An Integrated Approach of Face Recognition using Sparse and Template Matching

A face recognition system is a computer vision application for automatically identifying or confirm a person from a digital image. In this project an integrated technique is implemented in which face will recognize on both real time and on non-real time. Viola and Jones begins a method to precisely and quick detect faces within an appearance. This technique can be adjusted to precisely detect facial features. We consider the difficulty of automatically recognizing person faces from frontal views with varying expression and illumination, as well as occlusion and dissimulate. New premise from sparse linear representation offers the key to addressing this problem. Based on a sparse representation enumerated by `1-minimization, we propose a common classification algorithm for (image-based) object recognition technique. This innovative framework foresees novel insights into critical problems in face recognition i.e. feature extraction. The sparse representation-based classification (SRC) has been proved to be a robust facial recognition approach. However, its calculation difficulty is very high due to resolving a difficult -minimization problem.