Fuzzy Pattern-Based with Laplacianfaces Biometric Pattern Matching Algorithm for Face Recognition

In face recognition system, the efficiency and face detection rate is the key issue for designing the algorithm because the computer can recognize very large number of faces with high computational speed but it is not as efficient as human. There are many modern face recognition algorithms which efficiently detect face but also have false recognition. In this paper, we propose a new hybrid algorithm and method FPBL (Fuzzy Pattern-Based with Laplacianfaces) for face detection to enhance the true recognition rate and minimize the false recognition rate. We analyze and compare the performance and results of modern pattern matching and face recognition algorithms like Eigenfaces, Fisherfaces, Laplacianfaces and Fuzzy method with FPBL. The implementation result shows that performance and accuracy of our proposed method gives the efficient and significant face recognition rate compared to other algorithms.