A fuzzy approach to automatic target recognition applied to bare and camouflaged synthetic aperture targets
This thesis presents a feasibility study on the implementation of fuzzy logic system as opposed to using only a neural network system for target recognition. The work will be based on the work done by Bernardon & Carrick , in which an artificial neural network is used as a means of target discrimination. The primary aim of the proposed work will be to improve upon percentage of targets recognized using the strip-map data while, at the same time, maintaining the accuracy of the results reported for the spotlight mode SAR operation data. While the primary application of this thesis is synthetic aperture radar target recognition, the methods discussed will focus primarily on fuzzy logic as a tool to classify, or distinguish between, three different objects. As such, the details involved in acquiring SAR data will be dealt with in brevity, with enough attention given to the details that pertain to the system as a whole. The system will be implemented in five stages with the final output being decision as to which target is identified from the candidate image. These stages are as follows: the data acquisition stage, the preprocessing stage, the extraction stage, and lastly, the decision stage. Each candidate image will be tested using the system, with the system extracting key characteristics and processing these characteristics in a manner that will allow for accurate identification of the candidate image. ^
Engineering, Electronics and Electrical
Betancourt, Benjamin, "A fuzzy approach to automatic target recognition applied to bare and camouflaged synthetic aperture targets" (2007). ETD Collection for University of Texas, El Paso. AAI1449751.