Translator Disclaimer
2 December 2005 Automatic change detection based on least square image matching
Author Affiliations +
Proceedings Volume 6045, MIPPR 2005: Geospatial Information, Data Mining, and Applications; 604534 (2005)
Event: MIPPR 2005 SAR and Multispectral Image Processing, 2005, Wuhan, China
The key of automatic change detection between multi-temporal images is these images' matching accuracy should reach sub-pixel level. Nevertheless, due to the influence of various factors of geometrical distortion, remote sensing images after precise geometrical correction always still have pixel-level geometrical distortion. Therefore, ideal result of automatic change detection can't be achieved by these images directly. This paper proposes a new automatic change detection approach based on least square image matching. Using least square method to correct images' distortion, we get the Maximum correlation coefficient by iteration and obtain the optimal match point. By correlation coefficient, whether change or not can also be estimated. This approach resolves the multi-temporal image matching accuracy's influence to the change detection result, by which the accuracy of automatic change detection is guaranteed. The experiment results are provided, which indicate the effectiveness and the reliability of the approach.
© (2005) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Mi Wang, Baorong Zhang, and Xiaogang Ning "Automatic change detection based on least square image matching", Proc. SPIE 6045, MIPPR 2005: Geospatial Information, Data Mining, and Applications, 604534 (2 December 2005);

Back to Top