Paper
23 November 2011 A change detection algorithm for man-made objects based on remote sensing images
Wenwu Wang, Zhiguo Cao
Author Affiliations +
Proceedings Volume 8006, MIPPR 2011: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications; 800617 (2011) https://doi.org/10.1117/12.902078
Event: Seventh International Symposium on Multispectral Image Processing and Pattern Recognition (MIPPR2011), 2011, Guilin, China
Abstract
Radiometric difference, misregistration error and the determination of classification threshold for difference image seriously influenced the detection accuracy of traditional pixel-level change detection algorithms, and it is difficult to get the true changes of interest from various kinds of detected changes. Therefore, a novel change detection algorithm is proposed to detect changes of man-made objects in remote sensing images. Large-size images are divided into overlapping and multi-scale sub-images, and three kinds of multi-scale structural features (including interscale or intrascale features), such as central-shift moments, gradient-magnitude features, gradient-orientation features and line-length features are extracted by support vector machine (SVM) classification. Experimental results demonstrate the feasibility and effectiveness of the proposed algorithm.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Wenwu Wang and Zhiguo Cao "A change detection algorithm for man-made objects based on remote sensing images", Proc. SPIE 8006, MIPPR 2011: Remote Sensing Image Processing, Geographic Information Systems, and Other Applications, 800617 (23 November 2011); https://doi.org/10.1117/12.902078
Advertisement
Advertisement
RIGHTS & PERMISSIONS
Get copyright permission  Get copyright permission on Copyright Marketplace
KEYWORDS
Detection and tracking algorithms

Remote sensing

Image resolution

Image segmentation

Feature extraction

Image classification

Buildings

RELATED CONTENT


Back to Top