Facial Recognition System Using DWT, DCT, and Multilayer Sigmoid Neural Network Classifier

Abstract

Facial recognition systems have seen widespread use in numerous applications, including identity verification for phone security, missing person identification, and forensic investigations. The purpose of this study is to improve both the speed and accuracy of a facial recognition system, thus enhancing its suitability for real-world applications. The proposed system reduces overall computational complexity by using simple algorithms and transforms such as grayscaling, a two-dimensional discrete wavelet transform, and a two-dimensional discrete cosine transform. The classification algorithm increases accuracy by using a straight-forward multilayer sigmoid neural network, which better correlates the input and output data than existing methods. The recognition system is tested with four freely accessible datasets: the ORL, YALE, FERET-c, and FEI. A test set based on the combination of all datasets is also utilized to evaluate the system performance. Results show that the system still maintains high recognition rates despite reducing complexity compared to popular existing methods.