Suspect identification by matching composite sketch with mug-shot

摘要

The problem of automatically matching composite sketches to mug-shot is addressed in this paper. To determine the identity of a criminal, composite sketch to photograph is commonly used technique due to budgetary reason. The accuracy in detection is more using forensic sketches than composite sketches using facial composite software. Composite sketches have the same modality as that of the mug-shot has that changed the forensic world. In this paper emphasis is made on matching composite sketches to mug-shots that gives better performance over the more common representation method exists already. The component based framework presented in this paper consists of the following major steps. 1) Face normalization using a geometric transformation and color space conversion. 2) Facial component localization using Viola-John algorithm, followed by ASM. 3) Per component feature extraction using nearest neighbor symmetry and matrix normalize cross-correlation followed by computing length between facial constituent The proposed system has been evaluated on mugshot photographs-composite sketches pairs with hundreds of individuals. The results are encouraging.