Dynamically Reconfigurable Embedded Image Processing System

Image processing uses many data processing techniques to transform the raw data or information from a sensor system into useful information from which decisions can be made. Historically, the data processing systems associated with each image processing application were tuned or optimized to that application such as machine inspection, pattern recognition, etc. In today's environment it is desirable to quickly evaluate promising techniques while maintaining minimal manpower and/or capital penalties. In this paper we will describe the implementation of a Missile Automatic Target Recognition (ATR) based on Adaptive Computing Systems (ACS) / Model Integrated Computing (MIC) techniques developed at ISIS/Vanderbilt University. 1