Real-Time Intelligent Surveillance System

A Real-Time Surveillance System is proposed, enhancing traditional surveillance techniques with novel Face and Emotion recognition technologies. Primarily tested on a dataset containing people of South Asian descent, the system employs the DeepFace algorithm for 3D frontalization to accurately detect individuals' emotional states within a frame. In parallel, it uses Dlib, for reliable identity verification. Upon identifying potential threats, the system efficiently triggers push notifications to designated authorities, conveying essential information such as the individual's identity, emotional state, and location. Experiments demonstrated the system's effectiveness in emotion detection, recognition, and Face recognition, offering a comprehensive solution for enhancing security and surveillance capabilities.