Comparative survey analysis of the CNN and LBPH Face Recognition Learning Algorithms: A Survey of Learning Approaches and Performance Evaluation

Face recognition technology has made it possible for computers to recognize human faces. It is an application that uses algorithms to extract distinctive traits from facial images and compares them with a database of recognized faces. Facial recognition has numerous real-world uses in social networking, healthcare, access control, security and surveillance, and more. Due to the acquisition and processing of sensitive personal data, it has also sparked questions about ethics and privacy. The Deep Learning Convolutional Neural Network and the Local Binary Pattern Histogram (LBPH) are two face recognition algorithms that are compared in this paper. The study aims to evaluate the performance of these algorithms under different conditions, including variations in pose, expression and illumination. The relevant theories of LBPH and CNN are first briefly discussed, and the performance comparison method is then described. The choice between these two algorithms will depend on the particular needs of the application and the resources that are available. Overall, this work offers insightful information regarding the advantages and disadvantages of LBPH and CNN for face recognition, assisting in the selection of an algorithm for practical applications.