一种基于提升小波的SAR图像有损压缩算法  

An Algorithm for SAR Imge Lossy Compression Based on Lifting Scheme

作　　者：唐红[1] 禹卫东[1] 徐锋[1] 谢东东[1] 

机构地区：[1]中国科学院电子学研究所,北京100080

出　　处：《电子与信息学报》2005年第5期779-783,共5页Journal of Electronics & Information Technology

摘　　要：该文根据合成孔径雷达(SAR)图像局部相关性较弱,存在相干斑噪声的特点,提出了一种基于提升小波的SAR图像有损压缩算法。它可有效去除相干斑噪声,提高计算精度和速度,改进SAR图像的压缩性能。该算法采用了嵌入式零树小波的编码方法,很好地利用小波系数的特性,使得输出的码流具有嵌入特性。利用大量的实际SAR图像对所提算法进行了仿真验证,并与其它图像压缩算法进行了对比分析,表明所提算法是有效的。In this paper a new algorithm for SAR image lossy compression based on lifting scheme is brought forward, aiming at special characteristics of SAR imagery. The algorithm can reduce the speckle noise, improve the computation precision and time, and upswing the compression ability. The algorithm adopts embedded zerotree wavelets, making good use of the wavelet coefficients .Large numbers of SAR images validate the efficiency of the proposed algorithm.The results compare favorably with that of other image compression methods.

关 键 词：SAR图像压缩 相干斑去噪 小波变换 提升方案 嵌入式零树小波 

分 类 号：TN957.52[电子电信—信号与信息处理]