Face Detection and Recognition for Forensic Investigation

Abstract

As the crime rate is increasing day by day, it is important for forensic officers to keep records of all criminals and help the officers to identify them in future if they attempt any crime in spite of having changes in their facial features. System also identify the different images with beard, with glasses, with age increase etc. This thesis is divided into two approaches face detection and face recognition. Different methods were also analyzed for face detection and recognition. System uses OpenCV for the face detection and face recognition. Different face detection algorithms eigenface, fisherface, voila-joins were also tested on a different type of face images. In this system, face recognition system is done using principle component analysis(PCA) and support vector machine(SVM). Face detection is initial stage of face recognition, where it detects the face from a particular image. The face recognition system is also expanded into a pose-invariant face recognition system which is implemented and tested on facial images for subjects with different poses. In this system, we used face database that contains 165 grey-scale images in GIF configuration of 15 people. and all subjects contain a different type of pose. So in fully face recognition system was successfully released.