基于小波变换的多方位角SAR图像融合方法

石泉; 李景文; 杨威; 曾虹程; 张豪杰

doi:10.13700/j.bh.1001-5965.2016.0823

基于小波变换的多方位角SAR图像融合方法

doi: 10.13700/j.bh.1001-5965.2016.0823

1.
北京航空航天大学电子信息工程学院, 北京 100083
2.
北京电子工程总体研究所, 北京 100854

基金项目:

装备预研基金 A210301HK01043

详细信息

作者简介:
石泉男, 硕士研究生。主要研究方向:SAR信号仿真与处理

杨威男, 博士, 副教授。主要研究方向:SAR系统仿真与成像处理

通讯作者:
杨威, E-mail: yangweigigi@sina.com

中图分类号: TN975
计量
- 文章访问数: 682
- HTML全文浏览量: 76
- PDF下载量: 363
- 被引次数: 0
出版历程
- 收稿日期: 2016-10-26
- 录用日期: 2017-04-02
- 网络出版日期: 2017-10-20

Multi-aspect SAR image fusion method based on wavelet transform

1.
School of Electronic and Information Engineering, Beijing University of Aeronautics and Astronautics, Beijing 100083, China
2.
Beijing Institute of Electronic System Engineering, Beijing 100854, China

Funds:

Equipment Pre-Research Foundation A210301HK01043

More Information

Corresponding author: YANG Wei, E-mail: yangweigigi@sina.com

摘要

摘要:
针对目标散射特性空变效应以及多方位角图像信息冗余的问题，提出了一种基于小波变换和边缘检测的多方位角SAR图像融合方法。首先，将图像进行小波变换，完成低频信息和高频信息的分离，实现图像的多分辨表征及序贯图像多方位角信息的融合；然后，利用改进的Robinson算子增强图像的轮廓特征；最后，融合实验和定量评估结果证明了本文所提融合方法的有效性。
- 星载SAR /
- 多方位角图像 /
- 小波变换 /
- 边缘检测 /
- 图像融合
Abstract:
Considering the diversity of target scattering characteristics with different observation angle and information redundancy of multi-aspect images, this paper proposes a novel multi-aspect SAR image fusion method based on wavelet transform and edge detection. First, wavelet transform was performed to multi-aspect space-borne SAR image processing. The images were separated to different parts of frequency so that the multi-resolution representation and multi-aspect information of sequential images can be conveniently fused. Second, the improved Robinson edge detection algorithm was used to strengthen the energy of contour feature. Finally, fusion experiment and quantitative evaluation method were used to verify the effectiveness of this fusion imaging method.
- space-borne SAR /
- multi-aspect images /
- wavelet transform /
- edge detection /
- image fusion

HTML全文

图 1 多方位角观测示意图

Figure 1. Schematic of multi-aspect observation

下载: 全尺寸图片幻灯片

图 2 待融合图像及图像差异程度

Figure 2. Image to be fused and degree of image discrepancy

下载: 全尺寸图片幻灯片

图 3 提取出的差异区域

Figure 3. Extracted zone of difference

下载: 全尺寸图片幻灯片

图 4 灰度分布直方图

Figure 4. Gray level distribution histogram

下载: 全尺寸图片幻灯片

图 5 融合流程示意图

Figure 5. Flowchart of fusion method

下载: 全尺寸图片幻灯片

图 6 图像边缘检测流程图

Figure 6. Flowchart of image edge detection

下载: 全尺寸图片幻灯片

图 7 8个方向模板

Figure 7. Direction of 8 templates

下载: 全尺寸图片幻灯片

图 8 边缘检测及融合结果

Figure 8. Edge detection and fusion result

下载: 全尺寸图片幻灯片

图 9 融合结果

Figure 9. Fusion results

下载: 全尺寸图片幻灯片

表 1 8个方向模板对应的矩阵

Table 1. Corresponding matrixes of direction of 8 templates

方向模板	0°	45°	90°	135°	180°	225°	270°	315°
对应矩阵

下载: 导出CSV

表 2 锐化处理规则

Table 2. Sharpening processing rule

方向模板	锐化矩阵	锐化目标
0°		垂直方向
45°		对角方向
90°		水平方向
135°		对角方向
180°		垂直方向
225°		对角方向
270°		水平方向
315°		对角方向

下载: 导出CSV

表 3 实验参数

Table 3. Experiment parameters

参数	数值
分辨率/(m×m)	1×1
距离向入射角/(°)	40.69
方位向角度范围/(°)	±2.2
波段	X波段
成像模式	聚束模式
小波函数	Daubechies
小波阶数	1
融合图像数	6

