Semester

Summer

Date of Graduation

2006

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Lane Department of Computer Science and Electrical Engineering

Committee Chair

Hany Ammar

Abstract

Forensic dentistry is concerned with identifying people based on their dental records. Forensic specialists have a large number of cases to investigate and hence, it has become important to automate forensic identification systems. The radiographs acquired after a person is deceased are called the Post-mortem (PM) radiographs, and the radiographs acquired while the person is alive are called the Ante-mortem (AM) radiographs. Dental biometrics automatically analyzes dental radiographs to identify the deceased individuals. While, ante mortem (AM) identification is usually possible through comparison of many biometric identifiers, postmortem (PM) identification is impossible using behavioral biometrics (e.g. speech, gait). Moreover, under severe circumstances, such as those encountered in mass disasters (e.g. airplane crashes and natural disasters such as Tsunami) most physiological biometrics may not be employed for identification, because of the decay of soft tissues of the body to unidentifiable states. Under such circumstances, the best candidates for postmortem biometric identification are the dental features because of their survivability and diversity.;In my work, I present two different techniques to classify periapical images as maxilla (upper jaw) or mandible (lower jaw) images and we show a third technique to classify dental bitewing images as horizontally flipped/rotated or horizontally un-flipped/un-rotated. In our first technique I present an algorithm to classify whether a given dental periapical image is of a maxilla (upper jaw) or a mandible (lower jaw) using texture analysis of the jaw bone. While the bone analysis method is manual, in our second technique, I propose an automated approach for the identification of dental periapical images using the crown curve detection Algorithm. The third proposed algorithm works in an automated manner for a large number of database comprised of dental bitewing images. Each dental bitewing image in the data base can be classified as a horizontally flipped or un-flipped image in a time efficient manner.

Share

COinS