基于上下文的自适应二进制算术编码(Context-Adaptive Binary Coding,CABAC)是一种高效的熵编码方法,但是其高计算复杂度制约了该算法的编码速度,已经成为其应用的一个主要瓶颈。为了解决该问题,在分析CABAC算法及其计算复杂度的基础上,对其重归一化部分进行了改进,提出了一种提高其编码速度的有效方法。该算法根据重归一化次数细分编码下限的范围,并且根据重归一化次数出现的大小,选择最优的重归一化操作,除去了重归一化的循环过程,提高了编码速度。实验数据表明,该算法较以往的算法使CABAC的编码速度平均提高了18.7%~19.4%,并且编码效率没有下降[2]。%In the analysis of CABAC algorithm and computational complexity on the basis of it,the renormalization is im-proved in this paper,and an effective method is presented to improve the encoding speed.The algorithm based on the num-ber of renormalization divides the range of coding lower limit,and according to the size of renormalization number,it selects the optimal renormalization operation.So that it removes the cycle process of renormalization and improves the coding speed.
展开▼