Secure Access Control Using Face Recognition

• Version
• Download 5
• File Size 632.59 KB
• File Count 1
• Create Date 4 July, 2025
• Last Updated 4 July, 2025

Authors : Anagha Borkar , Sakshi Deote , Saloni Nimje , Vaishnavi Dandekar, Vaishnavi Gomase ,Dr. S.S. Golait , Ms. Ashwini Lokande

DOI : 10.46335/IJIES.2025.10.3.7

Abstract – Face recognition has emerged as a leading biometric technique for secure and efficient person authentication, offering a user-friendly alternative to traditional password-based systems. Th is project, titled "Face Recognition for Person Authentication Using Deep Learning," aims to develop a highly accurate and robust face recognition system for user authentication. Leveraging state-of-the-art deep learning models such as ResNet50, Efficient Net, VGG16, VGG19, and VGG Face, the system will recognize and authenticate users based on their facial features. The project workflow begins with the collection and cleaning of a facial dataset, followed by preprocessing steps such as image resizing, noise removal, and histogram equalization to enhance image quality. The dataset will be split into training, validation, and testing sets in an 80:10:10 ratio. Multiple deep learning models will be trained and evaluated to select the optimal architecture and hyperparameters, with key performance metrics like precision, recall, F1-score, and accuracy used to gauge model effectiveness. The system is designed to offer a multi-step authentication process. Users will log in using their credentials, followed by an OTP verification. Once authenticated, registered users will undergo face recognition to confirm their identity, while new users will be prompted to register by providing their email and real-time facial data. The web interface, developed using HTML, CSS, and JavaScript, is integrated with the backend via Flask to ensure seamless interaction between the user and the deep learning model. The system provides a secure, real-time authentication process, ensuring that only legitimate users gain access while alerting account owners of unauthorized access attempts. This project delivers a comprehensive solution combining advanced deep learning techniques with practical real-world applications in secure authentication.