Automatic Human Dental Identification based on Posterior and Anterior Teeth

Forensic dentistry is the subdivision of forensics that determines the identity of the victim based on dental features. In this paper, we propose a framework for human identification based on dental. The proposed framework consists of two main stages: the first stage is teeth classification and numbering, and the second stage is human recognition. In this study, a new feature has been proposed for each of these two stages: Crown mesiodistal neck and anatomic crown length for the first and a feature vector based on the geodesic distance between important points of teeth contours for the second. The proposed method is capable of solving principally, and automatically problems such as diagnosis of posterior and also anterior teeth, posterior and anterior teeth classification, diagnosis of number and kind of all kind pulled teeth from, which are overlooked or have been left with the simple premises in previous works. To evaluate the proposed method, we experiment on a set of bitewings, periapical and panoramic images. The practical results show an improvement in accuracy of classification with discovering anterior teeth, and also 9% improvement in the first rank of the accuracy of teeth recognition, in comparison with the previous works.