Oblique Aerial Image Matching Method Based on Dense SIFT and Improved Least Squares Matching Algorithm

doi:10.13474/j.cnki.11-2246.2018.0310

Abstract

Abstract: An automatic tie-points matching method of oblique image is proposed based on Dense SIFT and improved least squares matching to solve the problems of less matching points, uneven distribution and low matching efficiency. Firstly, the approximate orthophoto is obtained by the perspective transformation which the geometric deformation, scale and rotation between images is eliminated. Then pyramid image is built and split into blocks. Feature points are extracted using Dense SIFT algorithm from the top pyramid image block, and matching points are searched based on the KD tree, then the random sampling consistency algorithm is used to eliminate the mismatching points. An improved least squares matching algorithm is applied to correct the location of the initial matching point. Finally, the coordinate of the matching points are obtained by improved least square matching algorithm on all level pyramid images. Images of typical region is selected to test the matching effects, the experimental results show that the method can obtain dense and uniform matching points, and the efficiency is higher than SIFT and ASIFT algorithm.

Key words: oblique photogrammetry, perspective transformation, Dense SIFT, KD tree, least squares matching

CLC Number:

P237

YANG Xingbin, LÜ Jingguo, ZHANG Danlu, CHENG Zhe. Oblique Aerial Image Matching Method Based on Dense SIFT and Improved Least Squares Matching Algorithm[J]. 测绘通报, 2018, 0(10): 32-36,70.

References

[1] 袁修孝, 陈时雨. 倾斜航摄影像匹配方法探究[J]. 测绘地理信息, 2015, 40(6):1-6.
[2] MOREL J M, YU G. ASIFT:A New Framework for Fully Affine Invariant Image Comparison[J]. Siam Journal on Imaging Sciences, 2010, 2(2):438-469.
[3] MIKOLAJCZYK K, SCHMID C. Scale and Affine Invariant Interest Point Detectors[J]. International Journal of Computer Vision, 2004, 60(1):63-86.
[4] MIKOLAJCZYK K, SCHMID C. An Affine Invariant Interest Point Detector[C]//European Conference on Computer Vision.[S.l.]:Springer-Verlag, 2002:128-142.
[5] FORSSEN P E. Maximally Stable Colour Regions for Recognition and Matching[C]//IEEE Conference on Computer Vision and Pattern Recognition. Minneapolis:IEEE, 2007:1-8.
[6] 赵霞, 朱庆, 肖雄武,等. 基于同形变换的航空倾斜影像自动匹配方法[J]. 计算机应用, 2015, 35(6):1720-1725.
[7] 闫利, 费亮, 叶志云,等. 大范围倾斜多视影像连接点自动提取的区域网平差法[J]. 测绘学报, 2016, 45(3):310-317.
[8] HU H, Zhu Q, Du Z, et al. Reliable Spatial Relationship Constrained Feature Point Matching of Oblique Aerial Images[J]. Photogrammetric Engineering and Remote Sensing, 2015, 81(1):49-58.
[9] 肖雄武, 郭丙轩, 李德仁,等. 一种具有仿射不变性的倾斜影像快速匹配方法[J]. 测绘学报, 2015, 44(4):414-421.
[10] 肖雄武, 李德仁, 郭丙轩,等. 一种具有视点不变性的倾斜影像快速匹配方法[J]. 武汉大学学报(信息科学版), 2016, 41(9):1151-1159.
[11] 李德仁, 肖雄武, 郭丙轩,等. 倾斜影像自动空三及其在城市真三维模型重建中的应用[J]. 武汉大学学报(信息科学版), 2016, 41(6):711-721.
[12] VO T, TRAN D, MA W. Tensor Decomposition of Dense SIFT Descriptors in Object Recognition[C]//Proceedings of ESANN. Belgium:[s.n.], 2014.
[13] KAKDE H M. Range Searching Using Kd Tree[EB/OL].2005-08-25. http://www.cs.utah.edu/~lifeifei/cs6931/kdtree.pdf.
[14] 张祖勋, 张剑清. 数字摄影测量学[M]. 武汉:武汉大学出版社, 2012.
[15] LOWE D G. Distinctive Image Features from Scale-invariant Keypoints[J]. International Journal of Computer Vision, 2004, 60(2):91-110.
[16] 陶闯. 一种SPOT影像参数动态筛选的最小二乘匹配方法[J]. 武汉大学学报(信息科学版), 1993, 18(2):31-39.