Translator Disclaimer
31 December 2010 A multi-sensor image registration algorithm of window segmentation based on SIFT point and line features vector
Author Affiliations +
Proceedings Volume 7544, Sixth International Symposium on Precision Engineering Measurements and Instrumentation; 75443I (2010) https://doi.org/10.1117/12.885454
Event: Sixth International Symposium on Precision Engineering Measurements and Instrumentation, 2010, Hangzhou, China
Abstract
According to the deficiency of SIFT algorithm for matching multi-source remote sensing image feature points, this paper provides a multi-sensor image registration algorithm of window segmentation based on SIFT point and line feature vector. Firstly, we roughly matched the remote sensing images, and then we divided the remote sensing images into several separate temporary windows based on their characteristics. In the sub-window, we extracted SIFT feature point of remote sensing images and selected the eight directions points with the same name around the feature points to build line feature vectors. Finally, we ended up match point set by constraining the initial point of the SIFT algorithm. Tests show that this method achieved a high matching accuracy.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Yun Jiang and Jun Wang "A multi-sensor image registration algorithm of window segmentation based on SIFT point and line features vector", Proc. SPIE 7544, Sixth International Symposium on Precision Engineering Measurements and Instrumentation, 75443I (31 December 2010); https://doi.org/10.1117/12.885454
PROCEEDINGS
6 PAGES


SHARE
Advertisement
Advertisement
Back to Top