Date of Graduation

2001

Document Type

Thesis

Degree Type

MS

Committee Chair

Hany H. Ammar

Abstract

In this work we apply combined digital image processing and artificial neural network techniques in developing a prototype of an automatic dental identification system (ADIS) to be employed for postmortem forensic identification. The prototype ADIS follows a pyramidal image-matching architecture used in developing an analogous fingerprint comparison system. Owing to the differences in image natures, we used new techniques and modified some mathematical models. We introduce an adaptive technique for dental radiograph enhancement and an adaptive strategic searching technique for neural network training. The training and testing results we obtained are promising.

Share

COinS