SAR图像小波域多尺度模型.pdfVIP

41 4 Vol.4 1 No .4 2009 8 Journal of N anjing U niversity of Aeronautics Astronautics Aug . 2009 SAR 袁小红　朱兆达 ( , , 21001 ) : 针对面向ATR 应用的SAR 图像压缩需要将图像压缩与图像自动分析相结合, 本文研究小波域SAR 图像 MAR 模型 应用于图像鉴别。与SAR 复图像分辨单元相干平均形成多尺度图像序列建立多尺度模型不同, 本 文对SAR 对数检测图像小波变换形成多尺度图像序列建立MAR 模型。通过实例辨识了SAR 图像中自然杂波 ( 草地) 与人造物( 战略目标) 模型, 应用该模型推导了小波域多分辨率判别, 对M STAR 图像鉴别实验验证了模 型的有效性。其中 2 小波域多分辨率判别鉴别性能好而计算复杂度低。小波域多分辨率判别可用在图像压缩 db 有损量化前鉴别出人造物与自然杂波。 : 合成孔径雷达; 多尺度; 小波 :T P75:A : 1005-2 15(2009) 04-0510-0 Multiscale Model of SAR Imagery in Wavelet Domain Yuan X iaohong , Zhu Zhaoda ( College of Information Science and Technology, Nanjing University of Aeronautics Astronautics, Nanjing, 21001 , China) : ( ) Abstract Multiscale model of synthetic aperture radar SAR imagery in w avelet domain is used to discriminate man-made objects from the natural clutter under consideration of automatic target recognition ( ATR) based SAR image compression. It is based on multiresolution sequence through - , w avelet transform on SAR log detected image different from those based on multiresolution sequence through coherent averaging on the complex SAR image. Both natural clutter ( grass) and man-made objects ( tactical targets) models are identified and a multiresolution discriminant in the w avelet domain . , ( ) is derived Tests on moving and stationary target acquisition and recognition MST AR datasets show that the multiresolut