An Affine Invariant-based Match Propagation Method for Quasi-dense Image Registration
-
摘要: 待配准的两景影像之间存在非均匀形变时，需要采用非刚性模型进行校正，而密集的、均匀分布的控制点是构建非刚性模型的基础，对此提出一种仿射不变量支持的准稠密控制点匹配算法。首先从两景影像提取特征点并进行稀疏匹配，根据稀疏匹配点消除两景影像间的整体偏移；然后以稀疏匹配点作为三角形的顶点，迭代构建特征点间的近邻三角形，并以面积之比作为仿射不变量，据此判断三角形顶点之间是否满足匹配关系，实现匹配点传播，从而获得准稠密的匹配点集；最后采用HJ-Landsat数据与南极地区Landsat 7 ETM+数据进行实验，结果表明，本文方法能够提取均匀分布的准稠密控制点，在提取率等指标方面优于对比方法。Abstract: Non-rigid models are needed to correct the non-uniform deformation between two images; dense, uniformly distributed control points is the premise of constructing such non-rigid models. This paper puts forward an affine invariant-based match propagation method for quasi-dense image registration. Interest points are extracted from the input image and reference image, respectively; and sparsely matched. According to the sparsely matched points, the global offset between the two images is estimated and corrected. A corresponding adjacent triangle is established using the interest points in the two images interactively; and an affine invariant, the ratio of the triangle area, is employed to determine whether the vertices in the two triangles match or not. Through this method, the spare initial seed matching points propagate to other points of interest to obtain quasi-dense match. The HJ-Landsat image and ETM+ data of Antarctic data are taken in an experiment, the results show that our method can extract dense and uniformly distributed matching points while outperforming other state of art methods in true positive ratio.
-
-
![]()
图 3 仿射不变量支持的匹配点传播方法
Figure 3. Flow Chart of Match Propagation Based on an Affine Invariant
表 1 实验1影像的元数据
Table 1 The Metadata of Images Used in Experiment 1
影像 卫星传感器 轨道号 获取日期 HJ星 HJ1B-CCD 5/64 2012-08-29 TM影像A Landsat 5 TM 126-032 2010-08-31 TM影像B Landsat 5 TM 125-032 2010-07-05 TM影像C Landsat 5 TM 124-032 2010-08-15 
下载: 导出CSV
表 2 实验1的统计指标
Table 2 The Statistical Metrics of Experiment 1
编号 T/个 TR/% OR/% FR/% 本文方法 文献[13] 文献[10] 本文方法 文献[13] 文献[10] 本文方法 文献[13] 文献[10] 本文方法 文献[13] 文献[10] 1 6 243 5 997 1 026 90.34 87.53 83.32 8.59 12.20 - 9.66 12.47 16.68 2 3 357 2 686 2 202 92.17 85.63 82.14 9.54 27.63 - 7.83 14.37 17.86 3 5 215 4 068 3 336 93.72 90.62 67.21 10.50 30.19 - 6.28 9.38 32.79 4 3 819 3 055 2 505 89.17 80.16 79.65 10.58 28.47 - 10.83 19.84 20.35 5 4 915 3 932 3 224 88.63 79.15 75.62 7.99 26.39 - 11.37 20.85 24.38 6 2 785 2 228 1 827 85.13 80.25 73.62 13.59 30.87 - 14.87 19.75 26.38
下载: 导出CSV
-
[1] Zitova F J. Image Registration Methods:A Survey[J]. Image and Vision Computing, 2003, 21(11):977-1000 doi: 10.1016/S0262-8856(03)00137-9
[2] Mikolajczyk K, Schmid C. Scale & Affine Invariant Interest Point Detectors[J].International Journal of Computer Vision, 2004, 60(1):63-86 doi: 10.1023/B:VISI.0000027790.02288.f2
[3] Lowe D G. Distinctive Image Features from Scale-Invariant Keypoints[J].International Journal of Computer Vision, 2004, 60(2):91-110 http://cn.bing.com/academic/profile?id=0ad98e73d02a03f0751d0317dd8d714c&encoded=0&v=paper_preview&mkt=zh-cn
[4] Bay H, Ess A, Tuytelaars T, et al. Speeded-Up Robust Features (SURF)[J].Computer Vision and Image Understanding, 2008, 110(3):346-359 doi: 10.1016/j.cviu.2007.09.014
[5] 黄青青, 赵鸿志, 杨健, 等.环境星宽覆盖特征CCD影像几何定位及误差分析[J].武汉大学学报·信息科学版, 2014, 39(12):1425-1429 http://ch.whu.edu.cn/CN/abstract/abstract3135.shtml Huang Qingqing, Zhao Hongzhi, Yang Jian, et al. Geometric Rectification and Error Analysis for HJ-1 CCD Image with Wide Coverage Feature[J].Geomatics and Information Science of Wuhan University, 2014, 39(12):1425-1429 http://ch.whu.edu.cn/CN/abstract/abstract3135.shtml
