This work presents an experimental demonstration of an optical joint transform correlator based on the wavelet subband filter for texture pattern recognition. The optical wavelet subband filter is implemented using 4f filtering architecture and utilized to extract the texture features of fingerprints under noisy environments through frequency-and orientation-selective properties. The filtered texture features with noise reduction are applied to optimize recognition via joint transform correlation. Experimental results show that the optical wavelet subband filter enhances the significant texture features from corrupted fingerprints and increases the pattern discrimination of the joint transform correlator.
ASJC Scopus subject areas
- Atomic and Molecular Physics, and Optics