非正方形四叉树变换系数能量分布模型及其扫描方式改进
发布时间：2019-06-12 03:44
【摘要】：传统的非正方形四叉树变换(NSQT)系数的扫描方式并不能很好地适应系数的能量分布,这将降低熵编码的编码效率。为了提高编码效率,该文提出非正方形四叉树变换系数能量分布模型,并改进了其系数扫描方式。首先推导了非正方形变换块的空间频率域位置分布,基于推导的分布,从理论上提出了其能量分布模型,并用实验进行了证实,发现能量从左上角到右下角逐渐递减,而左下角的能量与右上角的能量接近,根据该能量分布模型,重建了非正方形四叉树变换中量化系数的扫描模式,以适应熵编码,而熵编码中的系数扫描方式倾向于系数上的降序。实验结果显示,相比NSQT中已有的系数扫描方式,该文方法提出的扫描方式在编码计算复杂度和客观视频质量损失可忽略的前提下,平均下降了1.32%~3.74%的比特率,对于非正方形变换块,下降了9.05%~13.19%的比特率。
[Abstract]:The traditional scanning method of (NSQT) coefficients of non-square quadtree transformation can not adapt to the energy distribution of coefficients well, which will reduce the coding efficiency of entropy coding. In order to improve the coding efficiency, this paper proposes a non-square quadtree transform coefficient energy distribution model, and improves its coefficient scanning method. Firstly, the spatial frequency domain position distribution of the non-square transformation block is derived. based on the deduced distribution, the energy distribution model is proposed theoretically, and the experimental results show that the energy decreases gradually from the upper left corner to the lower right corner, while the energy in the lower left corner is close to the energy in the upper right corner. According to the energy distribution model, the scanning mode of the quantitative coefficient in the non-square quadtree transform is rebuilt to adapt to the entropy coding. The coefficient scanning mode in entropy coding tends to the descending order of coefficients. The experimental results show that compared with the existing coefficient scanning methods in NSQT, the proposed scanning method reduces the bit rate by 1.32% 鈮，
本文编号：2497706
[Abstract]:The traditional scanning method of (NSQT) coefficients of non-square quadtree transformation can not adapt to the energy distribution of coefficients well, which will reduce the coding efficiency of entropy coding. In order to improve the coding efficiency, this paper proposes a non-square quadtree transform coefficient energy distribution model, and improves its coefficient scanning method. Firstly, the spatial frequency domain position distribution of the non-square transformation block is derived. based on the deduced distribution, the energy distribution model is proposed theoretically, and the experimental results show that the energy decreases gradually from the upper left corner to the lower right corner, while the energy in the lower left corner is close to the energy in the upper right corner. According to the energy distribution model, the scanning mode of the quantitative coefficient in the non-square quadtree transform is rebuilt to adapt to the entropy coding. The coefficient scanning mode in entropy coding tends to the descending order of coefficients. The experimental results show that compared with the existing coefficient scanning methods in NSQT, the proposed scanning method reduces the bit rate by 1.32% 鈮，
本文编号：2497706
本文链接：https://www.wllwen.com/kejilunwen/wltx/2497706.html
