一种基于LBT的分布式图像压缩算法

摘要

In wireless multimedia sensor networks（WMSNs） ,The amount of image data gathered by camera nodes is very large. They need to be compressed and processing before transmission.But the memory,computation and energy limitation of individual sensor nodes. In order to overcome the low-power and high image quality requirements of WMSN,this paper studies the Lapped Biorthogonal Transform （LBT） theory and proposes a distributed image compression algorithm based on LBT in WMSNs. In this algorithm,based on the clustering architecture,multi-nodes sharing the compression task of image compression results show that this distributed algorithm can greatly reduce the hardware cost and the energy consumptions in WMSNs.The lifetime of the WMSNs can also be prolonged many times in the case of the sensor nodes deployed densely.%无线多媒体传感器网络（WMSNs）中传感器节点采集的数据量非常大，在传输前需对大数据量的多媒体信息进行压缩处理。但是单节点能源受限，存储、处理能力相对较弱。针对无线多媒体传感器网络应用的高效、低耗能的需求这些问题．在图像压缩双正交重叠变换（LBT）的基础上，文中提出了一种基于此变换的分布式无线多媒体传感器网络图像压缩算法。即基于簇结构，把压缩任务分配给其他节点，通过多个节点相互协作，共同完成图像的压缩编码和传榆。实验结果表明，在传感器节点散布不均且较为密集的情况下，该算法在高质量、低复杂度和低功耗等方面都有了很大的性能提高。