Deep Unified Model for Face Recognition based on Convolution Neural Network and Edge Computing

CCTV, communication, and alarm systems use face recognition technologies. Face detection in photos is a popular topic in science for practical reasons and because it challenges computer-generated vision systems. The variety of shooting situations (position, lighting, hairdo, emotion, backdrop, etc.) and face traits requires versatility. Deep learning-based image identification methods beat machine learning methods in efficiency and information processing. Modern computer systems have major authentication issues. Internet-connected smart devices are producing more data every day. A new model is needed to handle its vast data output. Deep learning and edge computing process vast volumes of data with high precision. Many trust facial recognition systems. SIFT and accelerated robust features are used in traditional facial recognition algorithms (SURF). This paper presents a convolutional neural network-based face identification and recognition solution that outperforms established methods. Tagged photographs of people taken in the outdoors teach the face-recognition algorithm (LFW). The suggested system had 99.1% accuracy on test data.