Translator Disclaimer
20 April 2000 Remotely sensed pre- and post-disaster images for damage detection
Author Affiliations +
This study proposes a method to utilize the satellite, aerial and other remotely sensed pre- and post-disaster imagery data to perform geometric modeling and correlational analysis on the reconstructed models in order to detect the change associated, for example, with major regional and/or individual structural damage. Correlational analysis often fails to detect structural damage when only input images are utilized, especially if images are acquired under different illumination conditions. In fact, automatic detection in such cases becomes extremely challenging since making distinction of change due to structural damage from that associated with the difference in the illumination condition is extremely difficult. Many researchers have tackled this difficulty and proposed some methods of solutions including recursive hypothesis testing procedure. Although these methods provide a very useful basis for change detection, their applications are not universally successful for a variety of reasons. In order to achieve the required level of accuracy for the proposed application and locate the site of detected damage, it is proposed to use available GIS maps to register remotely sensed images. It is further necessary that a user-assisted three-dimensional model be reconstructed and correlational analysis performed. The algorithm performs successfully for change detection. However, issue of occlusion remains as a challenge that requires further investigation.
© (2000) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Masanobu Shinozuka and S. Ali Rejaie "Remotely sensed pre- and post-disaster images for damage detection", Proc. SPIE 3988, Smart Structures and Materials 2000: Smart Systems for Bridges, Structures, and Highways, (20 April 2000);

Back to Top