4D CHANGE DETECTION

Le résumé fourni par la source

The purpose of this paper is to show how the utilization of Digital Surface Models (DSMs) in the change detection problem can mitigate analyst fatigue, enhance data fusion, perform cross modality change detection, and be used to model illumination & shadow effects. We register two DSMs and subtract them from each other to obtain residual height information (named a Q-DSM). Implicitly there is a relationship between the Q-DSM and the original DSM; hence we can exploit this relationship to reduce fatigue for the analyst and minimize the search space for changes. Moreover, if the original Digital Surface Models are from disparate modalities then we can implicitly map 3D residual changes to 2D modality disparate imagery (e.g., Correlated DSM from Synthetic Aperture Radar & Electro-Optical Imagery). Finally, we can model illumination differences in 3-Space and project those changes into 2-Space such that if an image was collected with varying aspect angles then we can ascertain if there was true change or change attributed to shadow and illumination effects. Therefore, shadow-masked 2D and 3D change detection can be realized. Our end product can be viewed in 3D Visualization tools such as Harris InReality, Google Earth, and NASA Worldwind.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.