复杂纹理的视频,优化效果尤为明显,并且视频的编码质量得到显著提高。
To overcome the lack of reference to the complexity of image content in the existing bit allocation strategy of high efficiency video coding(HEVC) based on R-λ model bit rate control algorithm,and also improve the rate control accuracy and encoder rate distortion performance,a new frame-level code rate control algorithm based on the R-λ model is proposed. Firstly,an image complexity measure to improve the frame-level bit allocation algorithm is introduced,and the re-calculated target bit rate is adopted to improve the parameter accuracy when the rate-distortion model parameters are updated. Experimental results show that under the condition of constant bit rate,compared with the reference algorithm,the actual output bitrate of the proposed algorithm is closer to the target bit rate;and compared with the standard algorithm,the peak signal-to-noise ratio(PSNR) is improved by 0. 681 7 dB on average. In the test environment with low delay,the BD-Rate of the algorithm is-3. 3%,which shows the good rate distortion performance. For video with more complex texture,the optimization effect is particularly obvious,and the coding quality of video is significantly improved.
作者
常丹燕
杨静
CHANG Danyan;YANG Jing(College of Information Engineering,Shanghai Maritime University,Shanghai 201306,China)
出处
《电讯技术》
北大核心
2020年第4期455-459,共5页
Telecommunication Engineering
基金
国家自然科学基金资助项目(61601283)。
