论文:2016,Vol:34,Issue(2):362-366
引用本文:
赵伟, 田铮, 杨丽娟, 延伟东, 温金环. 图像稳健配准的非负子空间匹配方法[J]. 西北工业大学学报
Zhao Wei, Tian Zheng, Yang Lijuan, Yan Weidong, Wen Jinhuan. Robust Registration of Remote Sensing Images Using Nonnegative Subspace Matching[J]. Northwestern polytechnical university

图像稳健配准的非负子空间匹配方法
赵伟1, 田铮1,2, 杨丽娟1, 延伟东1, 温金环1
1. 西北工业大学 应用数学系, 陕西 西安 710129;
2. 中国科学院 遥感科学国家重点实验室, 北京 100101
摘要:
针对局部场景发生变化的多时相遥感图像配准,提出一种基于非负子空间匹配的配准方法。在图匹配的框架下,该方法同时考虑了特征点集的空间结构和特征点集之间的相似关系,提高了正确匹配率和图像配准精度。与传统图匹配方法相比,该方法增强了对特征点位置扰动和异常值的稳健性。最后,通过在模拟点集匹配和一组多时相遥感图像配准上与传统图匹配方法的对比分析,验证了该方法的有效性以及应用于多时相遥感图像的可行性。
关键词:    图像配准    遥感    图匹配    位置扰动    异常值    稳健性   
Robust Registration of Remote Sensing Images Using Nonnegative Subspace Matching
Zhao Wei1, Tian Zheng1,2, Yang Lijuan1, Yan Weidong1, Wen Jinhuan1
1. Department of Applied Mathematics, Northwestern Polytechnical University, Xi'an 710129, China;
2. The State Key Laboratory of Remote Sensing Science, Chinese Academy of Sciences, Beijing 100101, China
Abstract:
For the registration of multitemperal remote sensing images, a Nonnegative Subspace (NS) matching method is proposed. The method is under the framework of graph matching and describes the structure of the feature point set as well as the across-sets feature similarity and obtains the nonnegative subspace coordinates of the feature points. The matching precision and image registration accuracy are enhanced by the method. The NS method is more robust to position jitter and outliers as compared with classical graph matching methods. Experiments on simulated point sets matching and a pair of multitemperal remote sensing images registration verify its effectiveness and feasibility.
Key words:    computer simulation    convergence of numerical methods    efficiency    experiments    factorization    graph theory    image registration    iterative methods    Lagrange multipliers    matrix algebra    mean square error    optimization    remote sensing    robustness (control systems)    graph matching    Nonnegative Subspace (NS) matching    outliers    position jitter   
收稿日期: 2015-10-19     修回日期:
DOI:
基金项目: 国家自然科学基金(60972150、61201323、61301196)、遥感科学国家重点实验室开放基金(OFSLRSS201206)资助
通讯作者:     Email:
作者简介: 赵伟(1987-),西北工业大学博士研究生,主要从事遥感图像配准与模式识别的研究。
相关功能
PDF(4722KB) Free
打印本文
把本文推荐给朋友
作者相关文章
赵伟  在本刊中的所有文章
田铮  在本刊中的所有文章
杨丽娟  在本刊中的所有文章
延伟东  在本刊中的所有文章
温金环  在本刊中的所有文章

参考文献:
[1] Zhao M, An B, Wu Y, et al. A Robust Delaunay Triangulation Matching for Multispectral/Multidate Remote Sensing Image Registration[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12(4):711-715
[2] Scott G L, Longuet-Higgins H C. An Algorithm for Associating the Features of Two Images[C]//Proceedings of the Royal Society of London B:Biological Sciences, 1991, 244(1309):21-26
[3] Wang H, Hancock E R. A Kernel View of Spectral Point Pattern Matching[C]//Proceedings of Structural, Syntactic, and Statistical Pattern Recognition, 2004:361-369
[4] Lowe D G. Distinctive Image Features from Scale-Invariant Keypoints[J]. International Journal of Computer Vision, 2004,60(2):91-110
[5] Lee D D, Seung H S. Algorithms for Non-Negative Matrix Factorization[C]//Advances in Neural Information Processing Systems, 2001:556-562
[6] Yang Z, Oja E. Linear and Nonlinear Projective Nonnegative Matrix Factorization[J]. IEEE Trans on Neural Networks, 2010,21(5):734-749
[7] Cho M, Lee J, Lee K M. Reweighted Random Walks for Graph Matching[C]//Proceedings of 11th European Conference on Computer Vision, 2010:492-505
[8] Leordeanu M, Hebert M. A Spectral Technique for Correspondence Problems Using Pairwise Constraints[C]//Proceedings of 10th IEEE International Conference on Computer Vision, 2005, 2:1482-1489
[9] Torki M, Elgammal A. One-Shot Multi-Set Non-Rigid Feature-Spatial Matching[C]//2010 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2010, 3058-3065