Detecting changes in aerial views of man-made structures

Abstract

An important application of machine vision is to provide a means to monitor a scene over a period of time and report significant changes. Comparing intensity values of successive images is not effective as such changes do not necessarily reflect actual changes at a site but might be caused by changes in the view point and illumination. We also want to ignore seasonal changes and focus on structural changes instead. We take the approach of comparing a 3-D model of the site, prepared from previous images, with new images to infer significant changes. This task is difficult as the images and the models have very different levels of abstract representations. Our approach consists of several steps as follows: model to image registration aims to bring the site model and the images into close correspondence; model validation to confirm the presence of model objects in the image; structural change detection seeks to resolve matching problems and provide cues indicating possibly changed structures; model updating aims to suggest modified models for existing model structures; and to incorporate new models. Our system is able to detect changes, such as missing (or mis-modeled) buildings, changes in model dimensions, and new buildings under some conditions. Several experimental results are presented.