基于信源符号重排的抗丢包图像不等保护方法

Unequal Loss Protection of Image Based on Rearranging the Source Symbols

对内嵌图像数据采用交织器结构的不等保护,以抗击包丢失现象,是广泛采用的有效方案。本文提出在同一信道码率游程内,把信源符号的排列方式从原方案的逐行排列改成逐列排列能够增加接收端用于解码的信源符号数目,改善重建图像的质量。基于以上改进,导出了相应的期望PSNR计算式。在考虑重排效果和解码图像最小质量要求的前提下,提出了一种基于局部搜索的信源/信道码率分配新算法。采用双状态马尔可夫网络模型,通过对SPIHT和JPEG2000编码器产生的码流数据进行仿真实验。结果表明,在同样的传输失败概率下,新算法的期望PSNR值显著优于同类算法,特别在包数较少而包长度较大的情况下,期望PSNR性能的改善幅度更大。

It is an efficient system to protect the embedded image data unequally using the interleaver-based structure in order to combat the packet loss in packet erasure channel. In this paper, it is shown that the total number of the decoded source symbols can be increased by rearranging the source symbols within the runs of the same channel code rate in a vertical direction instead of the original horizontal direction, which results in the performance improvement to the original system. The relevant formulas are developed to describe the expected peak signal-to-noise ratio （PSNR） at the receiver based on the rearrangement. By considering the rearrange effect and the minimum quality requirement, a new algorithm based on local search is presented to allocate the transmission bit budget between the source and channel codes. Experiments are carried out with the source coders of SPIHT and JPEG2000 and with the network model of two-state Markov. The results show that the new algorithm can provide significantly higher expected PSNR especially in the cases of small number of packets and large packet length compared to the previous algorithm under the same probability of failure, i. e. , the probability that the quality of the decoded image is below the minimum quality require ment.