Share Email Print
cover

Proceedings Paper

Oblique low-altitude image matching using robust perspective invariant features
Author(s): Haiqing He; Jing Du; Xiaoyong Chen; Yuqian Wang
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper 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.

Paper Details

Date Published: 23 January 2017
PDF: 6 pages
Proc. SPIE 10322, Seventh International Conference on Electronics and Information Engineering, 103221S (23 January 2017); doi: 10.1117/12.2265167
Show Author Affiliations
Haiqing He, East China Univ. of Technology (China)
NASG (China)
Jing Du, East China Univ. of Technology (China)
Xiaoyong Chen, East China Univ. of Technology (China)
Yuqian Wang, East China Univ. of Technology (China)
NASG (China)


Published in SPIE Proceedings Vol. 10322:
Seventh International Conference on Electronics and Information Engineering
Xiyuan Chen, Editor(s)

© SPIE. Terms of Use
Back to Top