24 October 2013 Extraction of urban impervious surface information based on object-oriented technology
Author Affiliations +
Abstract
Impervious surface is an important part of urban underlying surface, as well as an important monitoring index for city ecological system and environment changes. However, accurate impervious surface extraction is still a challenge. This paper uses the color, shape and overall heterogeneity features from the high spatial resolution remote sensing image to extract the impervious surface. An edge-based image segmentation algorithm is put forward to fuse heterogeneous objects which integrates edge features and multi-scale segmentation algorithm and uses the edge information to guide image objects generation. Results showed that this method can greatly improve the accuracy of image segmentation. Accuracy assessment indicated that the overall impervious surface classification accuracy and a Kappa coefficient yield 87% and 0.84, respectively.
© (2013) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Aixia Liu, Aixia Liu, Xiaojie Zhao, Xiaojie Zhao, Jing Wang, Jing Wang, Ting He, Ting He, } "Extraction of urban impervious surface information based on object-oriented technology", Proc. SPIE 8893, Earth Resources and Environmental Remote Sensing/GIS Applications IV, 88931K (24 October 2013); doi: 10.1117/12.2029055; https://doi.org/10.1117/12.2029055
PROCEEDINGS
6 PAGES


SHARE
Back to Top