[6] Heid T, Kääb A. Evaluation of Existing Image Matching Methods for Deriving Glacier Surface Displacements Globally from Optical Satellite Imagery[J].Remote Sensing of Environment, 2012, 118:339-355 doi: 10.1016/j.rse.2011.11.024
[7] Zagorchev L, Goshtasby A. A Comparative Study of Transformation Functions for Nonrigid Image Registration[J]. IEEE Transactions on Image Processing, 2006, 15(3):529-538 http://cn.bing.com/academic/profile?id=372ab2bc65a198e224bc72648c882ab4&encoded=0&v=paper_preview&mkt=zh-cn
[8] Lhuillier M, Quan L. Match Propagation for Image-Based Modeling and Rendering[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(8):1140-1146 http://cn.bing.com/academic/profile?id=e02749661071413b9453a1815354aa40&encoded=0&v=paper_preview&mkt=zh-cn
[9] Lhuillier M, Quan L. A Quasi-Dense Approach to Surface Reconstruction from Uncalibrated Images[J].IEEE Transactions on Pattern Analysis and Machine Intelligence, 2005, 27(3):418-433 http://cn.bing.com/academic/profile?id=2a8bc4311f29717f8d64444feb4fe7d5&encoded=0&v=paper_preview&mkt=zh-cn
[10] 唐丽, 吴成柯, 刘侍刚, 等.基于区域增长的立体像对稠密匹配算法[J].计算机学报, 2004, 27(7):936-943 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjxb200407010 Tang Li, Wu Chengke, Liu Shigang, et al. Image Dense Stereo Matching by Technique of Region Growing[J].Chinese Journal of Computers, 2004, 27(7):936-943 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjxb200407010
[11] 许振辉, 张峰, 孙凤梅, 等.基于邻域传递的鱼眼图像的准稠密匹配[J].自动化学报, 2009, 35(9):1159-1167 http://www.nlpr.ia.ac.cn/2009papers/gnkw/nk6.pdf Xu Zhenhui, Zhang Feng, Sun Fengmei, et al. Quasi-Dense Matching by Neighborhood Transfer for Fish-Eye Images[J].Acta Automatica Sinica, 2009, 35(9):1159-1167 http://www.nlpr.ia.ac.cn/2009papers/gnkw/nk6.pdf
[12] Zhu Q, Wu B, Tian Y. Propagation Strategies for Stereo Image Matching Based on the Dynamic Triangle Constraint[J]. ISPRS Journal of Photogrammetry and Remote Sensing, 2007, 62(4):295-308 doi: 10.1016/j.isprsjprs.2007.05.010
[13] 杨化超, 张书毕, 张秋昭.基于SIFT的宽基线立体影像最小二乘匹配方法[J].测绘学报, 2010, 39(2):537-543 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201002014 Yang Huachao, Zhang Shubi, Zhang Qiuzhao. Least Square Matching Methods for Wide Base-Line Stereo Images Based on SIFT Features[J].Acta Geodaetica et Cartographica Sinica, 2010, 39(2):537-543 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=chxb201002014
[14] 姚国标, 邓喀中, 艾海滨, 等.倾斜立体影像自动准稠密匹配与三维重建算法[J].武汉大学学报·信息科学版, 2014, 39(7):843-849 http://ch.whu.edu.cn/CN/abstract/abstract3031.shtml Yao Guobiao, Deng Kazhong, Ai Haibin et al. An Algorithm of Automatic Quasi-dense Matching and Three-dimensional Reconstruction for Oblique Ste-reo Image[J].Geomatics and Information Science of Wuhan University, 2014, 39(7):843-849 http://ch.whu.edu.cn/CN/abstract/abstract3031.shtml
[15] 吴军, 姚泽鑫, 程门门, 等.融合SIFT与SGM的倾斜航空影像密集匹配[J].遥感学报, 2015, 19(3):431-442 http://www.cnki.com.cn/Article/CJFDTotal-YGXB201503007.htm Wu Jun. Yao Zexin, Cheng Menmen, et al. Airborne Oblique Stereo Image Dense Matching by Integrating SIFT and SGM Algorithm[J].Journal of Remote Sensing, 2015, 19(3):431-442 http://www.cnki.com.cn/Article/CJFDTotal-YGXB201503007.htm
[16] 李全文, 赵卫林, 杨晓梅, 等.基于FAST测点的大幅宽HJ星图像几何精纠正方法[J].国土资源遥感, 2014, 26(4):14-22 doi: 10.6046/gtzyyg.2014.04.03 Li Quanwen, Zhao Weilin, Yang Xiaomei, et al. FAST Algorithm-Based Geometric Accurate Rectification of Large HJ Satellite Image[J].Remote Sensing for Land & Resources, 2014, 26(4):14-22 doi: 10.6046/gtzyyg.2014.04.03
计量
- 文章访问数: 1493
- HTML全文浏览量: 140
- PDF下载量: 218
