Generic face recognition, feature extraction and edge detection using optimal DSNR expansion matching

Expansion matching optimizes a matching criterion called discriminative signal to noise ratio (DSNR) and has been shown to robustly recognize templates under conditions of noise, severe occlusion and superposition. The optimal DSNR matching filter for multiple templates is used to create a generic face filter, by requiring the designed filter to elicit equal responses to all the training faces. A new family of optimal DSNR edge detectors is introduced, based on the expansion filter of several edge models. The step edge detector is compared with the Canny edge detector (CED). Experimental comparisons show that the authors´ edge detector is superior to the CED in terms of DSNR, even under very noisy signal conditions. Expansion matching is also successful in extracting features such as corners from images. Experimental results of corner extraction are presented