QUANTIZATION AND TRUNCATION CONDITIONS OF FOURIER POWER SPECTRUM FOR GOOD PERFORMANCE IN BINARY SUBTRACTED JOINT TRANSFORM CORRELATOR

Binary subtracted joint transform correlator (BSJTC) provides sharp au tocorrelation peaks and better discrimination for similar targets even though many reference images are arranged regularly in an input scene . The effects of the number of reference patterns, the quantization le vels and truncation of the Fourier power spectra on the performance of BSJTC are investigated. The number of effective quantization levels t o obtain sharp and clear autocorrelation peaks is estimated by compute r simulations using the input scenes with many binary images (alphabet ic characters) and halftone images (human portraits). Experimental res ults of BSJTC are also shown using a hybrid system with a Bi12SiO20 Sp atial light modulator and a personal computer.