Analysis and Matching of Composite Sketches to Facial Photograph Using Component Based Approaches
- DOI
- 10.2991/iccasp-16.2017.74How to use a DOI?
- Keywords
- LBP ASM PCA
- Abstract
Facial Sketches are most widely used in law enforcement agencies for identification and apprehension of suspects, which may involve in several criminal activities. If the photographs of the suspect is stored in the mug shot data base, then the identity of the suspect is possible, but in most of the cases that it is a possibility that the photographs of the suspect is not available then two methods are used first is forensic sketch and the second is composite sketch software. The forensic sketch which is drawn by the expert artist according to verbal description provided by an eyewitness Unlike sketches, hand drawn by artist, composite sketches are drawn using one of the several facial composite software system (FACES and Mac-a-Mug etc). We propose a component based representation (CBR) approach that is used to measure the similarity between composite sketches and facial photographs. First the preprocessing technique is applied on both sketches as well as on face photo and after that components are extracted using 13 point marking and after that the multi scale local binary pattern (MLBP) is applied on each of the extracted region of face photo and sketch and finally the similarity which is obtained from individual facial photograph and sketch components are fused together so, that the total components similarity is obtained. Experimental results for matching 100 composites against 100 facial photographs show that the proposed method achieves (56% to 70%) performance as compared to a leading commercial face recognition system. We believe that our prototype system will be of great value in law enforcement agencies for catching a flitting suspect in a timely fashion
- Copyright
- © 2017, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - P. Mohod AU - S. Khandait PY - 2016/12 DA - 2016/12 TI - Analysis and Matching of Composite Sketches to Facial Photograph Using Component Based Approaches BT - Proceedings of the International Conference on Communication and Signal Processing 2016 (ICCASP 2016) PB - Atlantis Press SP - 501 EP - 508 SN - 1951-6851 UR - https://doi.org/10.2991/iccasp-16.2017.74 DO - 10.2991/iccasp-16.2017.74 ID - Mohod2016/12 ER -