Linear feature matching of satellite remote sensing images based on linear pairs
Abstract
In order to solve the problems such as incomplete line feature extraction and unstable gradient information in image edge region
in the process of line matching, this paper proposes a method of line feature matching for satellite remote sensing images based on line pairs. In
this algorithm, LSD method is used to extract the line features, and topological relationship is combined to construct the line pair features. Then,
combining affi ne transform constraint and distance similarity measure, the matching of the same name line pair is realized. Finally, we construct
some region descriptors of line offi ce respectively to realize the matching of the fi nal line with the same name. In this paper, two sets of Sentinel-2
and two sets of ZY-3 satellite remote sensing sub-images are selected for experiments. The experimental comparison results show that the proposed
method based on line pairs of satellite remote sensing images has certain eff ectiveness and robustness.
in the process of line matching, this paper proposes a method of line feature matching for satellite remote sensing images based on line pairs. In
this algorithm, LSD method is used to extract the line features, and topological relationship is combined to construct the line pair features. Then,
combining affi ne transform constraint and distance similarity measure, the matching of the same name line pair is realized. Finally, we construct
some region descriptors of line offi ce respectively to realize the matching of the fi nal line with the same name. In this paper, two sets of Sentinel-2
and two sets of ZY-3 satellite remote sensing sub-images are selected for experiments. The experimental comparison results show that the proposed
method based on line pairs of satellite remote sensing images has certain eff ectiveness and robustness.
Keywords
Straight line pairs; Satellite remote sensing image; Linear subarea descriptors
Full Text:
PDFReferences
[1] ZeqingLan, JingxueWang, LiqinWang. Multi-view line matching based on multi-view stereopsis and Leiden graph clustering [J]. Journal of
GeoInformation Science, 2019,26(07):1629-1645.
[2] HongruiZhao, JinchunQin, QifanTan. 3D reconstruction of Buildings based on linear feature optimization [J]. Journal of Building Science and
Engineering,2022,39(04):81-89.
[3] SuyanLiu, JingxueWang, ZhaoyuShen, et al. Linear matching algorithm combining geometric features of line pairs and single line descriptor constraints [J].
Journal of Wuhan University (Information Science Edition),2023,48(06):936-949.
[4] ALSHAHRI M,YILMAZ A.Line matching in wide-baseline stereo:a top-down approach[J].Image Processing IEEE Transactions on,2014,23(9):4199-
4210.
DOI: http://dx.doi.org/10.18686/modern-management-forum.v8i7.13582
Refbacks
- There are currently no refbacks.
