Suspect Identification By Matching Composite Sketch With Mug-Shot
To match composite sketch-to-mug shots several questions rose for consideration or solution, approach we propose
here will provide possible solution. We propose a component-based representation (CBR) approach to measure the similarity
between a composite sketch and mugshot photograph that evidences advances in matching performance over the more
common mug-shot to composite sketches matching methods. To determine the identicalness of a person who is guilty of crime
or serious offense composite sketch to facial image is commonly used technique. Composite sketches are composed using
facial composite software. In this paper emphasis is made on matching composite sketches to mug-shots that gives better
performance over the more common matching proficiency exists already. The component based theoretical account presented
in this paper chiefly consists of the following steps. 1) Renormalize mugshot using a geometric transformation, image
enhancement and color space conversion. 2) Facial constituent localization using a Viola-Jones algorithm and active shape
model (ASM) 3) per component feature extraction using nearest neighbor symmetry and matrix normalize cross-correlation
followed by computing length between facial constituent. 4) Per component similarity measurement followed by score
normalization and fusion. The proposed system has been evaluated on mugshot photographs–composite sketches pairs with
hundreds of individuals. The results are advocating.
Index Terms- ASM, CBR, Composite Sketch, Forensic Sketches, Mugshot, Viola-Jones Algorithm.