23 January 2017 Oblique low-altitude image matching using robust perspective invariant features
Author Affiliations +
Proceedings Volume 10322, Seventh International Conference on Electronics and Information Engineering; 103221S (2017) https://doi.org/10.1117/12.2265167
Event: Seventh International Conference on Electronics and Information Engineering, 2016, Nanjing, China
Abstract
Compared with vertical photogrammtry, oblique photogrammetry is radically different for images acquired from sensor with big yaw, pitch, and roll angles. Image matching is a vital step and core problem of oblique low-altitude photogrammetric process. Among the most popular oblique images matching methods are currently SIFT/ASIFT and many affine invariant feature-based approaches, which are mainly used in computer vision, while these methods are unsuitable for requiring evenly distributed corresponding points and high efficiency simultaneously in oblique photogrammetry. In this paper, we present an oblique low-altitude images matching approach using robust perspective invariant features. Firstly, the homography matrix is estimated by a few corresponding points obtained from top pyramid images matching in several projective simulation. Then images matching are implemented by sub-pixel Harris corners and descriptors after shape perspective transforming on the basis of homography matrix. Finally, the error or gross error matched points are excluded by epipolar geometry, RANSAC algorithm and back projection constraint. Experimental results show that the proposed approach can achieve more excellent performances in oblique low-altitude images matching than the common methods, including SIFT and SURF. And the proposed approach can significantly improve the computational efficiency compared with ASIFT and Affine-SURF.
© (2017) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Haiqing He, Haiqing He, Jing Du, Jing Du, Xiaoyong Chen, Xiaoyong Chen, Yuqian Wang, Yuqian Wang, } "Oblique low-altitude image matching using robust perspective invariant features", Proc. SPIE 10322, Seventh International Conference on Electronics and Information Engineering, 103221S (23 January 2017); doi: 10.1117/12.2265167; https://doi.org/10.1117/12.2265167
PROCEEDINGS
6 PAGES


SHARE
RELATED CONTENT


Back to Top