FACE RECOGNITION SYSTEM USING MULTIPLE RECOGNIZER

Abstract

Face recognition has been studied for many years and is expected to be widely used in daily identification systems, communication systems, public security systems, and in law enforcement systems. Inspired from the human vision system, we combined the conventional learning algorithms and image processing algorithms with predefined rules to increase the intelligence of machine recognition systems. As the first step, face detection is implemented by an industrial image-based face detector combined with novel temporal differencing algorithms. The face detection result, an industrial image-based classifier, temporal filtering and video context related rules are all combined for face recognition. Simulation results show that, the proposed system, which is based on mixtures of subspaces, is effective for face detection. It can also be concluded that the subspace methods can be used effectively, in general object detection problems. Moreover, for colour images, observations show that, the proposed pre-processing step enhances the performance of the system. Here, we are proposing an algorithm for the face recognition based on fusion of multiple recognizers namely Fisher’s linear discriminant (FLD) and eigen-face so that we can overcome the limitation of single recognizer and improve the performance of the overall recognition system. The images of a human face lie in a complex subset of the image space that is unlikely to be modelled by a single linear sub-space; we use a mixture of linear subspaces to model the distribution of face and non-face patterns. This approach is used to overcome the drawback of the eigen-face approach by integrating Fisher’s linear discriminant (FLD) criteria, while retaining the idea of the eigen-face in projecting faces from a high-dimension image space to a significantly lower-dimensional feature space.