Face Recognition Techniques and Analysis

This book presents standard as well as novel face recognition methods. These methods utilize Principal Component Analysis, Linear Discriminant Analysis, Independent Component Analysis, Gabor Wavelets, Neural Networks, Hidden Markov Models, Graph Matching, etc. Emphasis is given to the popular Eigenfaces algorithm, which is presented analytically in detail and a framework is presented for its experimental evaluation. In addition, this book presents a novel face recognition method that is computationally efficient and can be implemented as a real-time process. This method operates in quantized block histogram face spaces. Next, a classification algorithm, which inherently applies the optimum classification measure in these spaces, is mathematically derived. The development of this algorithm was motivated by the practical limitations that impair the performance of the Eigenfaces method. To overcome these limitations, theoretical and experimental statistical criteria are derived in order to achieve high recognition rates. Thus, a novel and potent face recognition framework is presented along with other standard face recognition methodologies.