Face Detection System for Criminal

In this paper a criminal detection agenda that could help police to know the face of a criminal or a suspicious is planned. The agenda is a client-server video based face recognition investigation in the actual. The agenda smears face detection and tracking using Android mobile devices at the client side and video based face recognition at the server side. This paper attentions on the growth of the client side of the future agenda, face detection and tracking using Android mobile devices. For the face detection phase, strong Viola-Jones algorithm that is used. The face tracking phase is founded on Optical Flow algorithm. Optical Flow is applied in the future agenda with two feature extraction methods, Fast Corner Features, and Regular Features. The planned face detection and tracking is executed using Android studio and OpenCV library, and verified using Sony Xperia Z2 Android 5.1 Lollipop Smartphone. Trials show that face tracking using Optical Flow with Regular Features attains a higher level of accurateness and effectiveness than Optical Flow with Fast Corner Features.