BINARY NONLINEAR JOINT TRANSFORM CORRELATION WITH MEDIAN AND SUBSET MEDIAN THRESHOLDING

We investigate the performance of the binary joint transform correlator (JTC) for three types of thresholding used to binarize the joint power spectrum. The first thresholding method uses the median of the joint power spectrum of the reference image, which is independent of the effect of input scene noise. The second thresholding method uses the median of the joint power spectrum of the input scene and the reference image. The third method is the subset median thresholding. The subset median is computed for each segment of the joint power spectrum and is used to threshold the same segment. The threshold computed by the second or third method is dependent on the effect of the input scene noise. The correlation performance of the binary JTC is determined for each thresholding method and is compared with the linear (conventional) JTC. Computer simulation is used to determine the correlation peak intensity, peak-to-sidelobe ratio, and correlation width for various thresholding techniques. The results indicate that the thresholding of the joint power spectrum using the methods that take into account the input scene noise produces a reasonably good correlation performance.