Paper
11 October 2023 Multi-angle SAR image registration based on Demons
Maixue Chen, Dong Feng, Liang Shen, Xiaotao Huang
Author Affiliations +
Proceedings Volume 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023); 1280056 (2023) https://doi.org/10.1117/12.3004015
Event: 6th International Conference on Computer Information Science and Application Technology (CISAT 2023), 2023, Hangzhou, China
Abstract
The inconsistent geometric distortion of Synthetic Aperture Radar (SAR) images when detecting from different angles makes multi-angle SAR image registration difficult. In this paper, the Demons algorithm, a non-rigid grey-scale based registration algorithm, is used for the first time in multi-angle SAR image registration. The algorithm calculates the displacement offset of the image to be aligned which influenced by the difference between the grey-scale of the reference image and the image to be aligned as well as the gradient of the reference image. Then Demons iteratively optimizes the offset through smooth interpolation with a Gaussian filter to ensure that image is accurately aligned. Three sets of SAR multi-angle images of different mountainous areas are used for experimental validation, and the analysis is carried out from both qualitative and quantitative metrics. The results show the Demons algorithm is more effective and more accurate than other non-rigid registration algorithms.
(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.
Maixue Chen, Dong Feng, Liang Shen, and Xiaotao Huang "Multi-angle SAR image registration based on Demons", Proc. SPIE 12800, Sixth International Conference on Computer Information Science and Application Technology (CISAT 2023), 1280056 (11 October 2023); https://doi.org/10.1117/12.3004015
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Image registration

Synthetic aperture radar

Deformation

Distortion

Image processing

Rigid registration

Image restoration

Back to Top