22 October 2010 Research on IKONOS shadow extraction in urban region based on the principal component fusion information distort
Author Affiliations +
Abstract
Shadow exists obviously in high resolution remote sensing images. Automatic extracting shadow is quite important for removing shadow as noise or for mining shadow information. A new method of IKONOS shadow extraction in urban region was presented in this paper based on the principal component (PC) fusion information distort. First, the NIR (near infrared) band with more shadow information was selected for shadow extraction, and the information distort of PC fusion was assessed; it was found that shadow was sensitive to difference index. Second, a relative difference index was structured to enhance shadow information, as a result the values of relative difference index in shadow region were higher and the ones in non-shadow region were lower. Third, possible shadow was distinguished from non-shadow by threshold. Finally standard deviation was used to differentiate shadow from water for possible shadow, and the shadow was extracted. The results show that this shadow extraction method was simple with high accuracy, not only the shadow of high building but also that of low trees were all detected.
© (2010) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Cunjun Li, Cunjun Li, Jihua Wang, Jihua Wang, Qian Wang, Qian Wang, Wenjiang Huang, Wenjiang Huang, Xingang Xu, Xingang Xu, } "Research on IKONOS shadow extraction in urban region based on the principal component fusion information distort", Proc. SPIE 7824, Remote Sensing for Agriculture, Ecosystems, and Hydrology XII, 78241U (22 October 2010); doi: 10.1117/12.865156; https://doi.org/10.1117/12.865156
PROCEEDINGS
10 PAGES


SHARE
Back to Top