A Systematic way of Hybrid model design and comparative analysis of EBGM and eigen values for biometric face recognition using neural network

Face recognition plays an essential role in humanmachineinterfaces and naturally an automatic facerecognition system is an application of greatinterest. Although the roots of automatic facerecognition trace back to the 1960, a completesystem that gives satisfactory results for videostreams still remains an open problem. Research inthe field has been intensified the last decade due toan increasing number of applications that can applyrecognition techniques, such as security systems,ATM machines, “smart rooms” and other humanmachineinterfaces. Elastic Bunch Graph Matching(EBGM) [3] is a feature-based face identificationmethod. The algorithm assumes that the positions ofcertain fiducial points on the faces are known andstores information about the faces by convolving theimages around the fiducial points with 2D Gaborwavelets of varying size. The results of allconvolutions form the Gabor jet for that fiducialpoint. EBGM treats all images as graphs (calledFace Graphs), with each jet forming a node. Thetraining images are all stacked in a structure calledthe Face Bunch Graph (FBG), which is the modelused for identification. For each test image, the firststep is to estimate the position of fiducial points onthe face based on the known positions of fiducialpoints in the FBG. Eigenfaces are a setof eigenvectors used in the computer visionproblem of human face recognition. The approachof using eigenfaces for recognitionwas developed bySirovich and Kirby (1987) and used byTurk and Alex Pentland in face classification. It isconsidered the first successful example of facialrecognition technology. The purpose of this paper isthe implementation of various methods from Twodifferent families of face recognition algorithms,namely the the EBGM and eigenvalues forbiometric face recognition