基于仿射不变SIFT特征的SAR图像配准

刘向增; 田铮; 温金环; 武建明; 张朝阳

光电工程, 2010, 37 (11): 121, 网络出版: 2011-01-05

基于仿射不变SIFT特征的SAR图像配准

SAR Image Registration Based on Affine Invariant SIFT Features

论文大纲

刘向增 ^1,*田铮 ^1,2温金环 ¹武建明 ¹张朝阳 ¹

作者单位

¹ 西北工业大学应用数学系，西安 710129

² 中国科学院遥感应用研究所，北京 100101

SAR 图像配准相似矩阵 SIFT SIFT SAR image registration SVD SVD proximity matrix

摘要

针对SAR (Synthetic Aperture Radar) 图像全自动配准问题，本文提出一种基于仿射不变SIFT (Scale InvariantFeature Transform) 特征的精确配准方法。该方法首先对传统SIFT 方法改进构建具有仿射不变性的SIFT 描述子，并利用该描述子对提取的控制点进行粗匹配，然后由粗匹配点对的尺度比和方位差及其邻域的灰度相似性构建新的相似矩阵，最后利用SVD (Singular Value Decomposition)方法确定精确匹配点对，求出变换参数从而实现图像的精确配准。实验结果表明该方法优于传统的SIFT 方法和SIFT＋SVD 方法并且可以达到亚像素的配准精度。

Abstract

Referring to the problems of Synthetic Aperture Radar (SAR) image registration, an approach of SAR image registration based on affine invariant Scale Invariant Feature Transform (SIFT) features is proposed. First, the affine invariant SIFT descriptors are constructed by improving the original SIFT descriptors and the control points are rough matched by using the improved SIFT descriptors. Then, a new proximity matrix is built according to the scale ratio and the difference of orientations among the rough matched points and the similarity of the intensity around the rough matched points. Finally, the exact matching points are determined by using Singular Value Decomposition (SVD) method, and the precise image registration is performed. Experimental results show that the proposed algorithm has a better performance than original SIFT algorithm and SIFT+SVD method, and the accuracy of our algorithm can achieve sub-pixel level.

参考文献

[1] . A survey of image registration techniques[J]. ACM Computing Surveys (S0360-0300), 1992, 24(4): 326-376.

[2] . Image registration methods:A survey[J]. Image and Vision Computing (S0262-8856), 2003, 21(11): 977-1000.

[3] . Distinctive image features from scale-invariant keypoints[J]. International Journal of Computer Vision(S0920-5691), 2004, 60(2): 91-110.

[4] Ke Y，Sukthankar R. PCA-SIFT：A more Distinctive Representation for Local Image Descriptors [C]// Proceedings of Conference on Computer Vision and Pattern Recognition，Washington，USA，June 27-July 2，2004，2：II-506 - II-513.

[5] . A performance evaluation of local descriptors[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence (S0162-8828), 2005, 27(10): 1615-1630.

[6] . How to make local image features more efficient and distinctive[J]. IET Computer Vision(S1751-9632), 2008, 2(3): 178-189.

[7] . Robust scale-invariant feature matching for remote sensing image registration[J]. IEEE Geoscience and Remote Sensing Letters (S1545-598X), 2009, 6(2): 287-291.

[8] . SVD-matching using SIFT features[J]. Graphical Models (S1524-0703), 2006, 68(5): 415-431.

[9] . An algorithm for associating the features of two images[J]. Proceedings of Royal Society London Series B (S0962-8452), 1991, 244(1309): 21-26.

[10] . Feature-based correspondence:An eigenvector approach[J]. Image and Vision Computing(S0262-8856), 1992, 10(5): 283-288.

[11] Pliu M. A direct method for stereo correspondence based on singular value decomposition [C]// Proceedings of Conference on Computer Vision and Pattern Recognition，San Juan，Puerto Rico，June 17-19，1997：261-266.

[12] . Shape-adapted smoothing in estimation of 3-D shape cues from affine deformation of local 2-D brightness structure[J]. Image and Vision Computing (S0262-8856), 1997, 15(6): 415-434.

刘向增, 田铮, 温金环, 武建明, 张朝阳. 基于仿射不变SIFT特征的SAR图像配准[J]. 光电工程, 2010, 37(11): 121. LIU Xiang-zeng, TIAN Zheng, WEN Jin-huan, WU Jian-ming, ZHANG Zhao-yang. SAR Image Registration Based on Affine Invariant SIFT Features[J]. Opto-Electronic Engineering, 2010, 37(11): 121.

基于仿射不变SIFT特征的SAR图像配准

关于本站 Cookie 的使用提示

全站搜索

基于仿射不变SIFT特征的SAR图像配准

相关论文

相关资讯

关于本站 Cookie 的使用提示

全站搜索