Wavelet-based non-zero order joint transform corrlator for target recognition in noisy scenes

A novel joint transform correlator for target recognition under noisy conditions is proposed. In this design, strong zero order term is eliminated by modifying joint transform power spectrum through wavelet transform. To obtain sharp correlation peak and noise resistance ability, logarithmic algorithm is applied and yields far better results compared with widely used modified fringe adjusted filter. Computer simulation results are presented to confirm the performance of the proposed technique in additive Gaussian noise environment.