Face Sketch Construction & Suspect Identification for Forensic Analysis

Abstract

: The rising crime rate in today’s society requires quicker and more effective ways to identify suspects. Traditional methods, like manually drawn forensic sketches, have limitations due to the lack of skilled artists and lengthy processes. This paper presents a self-contained, cloud-based platform that allows law enforcement to create accurate composite facial sketches without needing artistic skills. A drag-and-drop interface helps users assemble facial features, which are then automatically matched against criminal databases using deep learning algorithms. The system includes secure authentication through machine locking and two-factor verification, ensuring data privacy. By using cloud technology, forensic sketch creation, and neural networks, the platform greatly reduces identification time, improves accuracy, and remains compatible with hand-drawn sketches. Experimental results with the CUFS dataset show better construction performance compared to traditional methods. This innovation gives law enforcement agencies a scalable and reliable tool for real-time suspect tracking, allowing for faster and more accurate criminal identification.