Investigations on dental image based Human identification in forensic Odontology and oral disease Detection using deep learning Approach

Abstract

Biometric are sources of information obtained from the characteristics of individual. This helps in retrieving the identity of human beings. There are several biometrics such as DNA, fingerprint, and face which could be used for human identity retrieval. But the major drawback of this biometrics is that they cannot survive extreme conditions like fire or flood. In situations of mass disasters like Tsunami or Earthquake, the number of casualties caused is very high. Hence, there is a need for identification of victims by some means. To manage this after effects of disaster, the Government of India has initiated Disaster Management Act in the year 2005 which includes Disaster Victim Identification (DVI), so as to identify the fatalities as early as possible. newlineDVI involves the expertise of many departments such as Forensic Odontology, Pathology, Forensic Sciences, Anthropology, and Entomology to identify the deceased individual. Since the casualties caused are large in number, collecting biometrics such as DNA, fingerprint or face is not possible. The only biometric which could withstand any weather conditions is dental features. Forensic Sciences seek assistance from Forensic Odontologists in identifying the victims. Odontologists are those who use dental biometrics to identify human by manual comparison between Ante-Mortem and Post-Mortem radiographs. This manual comparison is time consuming and is subject to error. Hence, there is need to automate this process of human or victim identification. The Tsunami disaster report state that more number of victims was identified using dental features as the only source of information. So to enable Odontologists identify the victims with ease, Automated Dental Identification System (ADIS) was developed newline