
浏览全部资源
扫码关注微信
1. 中国科学院长春光学精密机械与物理研究所,吉林 长春,130033
2. 吉林省送变电工程公司, 吉林 长春 130033
3. 长春长光奥立红外技术有限公司, 吉林 长春 130031
4. 国网福建龙岩市新罗区供电有限公司, 福建 龙岩 364000
收稿日期:2013-12-20,
修回日期:2014-01-04,
网络出版日期:2014-01-07,
纸质出版日期:2014-04-25
移动端阅览
王丽, 魏巍, 吴林钢, 王灿进. SAR 图像目标识别新方法[J]. 液晶与显示, 2014,29(3): 429-434
WANG Li, WEI Wei, WU Lin-gang, WANG Can-jin. Novel target recognition method for SAR images[J]. , 2014,29(3): 429-434
王丽, 魏巍, 吴林钢, 王灿进. SAR 图像目标识别新方法[J]. 液晶与显示, 2014,29(3): 429-434 DOI: 10.3788/YJYXS20142903.0429.
WANG Li, WEI Wei, WU Lin-gang, WANG Can-jin. Novel target recognition method for SAR images[J]. , 2014,29(3): 429-434 DOI: 10.3788/YJYXS20142903.0429.
提出一种SAR图像目标识别新方法。首次引入BM3D方法,用于滤除原始图像中的相干斑噪声,BM3D结合了空间域和变换域去噪的优势,滤波性能优异。在特征提取步骤,将低阶Hu矩与高阶Zernike矩组合,Hu矩描述目标的粗略信息,高阶Zernike矩描述目标的细节信息,因此组合矩能够更加全面而细致地表达目标特性。使用组合矩特征训练SVM分类器,对含噪的SAR图像进行识别实验。实验结果表明:本文方法的识别率高达98.90%,优于已有的SAR目标识别方法。
A new target recognition method for SAR images is proposed. An advanced denoising method named BM3D is introduced for the speckle noise suppression at the first time. BM3D filter combines advantages in space-domain denoising with those in transform-domain denoising
which leads to better performance. For the feature vector extraction step
Hu moments
which describe the target's cursory information
and Zernike moments
which describe target's fine information
are fused to construct the combination moments. In this way
the combination moments can express the target more comprehensively and precisely. In our experiment the moments are extracted from SAR images and used to train SVM classifier
and the experiment results indicate that the proposed method can obtain a recognition rate of 98.90%
which outperforms other SAR recognition methods.
黄世奇,刘代志.侦测目标的SAR图像处理与应用 [M].北京:国防工业出版社,2009. Huang S Q, Liu D Z.SAR Image Processing and Application of Reconnaissance Measurement Target [M].Beijing: Defense Industry Press,2009. (in Chinese)[2] 薛笑荣. SAR图像处理技术研究[D].西安:西北工业大学,2004. Xue X R. Research on SAR image processing technology[D].Xian: Northwestern Polytechnical University,2004.(in Chinese)[3] 刘向增, 田铮, 史振广,等. 基于FKICA-SIFT特征的合成孔径图像多尺度配准[J]. 光学精密工程, 2011, 19(9): 2186-2196. Liu X Z, Tian Z, Shi Z G et al.SAR image multi-scale registration based on FKICA-SIFT features [J].Opt. Precision Eng.,2011,19(9): 2186-2196.(in Chinese)[4] 高珊,马艳会. 基于边缘保持的SAR图像滤波算法研究[J]. 液晶与显示, 2013,28(2):290-294. Gao S, Ma Y H. Algorithm research of filtering for sar images based on edges-preserving [J].Chinese Journal of Liquid Crystals and Displays,2013, 28(2): 290-294. (in Chinese)[5] Lee J S. Digital image enhancement and noise filtering by use of local statistics [J].IEEE Trans Pattern Analysis and Machine Intelligence, 1980, 2(2): 165-168.[6] Frost V S, Stiles J A, Shanmugan K S, et al.A mode for radar images and its application to adaptive digital filtering of multiplicative noise [J].IEEE Trans Pattern Analysis and Machine Intelligence,1982, 4(2): 157-165.