FACE RECOGNITION ON A STREAMING VIDEO SERIES USING THE OPENCV LIBRARY

Abstract

Face Detection System is a technology that can match a human face to a digital image or video frame to a database of individuals, commonly used to authenticate users through identity verification services, and works by accurately identifying and measuring facial features in a given image. Face recognition systems are used today by governments and private companies around the world, their effectiveness varies, and some systems have previously been written off due to their inefficiency. Thus, the creation of a program for human face recognition is a topical issue. The aim of the article is to study the theoretical aspects of the development of the human face recognition system and the practical implementation of the relevant software package. The face recognition procedure simply requires that any device equipped with digital photographic technology generate and receive the images and data necessary to create and record a biometric image of the person to be identified. The main methods of face recognition are considered: geometric methods, principal components method, flexible comparison method on graphs, Viola-Jones method, binary templates, neural networks. The implementation of the algorithm of the face recognition system is proposed. This paper analyzes the existing algorithms and systems for face detection and recognition, weighing their advantages and disadvantages. The percentage of human face recognition accuracy and performance were analyzed in practice, taking into account such factors as lighting, image quality, number of faces in the image, human face recognition using Local Binary Patterns (LBP) using the OpenCV computer vision library.