21-08-2014, 10:28 AM
With the advancement of bio-metric security solutions, face recognition is fast catching up as one of the most preferred methods of security. It has the advantages of being fast and user friendly and still gives accuracy in performance that may outshine other complex methods. Since the early 60’s great research work has been made in this field with various algorithms and procedures developed to implement face recognition under varying image conditions with a prime focus to reduce computational requirements and processing time without compromising on accuracy.
The Principal Component Analysis (PCA) is one of the most successful techniques that have been used to recognize faces in images. The PCA algorithm makes use of Eigen faces to find the best match for a given image. There is evidence that PCA can outperform over many other techniques when the size of the database is small. Presented here is an implementation of PCA for face recognition and its corresponding VHDL code. In this project we use Matlab™ for base programming of the algorithm and then translate it into C and VHDL and then model the code for FPGA board. For data communication between the PC and the FPGA board, we use the Ethernet protocol using a P160 Communication Board as the interface. Standard USB and JTAG standards fail because of the increased data stream rate. This calls for a program for the Ethernet protocol which is implemented in C# and stored on the controller flash memory.