[7] Kuan D T, Sawchuk A, Strand T et al.Adaptive restoration of images with speckle [J].IEEE Trans. Acoust., Speech, Signal Process,1987,35(3):373-383.[8] Lopes A, Nezry E, Touzi R et al.Maximum a posteriori speckle filtering and first order texture models in SAR images[C]//Proceedings of the 10th Annual International Symposium on Geoscience and Remote Sensing,New York, USA,1990:2409-2412.[9] Lee J S. Refined filtering of image noise using local statistics [J].Computer Graphic and Image Processing,1981,15(2):380-389.[10] Yu Y, Acton S T. Speckle reducing anisotropic diffusion [J].IEEE Transactions on Image Processing,2002, 11(11):1260-1270.[11] 孙海江, 王延杰, 陈小林. 基于自适应梯度阈值各向异性滤波抑制红外复杂背景[J]. 光学精密工程, 2014,22(1):146-151. Sun H J, Wang Y J, Chen X L.Suppression of infrared complex background based on adaptive gradient threshold anisotropic Filtering [J].Opt. Precision Eng.,2014, 22(1): 146-151.(in Chinese)[12] 于长淞, 方超. 基于小波变换的 ESPI 图像去噪及边缘提取[J]. 液晶与显示, 2011, 26(6): 818-822. Yu C S, Fang C.ESPI image denoising and edge extraction based on wavelet transform [J].Chinese Journal of Liquid Crystals and Displays,2011, 26(6): 818-822.(in Chinese)[13] Walessa M, Datcu M. Model-based despeckling and information extraction from SAR images [J].IEEE Transactions on Geoscience and Remote Sensing,2000, 38(6): 2258-2269.[14] 李家德, 张叶, 贾平. 采用非局部均值的超分辨率重构[J]. 光学 精密工程, 2013, 21(6): 1576-1585. Li J D, Zhang Y, Jia P.Super-resolution reconstruction using nonlocal means [J].Opt. Precision Eng.,2013, 21(6): 1576-1585.(in Chinese)[15] Dabov K, Foi A, Katkovnik V,et al.Image denoising by sparse 3D transform domain collaborative filtering [J].IEEE Transactions on Image Processing,2007, 16(8):2080-2095.[16] 付燕,詹新光. 一种有效的SAR图像目标识别方法[J]. 计算机工程与应用, 2010,46(15):156-158. Fu Y, Zhan X G.Efficient method of SAR image target recognition [J].Computer Engineering and Applications,2010, 46(15):156-158.(in Chinese)[17] 王义敏,秦永元,安锦文. 基于特征向量的SAR图像目标识别方法研究[J]. 计算机工程与应用, 2008,44(30):186-188. Wang Y M, Q Y Y, An J W. Target Recognition in SAR images based on feature vector [J].Computer Engineering and Applications,2008,44(30):186-188.(in Chinese)[18] Teague M R. Image analysis via the general theory of moment [J].Opt.Soc.Am,1980, 70(8):920-930.[19] 陶李, 王珏, 邹永宁, 等. 改进的Zernike矩工业CT图像边缘检测[J]. 中国光学, 2012, 5(1): 48-56. Tao L, Wang Y, Zou Y N,et al.Improved Zernike moment method for industrial CT image edge detection [J].Chinese Optics,2012, 5(1): 48-56.(in Chinese)[20] 赵建川, 王弟男, 陈长青, 等. 红外激光主动成像和识别[J]. 中国光学, 2013, 6(5): 795-802. Zhao J C, Wang D N, Chen C Q,et al.Infrared laser active imaging and recognition technology [J].Chinese Optics,2013,6(5): 795-802.(in Chinese)
0
浏览量
33
下载量
6
CSCD
关联资源
相关文章
相关作者
相关机构
京公网安备11010802024621