Real Time Embedded Implementation of Face Recognition Using Beagle Board

This Paper presents a Standalone low cost face recognition system. When compared to other bio-metrics systems such as finger print, palm print and Iris, face recognition has distinct advantages because of its non contact process. To secure unauthorized access of the data, this face recognition process will be more secure to develop in future for longer security life. This paper investigates alternative methods to be used for face recognition system by different process. It involves face detection and verification process using Gabor wavelets for huge data sets taken under this method. Then for feature extraction process, we extracted Color-based model, Coarse to fine method, and Gabor wavelets are applied. It has been designed to improve the accuracy of recognition systems in a huge data set, particularly focused on verification of the person rather than identification.