ATR via coregistration of visual thermal range (VTR)-augmented radar images

摘要

Abstract: A method to automatically detect targets from sets of pixel- registered visual, thermal, and range images is outlined. It uses operations specifically designed to work on the different kinds of images to explore the information given by each of them. Five features are used to distinguish the targets from the clutter; texture, brightness, temperature, surface planarity, and height. We describe two different schemes. The first one involves logically combining the results of individual detectors with binary detection information as output. A morphological operation called `erosion of strength n' is introduced and utilized as a powerful tool for removal of spurious information. The second scheme utilizes the concept of `images of interest' to combine the information provided by the different images. A simple linear combination of these images yields excellent detection results. The success of this scheme supports its suitability for other ATR (Automatic Target Recognition) problems.!8