下载: 导出CSV

表 4 评估结果

Table 4. Evaluation results

融合方法	信息熵	平均交叉熵	均方根交叉熵	信噪比/ dB
均值融合^[13]	6.672 3	0.126 5	0.141 9	9.264 2
加权均值融合^[13]	6.684 0	0.119 4	0.134 9	9.206 2
传统小波融合^[14]	6.604 2	0.391 0	0.428 7	8.544 0
本文方法	6.734 1	0.083 7	0.093 8	8.930 5

下载: 导出CSV

参考文献(27)

[1]	KIM J H, HEER C, SCHAEFER C.Astrium technology development for next generation SAR[C]//2013 IEEE International Asia-Pacific Conference on Synthetic Aperture Radar(APSAR), 2013:24-26.
[2]	GANTERT S, KERN A, DVRING R, et al.The future of X-band SAR:Terra-SAR-X next generation and WorldSAR constellation[C]//2013 IEEE International Asia-Pacific Conference on Synthetic Aperture Radar(APSAR), 2013:20-23.
[3]	PAPSON S.The exploitation of multi-look synthetic aperture radar and inverse synthetic aperture radar images for non-cooperative target recognition[D].State College:The Pennsylvania State University, 2007:1-20.
[4]	RUNKLE P, NGUYEN L H, MCCLELLAN J H, et al.Multi-aspect target detection for SAR imagery using hidden Markov models[J].IEEE Transactions on Geoscience & Remote Sensing, 2001, 39(1):46-55. http://adsabs.harvard.edu/abs/2001ITGRS..39...46R
[5]	张豪杰, 陈杰, 杨威, 等.基于FDTD的高保真SAR回波信号仿真方法[J].系统工程与电子技术, 2016, 38(1):45-52. doi: 10.3969/j.issn.1001-506X.2016.01.08 ZHANG H J, CHEN J, YANG W, et al.High-precision simulation of SAR echo using FDTD method[J].Systems Engineering and Electronics, 2016, 38(1):45-52(in Chinese). doi: 10.3969/j.issn.1001-506X.2016.01.08
[6]	WANG K, CHEN J, YANG W, et al.High accuracy SAR echo generation approach using space-time-variant backscattering characteristics[C]//IEEE International Geoscience and Remote Sensing Symposim.Piscataway, NJ:IEEE Press, 2014, 30:656-659.
[7]	DAVIDSON G W, CUMMING I G, ITO M R.A chirp scaling approach for processing squint mode SAR data[J].IEEE Transactions on Aerospace & Electronics Systems, 1996, 32(1):121-133. http://adsabs.harvard.edu/abs/1996ITAES..32..121D
[8]	李悦丽, 梁甸农, 黎向阳.一种改进方位向非线性CS大斜视角SAR成像算法[J].国防科技大学学报, 2008, 30(5):62-67. http://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ200805012.htm LI Y L, LIANG D N, LI X Y.A refined azimuth nonlinear chirp scaling algorithm for high squint SAR imaging[J].Journal of National University of Defense Techology, 2008, 30(5):62-67(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-GFKJ200805012.htm
[9]	DAILY M, FARR T, ELACHI C.Geologic interpretation from composited radar and Landsat imagery[J].Photogrammetric Engineering and Remote Sensing, 1979, 45(8):1109-1116.
[10]	宦若虹, 杨汝良.多方位角图像决策融合的SAR目标识别[J].遥感学报, 2010, 14(2):252-261. http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015 HUAN R H, YANG R L.SAR target recognition using multiple views decision fusion[J].Journal of Remote Sensing, 2010, 14(2):252-261(in Chinese). http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015
[11]	BROWN M Z.Analysis of multiple-view Bayesian classification for S-SAR ATR[J].Proceedings of SPIE-The International Society for Optical Engineering, 2003, 5095:265-274. http://adsabs.harvard.edu/abs/2003SPIE.5095..265B
[12]	SNYDER W C.Model-based fusion of multilook SAR for ATR[J].Proceedings of SPIE-The International Society for Optical Engineering, 2002, 4727:277-289. http://adsabs.harvard.edu/abs/2002SPIE.4727..277E
[13]	郭雷, 李晖晖, 鲍永生.图像融合[M].北京:电子工业出版社, 2008. GUO L, LI H H, BAO Y S.Image fusion[M].Beijing:Publishing House of Electronics Industry, 2008(in Chinese).
[14]	李卫斌, 张顺利, 解争龙, 等.基于小波分析与方向模板的SAR图像融合技术研究[J].计算机应用研究, 2007, 24(3):280-282. http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015 LI W B, ZHANG S L, XIE Z L, et al.Application of SAR image fusion based on wavelet and directional template[J].Application Research of Computers, 2007, 24(3):280-282(in Chinese). http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015
