基于方差的WMSN图像非均匀分块自适应压缩算法

Non-uniform Block Adaptive Compression Algorithm for WMSN Image Based on Variance

针对无线多媒体传感器网络数据量较大和储存能力有限等问题,提出了基于方差的图像非均匀分块自适应压缩算法。在传统的压缩方法中,均匀分块的稀疏变换对图像信号进行压缩,会带来图像重构的块状效应,而且无法有效地分离出重要区域或者背景区域。给出一种基于方差的图像非均匀分块自适应压缩算法,以图像方差值作为块大小分割依据,利用四叉树算法进行图像块大小的划分。并将分块大小结果分成快速变化块、过度块和缓慢变化块3类,适时做不同程度的压缩。实验结果表明:本文所提出的算法在图像重构质量方面优于传统的均匀分块压缩方法。

To solve the problems of large amount of data and limited storage capacity in wireless multimedia sensor network,an image non-uniform block adaptive compression algorithm is proposed based on variance.In the traditional compression method,the image signal is compressed by the sparse transform of uniform block,which will bring the block effect of image reconstruction,and the important area or background area cannot be separated effectively.An image non-uniform block adaptive compression algorithm is presented based on variance,and the image variance value is used as the basis for block size segmentation,and the quadtree algorithm is applied to divide the image block size.The results of block size is divided into three types:fast changing block,excessive block and slow changing block,and different degrees of compression are done at the right time.Experimental results show that the algorithm is superior to the traditional uniform block compression method in terms of image reconstruction quality.