论文部分内容阅读
阐述了码率变换再量化过程中再量化误差的产生及其实质 ,给出了应用最小均方误差(MSE)准则减少再量化误差的原理。在码率变换再量化器一端 ,根据首次量化值在量化区间的概率分布中心决定该量化值的区间 ,有效地消除了量化区间跨区交叠所引入的再量化误差。给出了减少再量化误差方案的实现步骤。实验和模拟结果均表明 :该方案有效地减少了再量化误差 ,获得了较好的图像质量。
The generation and essence of re-quantization error in the process of re-quantization of bit-rate conversion are described. The principle of reducing the re-quantization error by using minimum mean square error (MSE) criterion is given. Based on the probability distribution center of the first quantization value in the quantization interval, the quantization range is determined at the end of the bit-rate conversion and re-quantizer, which effectively eliminates the re-quantization error introduced by the cross-region overlap in the quantization range. The realization steps of reducing re-quantization error are given. Experimental and simulation results show that this scheme can effectively reduce the re-quantization error and achieve better image quality.