Face recognition using PCA on FPGA based embedded platform

Applications such as surveillance, access management and law enforcement demands a real time embedded platform for face recognition. The algorithms like Linear Discriminant Analysis (LDA), Discrete Cosine Transform (DCT) and Principal Component Analysis (PCA) are being used for face recognition. Among this algorithm PCA is chosen in this work because of its better performance for face recognition. Hence, in this work, an FPGA based embedded platform using TSK 3000a processor has been developed to implement PCA algorithm for real time face recognition. This proposed model captures image of a person through a camera and execute in embedded processor to recognize image of the person stored in database and displays it on screen. Experiments are performed to validate the proposed embedded system for face recognition on various persons. Since this a is re-configurable system, it can be used to accommodate more number of users to this proposed system.