Distortion-invariant pattern recognition with nonlinear correlationfilters

摘要

Classical correlation-based methods for pattern recognition are very sensitive to geometrical distortions of objects to berecognized. Besides, most captured images are corrupted by noise. In this work we use novel nonlinear composite filtersfor distortion-invariant pattern recognition. The filters are designed with an iterative algorithm to reject a backgroundnoise and to achieve a desired discrimination capability. The recognition performance of the proposed filters is comparedwith that of linear composite filters in terms of noise robustness and discrimination capability. Computer simulationresults are provided and discussed.