基于交织抽取与分块压缩感知策略的图像多描述编码方法  

Image Multiple Description Coding Method Based on Interleaving Extraction and Block Compressive Sensing Strategy

作　　者：赵春晖[1] 刘巍[1] 

机构地区：[1]哈尔滨工程大学信息与通信工程学院,哈尔滨150001

出　　处：《电子与信息学报》2011年第2期461-465,共5页Journal of Electronics & Information Technology

基　　金：国家自然科学基金(61077079)资助课题

摘　　要：该文基于交织抽取和分块压缩感知(Interleaving Extraction and Block Compressive Sensing,IEBCS)理论,提出了一种可以在成像过程中实时实现的多描述编码方法(IEBCS-MDC)。首先利用交织抽取将图像划分成若干个子图像,然后对各个子图像进行分块压缩感知形成多个描述码流,接收端通过求解优化问题重建原图像。分块策略保证了观测过程的复杂程度不因图像尺寸而改变,所以该方法结构简单易于实现,适合处理高分辨率图像,另外特有的自恢复能力提升了算法的抗丢包性能。实验表明,在相同的硬件环境下,该文方法可以处理的图像尺寸远远大于CS-MDC方法,在同样的丢包率下重构质量也优于CS-MDC方法。Based on Interleaving Extraction and Block Compressive Sensing（IEBCS）,a new Multiple Description Coding method（IEBCS-MDC） which can be achieved real-timely during imaging process is presented.The method is first partitions an image into several sub-images using interleaving extraction,then measures each sub-image with block compressive sensing and forms multiple descriptions.At the decoding terminal,the method reconstructs the original image by solving an optimization problem.Block strategy ensures that the complexity of measurement process does not change due to image size,so the method is simple and easy to implement,suitable for handling high-resolution images,and the characteristic self-recovery capability enhances the ability against packet loss.Experimental results show that,compared to CS-MDC,the proposed method can handle much bigger images in the same hardware environment and the reconstruction quality is also better than CS-MDC with the same packet loss probability.

关 键 词：多描述编码 压缩感知 交织抽取 分块策略 自恢复 

分 类 号：TN911.2[电子电信—通信与信息系统]