Human Identification using the Individual's Dental Radiographs Record

O. Nomir (Egypt) and M. Abdel-Mottaleb (USA)


Forensic dentistry, biometrics, segmentation, Antemortem(AM) dental radiographs, postmortem (PM) dental radiographs.


Dental biometrics automatically analyzes dental X-ray radiograph records to identify the deceased individuals based on their dental features. Human identification based on dental features has always played a very important responsibility in forensics. Dental Biometrics relies on dental radiographs to identify victims in situations (victims of motor vehicle accidents, violent crimes, and work place accidents.) where conventional biometric features, i.e., DNA, fingerprint, face, and iris may not be available. The dental features are considered as the best candidates for a PM biometric identification. In dental biometrics, given a dental postmortem (PM) dental radiograph records, the identification procedure is carried out by searching a database of antemortem (AM) radiographs to determine the identity of the individual which associated with the given PM dental records. In our research, we developed a human identification system based on dental features. In this system, we relied on the bitewing images to identify individuals. In this paper, we address the problem of identifying individuals based on more than one type of dental radiographs for the first time. To achieve that, we enlarge the dental database, where different types of radiographs for a single person are available. Also, we extend our previously introduced identification system to handle panoramic and periapical radiographs. In order to improve the performance of our identification system, we rely on the three different types of dental radiographs. The identification procedure is carried out by integrating the results of using the individual’s three available types of dental radiographs. The experimental results show that using the three types of dental radiographs enhances the overall human identification procedure.

Important Links:

Go Back