In the application of achieving the correlation recognition of moving target with hybrid optoelectronic joint transform correlator (HOJTC), the differences resulting from the movement between the target and the template and the cluttered background influence the correlation ratio of the correlator. A combined method of wavelet multi-scale edge extraction which uses the first order derivative of Gaussian function and edge processing with morphological dilation is applied. This method can fully use the wavelet multi-scale characteristics. After extracting the edge, the noise can be suppressed and more details can be retained. It can improve the detection efficiency of the moving target in cluttered background apparently. The optical correlation experiments on a low contrast and small moving target show that this method can enhance the brightness of the correlation peaks and verify that the algorithm is feasible to the moving target recognition in cluttered background.