Face recognition surveillance web application

• Main Author: Hemingway, Josephine Agatha
• Other Authors: Lin Weisi
• Format: Final Year Project
• Language: English
• Published: Nanyang Technological University 2023
• Subjects: Engineering::Computer science and engineering
• Online Access: https://hdl.handle.net/10356/166158
• Institution: Nanyang Technological University

Facial recognition technology is a rapidly evolving scene in today’s world that allows for the automatic identification and authentication of individuals by their facial features. It has a wide range of use cases, particularly in security, surveillance, and access control, as it can be leveraged as a tool to heighten security and safety in many aspects. Currently, there exists some surveillance software in the market that leverages facial recognition technology to detect authorised individuals to grant access or entry into restricted premises. However, these solutions often target security for private organisations and buildings and tend to have an incomplete set of features required for effective surveillance software. This project aims to deliver a cost-effective and comprehensive web-based surveillance application, called iRecognise, that addresses public safety requirements. iRecognise is a video management system that incorporates facial recognition technology for identifying blacklisted individuals in both live surveillance footage and uploaded pre-recorded videos. In addition, iRecognise boasts a user-friendly interface with a variety of novel features, including its customisable interface and real-time alerting feature through the iRecognise Alerts Telegram bot, which alerts users to potential threats. This facilitates the effective and prompt dispatching of relevant authorities for targeted intervention. iRecognise was developed with the React TypeScript library for its front-end components and Flask Python web framework for its back-end component, with its application data stored in MongoDB and Amazon Web Services (AWS) Simple Storage Service (S3). It incorporated state-of-the-art deep facial recognition models from Deepface to recognise blacklisted individuals. This report will discuss the problem in depth and walk through the end-to-end process of designing the web application down to its implementation details. Evaluation of the product was also conducted via comprehensive testing.