A Comparative Implementation of PCA Face Recognition Algorithm

Abstract

This paper presents PCA algorithm used in face recognition system and its implementation on different architectures in order to choose the best solution for designing a real time face recognition system. Benchmarks and comparisons will be given for PC, DSP and FPGA and results will show that FPGA soft core is too slow for this computation. An IP solution is the best choice. But DSP is also a good compromise for implementing easily the algorithm and having acceptable processing time.