| The main purpose of this research was to develop performance metrics for evaluating the performance of joint transform correlator (JTC) based pattern recognition techniques. Existing metrics such as the correlation peak amplitude and implied weighting of the templates using edges are not adequate. Therefore, the focus of this research was to develop an effective and comprehensive set of performance metrics for measuring the performance of JTC techniques. The performance of all important JTC techniques such as classical JTC, fringe-adjusted JTC, and reference phase-encoded fringe-adjusted JTC were determined using performance metrics such as correlation peak intensity, peak-to-correlation energy, peak-to-background correlation energy, peak-to-sidelobe ratio, peak-to-clutter ratio and Fisher ratio. These performance metrics were used to estimate the reliability of pattern recognition techniques with respect to clutter, noise and other associated distortions. Simulation results are presented to verify the effectiveness of these performance metrics on measuring the performance of various JTC based pattern recognition techniques. |
