Detecting Chemical and Biological Munitions Using Infrared Imaging Systems, Advanced Image Processing and Feature Extraction Methods, and Linguistic-Fuzzy Classifiers

This report is one in a series of reports describing the Disparate Sensor Integration (DSI) program. The purpose of the program is to determine if employment of CB weapons can be detected by existing (non-CB) sensors in a dual use mode. Specifically, this report describes how an existing Air Force IR imager can successfully discriminate between HE and CB artillery rounds during detonation. The report also describes how a newly developed data fusion technique, the Liguistic Fuzzy Logic, can be used to fuse together several discriminating features of the IR images to achieve this discrimination with a high degree of reliability. The reliability of the discrimination exceeds 90%. In addition, we are able to determine if the CB round detonated in the air or upon ground impact. This information is important for estimating the footprint of the chemical/biological plume. The original document contains color images. Prepared in collaboration with Geo-Centers, Inc., Newton Centre, MA.