Pattern Recognition 38 (2005) 1295 – 1305 www.elsevier.com/locate/patcog A system for human identification from X-ray dental radiographs  Omaima Nomir, Mohamed Abdel-Mottaleb ∗ Department of Electrical and Computer Engineering, University of Miami, 1251 Memorial Dr., Coral Gables, FL 33146, USA Received 3 May 2004; received in revised form 16 December 2004; accepted 16 December 2004 Abstract Forensic odontology is the branch of forensics that deals with human identification based on dental features. In this paper, we present a system for automating that process by identifying people from dental X-ray images. Given a dental image of a postmortem (PM), the proposed system retrieves the best matches from an antemortem (AM) database. The system automatically segments dental X-ray images into individual teeth and extracts the contour of each tooth. Features are extracted from each tooth and are used for retrieval. We developed a new method for teeth separation based on integral projection. We also developed a new method for representing and matching teeth contours using signature vectors obtained at salient points on the contours of the teeth. During retrieval, the AM radiographs that have signatures closer to the PM are found and presented to the user. Matching scores are generated based on the distance between the signature vectors of AM and PM teeth. Experimental results on a small database of dental radiographs are encouraging.  2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved. Keywords: Dental radiograph; Forensic odontology; Human identification; Image segmentation 1. Introduction Under severe circumstances, such as those encountered in mass disasters (e.g., airplane crashes) or if identification is being attempted more than a couple of weeks postmortem, most physiological biometrics may not work, because of the decay of soft tissues of the body. Therefore, a postmortem biometric identifier has to resist the early decay that affects body tissues. Because of their survivability, the best candi- dates for postmortem biometric identification are the dental  This research is supported in part by the US National Science Foundation under Award number EIA-0131079, the research is also supported under Award number 2001-RC-CX-K013 from the Office of Justice Programs, National Institute of Justice, US Department of Justice. ∗ Corresponding author. Tel.: +1 305 284 3825; fax: +1 305 284 4044. E-mail addresses: onomir@umsis.miami.edu (O. Nomir), mottaleb@miami.edu (M. Abdel-Mottaleb). 0031-3203/$30.00  2005 Pattern Recognition Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.patcog.2004.12.010 features. Forensic odontology [1] is the branch of forensics concerned with identifying humans based on their dental features obtained from dental radiographs. Dental features are very important as the best candidates for postmortem biometric identification [2]. Currently, the identification is carried out manually by comparing extracted features from a postmortem (PM) dental record to extracted fractures from a database of antemortem (AM) records. The forensic experts confirm the identity based on distinctive features such as dental restoration, dental work features, root morphology and teeth morphology [3,4]. In many cases these features are not enough to correctly identify individuals and this manual identification approach is useful only for verifying individuals in a small database. The goal of our research is to automate the process of extracting a representation from the dental radiographs and to automate the process of matching PM and AM records [5–8]. This will facilitate for forensic odontologists to search through large numbers of AM records and only manually verify the few best matches.