[15]	MALLAT S G.Multiresolution approximations and wavelet orthonormal bases of L²(R)[J].Transactions of the American Mathematical Society, 1989, 315(1):69-87.
[16]	韦春桃, 吕健刚, 杨先武, 等.一种利用方向模板的遥感影像道路提取方法[J].桂林理工大学学报, 2010, 30(3):378-383. http://www.cnki.com.cn/Article/CJFDTOTAL-GLGX201003013.htm WEI C T, LV J G, YANG X W, et al.Road extraction method in remote sensing image based on direction template[J].Journal of Guilin University of Technology, 2010, 30(3):378-383(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-GLGX201003013.htm
[17]	秦宣云, 卜英勇.多方向模板的构造及匹配的快速算法[J].计算机工程与设计, 2003, 24(6):91-92. http://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ200306026.htm QIN X Y, BU Y Y.Construction of multi-direction template and fast algorithm for template matching[J].Computer Engineering and Design, 2003, 24(6):91-92(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-SJSJ200306026.htm
[18]	张洁. 数字图像边缘检测技术的研究[D]. 合肥: 合肥工业大学, 2009. http://cdmd.cnki.com.cn/Article/CDMD-10359-2009155288.htm ZHANG J.Research on the techniques of digital image edge detection[D].Hefei:Hefei University of Technology, 2009(in Chinese). http://cdmd.cnki.com.cn/Article/CDMD-10359-2009155288.htm
[19]	DELEDALLE C A, DENIS L, TUPIN F.Iterative weighted maximum likelihood denoising with probabilistic patch-based weights[J].IEEE Transactions on Image Processing, 2009, 18(12):2661-2672. doi: 10.1109/TIP.2009.2029593
[20]	杨静, 田金文.整数Haar小波变换及其在无失真图像压缩中的应用[J].贵州大学学报(自然科学版), 1999, 16(3):161-167. http://www.cnki.com.cn/Article/CJFDTOTAL-GZDI199903000.htm YANG J, TIAN J W.Integer Haar wavelet transforms and lossless image compression[J].Journal of Guizhou University (National Science), 1999, 16(3):161-167(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-GZDI199903000.htm
[21]	GOUD P A, BINULAL G S.Simplified method of designing daubechies wavelets in class room[J].International Journal of Recent Trends in Engineering, 2009, 1(4):52-54.
[22]	COHEN A, DAUBECHIES I, FEAUVEAU J.Biorthogonal bases of compactly supported wavelets[J].Communications on Pure & Applied Mathematics, 1988, 41(7):909-996.
[23]	DAUBECHIES I, HEIL C.Ten lectures on wavelets[J].Computers in Physics, 1992, 6(6):697. http://adsabs.harvard.edu/abs/1993ASAJ...93.1671D
[24]	HERIC D, ZAZULA D.Combined edge detection using wavelet transform and signal registration[J].Image & Vision Computing, 2007, 25(5):652-662. http://www.sciencedirect.com/science/article/pii/S0262885606001557
[25]	贺国旗, 凌凤彩, 林晓.基于Canny思想的Robinson边缘检测算法[J].计算机工程与科学, 2016, 38(4):755-760. http://www.cnki.com.cn/Article/CJFDTOTAL-JSJK201604022.htm HE G Q, LING F C, LIN X.A Robinson edge detection algorithm based on Canny idea[J].Computer Engineering and Science, 2016, 38(4):755-760(in Chinese). http://www.cnki.com.cn/Article/CJFDTOTAL-JSJK201604022.htm
[26]	韩瑜, 蔡云泽, 曾清.图像融合的客观质量评估[J].指挥控制与仿真, 2013, 35(4):1-7. http://youxian.cnki.com.cn/yxdetail.aspx?filename=KZYC20170831001&dbname=CAPJ2015 HAN Y, CAI Y Z, ZENG Q.Objective image fusion performance assessment[J].Command Control & Simulation, 2013, 35(4):1-7(in Chinese). http://youxian.cnki.com.cn/yxdetail.aspx?filename=KZYC20170831001&dbname=CAPJ2015
[27]	周芳, 王鹏波, 李春升.遥感图像融合效果评估方法[J].现代雷达, 2013, 35(3):19-23. http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015 ZHOU F, WANG P B, LI C S.Evaluation method of remote sensing image fusion effect[J].Modern Radar, 2013, 35(3):19-23(in Chinese). http://youxian.cnki.com.cn/yxdetail.aspx?filename=BJHK20170622000&dbname=CAPJ2015