Influence of input target characteristics to correlation peak recognition in optical correlation system

Optical correlator technologies play the important role in image processing systems, such as Vander Lugt correlator and Joint Transform Correlator. Compared with traditional computer image processing techniques, optical correlation system have an inherent capability for parallel processing, which can transmit rapidly mass information. Meanwhile, it changes the complex images processing problem into spot recognizing in output plane. As a kind of important image recognizing tool, JTC can be built easily in lab. Abundant research results show the correlation peak of joint transform correlator should be a sharp circle spot when input the same target picture and reference picture. However, experimental result shows the Characteristics of input target influences the intensity distribution of correlation peak in the paper. A group of correlation peak is obtained which is different distribution in the vertical direction and horizontal direction when change length-width ratio of rectangles which is used as the target picture and reference picture. Furthermore, the experimental result proves the distribution of correlation peak change continuously with the different characteristics of rectangle. Under the special condition in experiment course, length-width ratio of correlation peak will climb up to 3.8 when length-width ratio of two input rectangle is 30 to 1. The possible cause is analyzed in paper. According to the above results, more strict identifying condition must be adopted in correlation peak recognition in order to avoid the occurrence of misjudgment.