Abstract: In the present study we present an innovative approach towards countering the problem of partial occlusion in face recognition scenario. The partial occlusion can be caused by various objects such as scarfs, sunglasses etc., and its effects are confounding in the performances of the recognition rates. The advantage that the adopted pre-processing algorithm poses before face recognition steps is to eliminate the distortions due to the variance in light illumination field at the given instance when the facial image is recorded or captured. The framework tends to mathematically model the curvature and other essential features of the face such as micro-expression and the curves of the facial regions. This, significantly enhances the probability of matching the parent image to that of the occlude image that is how multiple object recognition using hybrid approach. The presented algorithm is tested over Extended Yale B & CMU PIE standardized datasets. Over the years biometrics has gained unparalleled popularity in digital medium and has proven its usefulness for several applications concerned with the threats and crime or security purposes. Face Recognition is a widely emerging biometric for automating the surveillance, as it has aid in strengthening the security from several types of terrorist or criminal threats. Though, there are several face recognition techniques which are categorized based on its error rates in recognition but there are few that gives the marginal rate for sufficient and validated recognition rates for occlude images.
Keywords: Hybrid Neuro Fuzzy Network, Face Recognition, Neural Network, Normalized Rapid Descriptor (NBD), PCA, LDA, NMF, LNMF, ICA.
|
DOI:
10.17148/IJARCCE.2020.9707
