Novel Optical Wavelet Correlator for Cluttered Target Identification

We propose to develop a novel optical wavelet correlator for automatic recognition of cluttered targets. Instead of computing the correlation between the target and the reference patterns, the optical wavelet correlator computes the correlation between the wavelet transform coefficients of the target and reference patterns. It incorporates the efficient feature detection and noise suppression capabilities of wavelet transformation and the inherent high speed and two-dimensional nature of optics. The wavelet matched filter (WMF) is synthesized with the Fourier spectra of the reference pattern and the selected wavelets. This WMF replaces the conventional matched filter (CMF) in the optical wavelet correlator. When an input signal enters the wavelet correlator, the feature extraction and target identification operations are performed in a single step to produce the correlation between the WTs of the input scene and the refernce pattern. By suitably choosing the wavelet functions and their dilation factors, the proposed wavelet correlator is capable of effectively recognizing targets in the presence of clutter and noise.