摘 要:針對高效率視頻編碼(High Efficiency Video Coding,HEVC)在編碼過程中的幀層比特分配未考慮視頻內容特性與緩沖區的狀態而導致幀層比特分配不合理及編碼性能低等問題,本文提出一種基于R-λ模型的幀層碼率控制優化算法。首先,通過計算得到一種度量幀層圖像復雜度的綜合因子,并作為新幀層比特分配權重;其次,根據目標緩沖級與緩沖區剩余比特計算出緩沖區反饋比特;最后,根據計算獲得的新幀層比特分配權重和緩沖區反饋比特對幀層進行比特分配。本文算法在HM16.0上進行了實現和性能測試,結果表明,與HEVC原始的碼率控制算法相比,優化算法的碼率控制誤差平均下降了2.008%,峰值信噪比平均提高了0.21dB。
關鍵詞:高效率視頻編碼;碼率控制;R-λ模型;綜合因子;緩沖區反饋比特
中圖分類號:TN919.81 文獻標識碼:A 文章編號:2096-4706(2018)12-0059-03
Frame Level Rate Control Optimization Algorithm Based on R-λ Model in
High Efficiency Video Coding
ZHENG Qiuju1,XU He2
(1.College of Mobile Telecommunications,Chongqing University of Posts and Telecommunications,Chongqing 401520,China;
2.School of Communication and Information Engineering,Chongqing University of Posts and Telecommunications,Chongqing 401520,China)
Abstract:Aiming at the problems of unreasonable frame layer bit allocation and low coding performance caused by the absence of video content characteristics and buffer state in frame layer bit allocation of high efficiency video coding (HEVC),this paper proposes a frame layer rate control optimization algorithm based on R-lambda model. Firstly,a comprehensive factor to measure the complexity of the frame layer image is calculated and used as the new frame layer bit allocation weight. Secondly,the buffer feedback bit is calculated according to the target buffer level and the residual bit of the buffer. Finally,the new frame layer bit allocation weight and the buffer feedback bit are calculated to advance the frame layer. Row bit allocation. The algorithm is implemented and tested on HM16.0. The results show that the rate control error of the optimized algorithm is reduced by 2.008% and the peak signal-to-noise ratio is increased by 0.21dB on average compared with the original rate control algorithm of HEVC.
Keywords:high efficiency video coding;rate control;R-lambda model;synthesis factor;buffer feedback bit
0 引 言
JCTVC在2013年4月頒布了新一代的視頻編碼標準—高效率視頻編碼(HEVC)[1]。與H.264/AVC[2]相比,在視頻編碼質量相同的情況下,HEVC的壓縮率提高了50%以上。HEVC的應用前景備受視頻消費者與視頻內容集成服務提供商的青睞。目前,傳輸帶寬和存儲空間仍然是視頻應用中的最關鍵資源,實現在有限的信道資源中獲得最佳的視頻體驗一直是專家、學者和視頻內容提供者的不懈追求。
視頻編碼標準大多采用預測、變換加熵編碼的混合編碼框架結構,由于編碼端的輸出碼率編碼方式不同,或由圖像運動劇烈程度和紋理復雜度的變化而導致存在著較大差異。為解決由碼流大……