Preliminary Findings: Use of CNN Powered Criminal Identification System

Face is the principal distinguishing attribute of the human body in recognizing an individual. Face recognition is vital for the law enforcement authorities in primary detection and identification of the criminals. However, it becomes very difficult and time-consuming to manually identify every individual in highly populous cities and public places. In this paper, we propose an automatic criminal identification technique face recognition of the criminals using history of their criminal activities. The proposed system will detect the detailed features of human face and recognize the suspect entering a public location. For this purpose, we have built a facial recognition system to apply in Convolutional Neural Networks (CNN). In the simulation, we use the Labeled Faces in the Wild (LFW) dataset to train and test the CNN-powered face recognition system. Preliminary findings prove the feasibility of implementing this system in effective detection and identification of individuals. Such an automatic system will be operative to identify and catch the criminals in public places in a more efficient way.