Authors: Assistant Professor T.Balanandhi, E. Pravalika
Abstract: The increasing number of missing and unrecognized persons has become a major social and public safety concern worldwide. Traditional methods of identifying missing individuals rely heavily on manual investigations, newspaper advertisements, police records, and public notifications, which are often time-consuming and inefficient. Recent advancements in artificial intelligence and computer vision have enabled the development of intelligent face recognition systems capable of automatically identifying individuals with high accuracy. This paper presents an intelligent web-based missing person identification system that integrates an optimized face recognition algorithm with the Django web framework to facilitate rapid identification of missing and unrecognized persons. The proposed system utilizes Histogram of Oriented Gradients (HOG) for face detection, dlib facial landmark extraction, 128-dimensional face encoding, and Euclidean distance-based face matching to compare uploaded images against a centralized database of known individuals. The web application incorporates Google Authentication, SQLite database management, secure user registration, anonymous reporting, and automated alert generation to improve accessibility and public participation. The system maintains separate databases for known and unknown individuals while allowing authorized users to upload missing person information and register unidentified persons through an intuitive web interface. Experimental evaluation demonstrates that the optimized face recognition model effectively identifies matching facial images while rejecting blurred, duplicate, fake, and non-face images, thereby improving the reliability of the identification process. The integration of artificial intelligence, face recognition, and web technologies provides an efficient, secure, and scalable solution for assisting law enforcement agencies and the general public in locating missing persons while reducing the time and effort required for manual investigations. The proposed framework offers significant potential for deployment in public safety applications and future integration with real-time surveillance systems for continuous missing person identification.
