基於偵幅複雜度畫面內編碼在H.264的位元率控制演算法

Abstract

為了讓視訊編碼後的位元率能維持在頻寬的限制之內，並且達到良好與穩定的畫面品質，位元率控制是相當重要的。然而目前大多數的研究都集中在畫面間編碼(inter frames)而不是容易造成緩衝區溢位的畫面內編碼(intra frames)。此外Intra-only這種編碼方式也已經納入H.264新的profile當中，它較傳統GOP編碼更適合應用在特別要求畫面品質的產品上。 在此論文中，我們提出一個於Intra-only編碼的位元率控制演算法。首先我們提出基於R-D-Q model最佳化的QP決定方式，藉由位元率與PSNR預測模型，我們利用R-D-Q model最佳化來找出可以平衡畫面品質與編碼效能的QP最佳解。另外，在Intra-only編碼上為了解決場景變換對位元率控制所產生的影響，我們所利用的R-D-Q model可以直接偵測場景變換並做適當QP選擇，避免緩衝區溢位。實驗結果顯示，提出的方法可以達到更好更穩定的畫面品質，且緩衝內含量也都維持在較低的水平。

Rate control serves as an important technique to constrain the bit rate of video transmission over a limited bandwidth and to control the bit allocations within a video sequence to maximize its overall visual quality. However, most of rate control researches focus on inter coding frames instead of intra coding frames which are more possible to cause buffer overflow problem. Besides, H.264 Intra-only compression scheme has been standardized as H.264 profiles which are more proper for professional applications than traditional GOP compression scheme. In this thesis, we propose an improved rate control scheme which is appropriate for Intra-only compression. First, we present a R-D-Q model based QP determination scheme for I-frames and P-frames . By the estimation models for rate and PSNR of I-frames, the best quantization parameters can be determined by R-D-Q model method. In order to deal with the specific intra frames caused by scene transitions, our R-D-Q model can directory detect scene change without extra detection to determine appropriate QPs for avoiding buffer overflow and saving bit budget. Simulation results show, that compared to other reference algorithms, our approach achieves better and stable quality with low buffer fullness.