Various matched filter based architectures have been proposed over the last two decades to optimize the target detection and recognition performance. While these techniques provide excellent performance with respect to one or more parameters, a unified and synergistic approach to evaluate the performance of these techniques under the same constraints is yet to be done. Consequently, in this paper, we used a set of generalized performance metrics for comparing the performance of the recently reported matched filter based techniques using various types of infrared and SAR datasets. Test results obtained using the aforementioned datasets and performance metrics provide excellent information with respect to the suitability of existing filter based techniques for various target detection and tracking practical applications.

Date Published: 5 May 2014
PDF: 12 pages
Proc. SPIE 9094, Optical Pattern Recognition XXV, 90940C (5 May 2014); doi: 10.1117/12.2054358

Published in SPIE Proceedings Vol. 9094:
Optical Pattern Recognition XXV
David Casasent; Tien-Hsin Chao, Editor(s)