Semester

Spring

Date of Graduation

2009

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

Yenumula V Reddy

Committee Co-Chair

Hany H Ammar

Abstract

Forensic scientists are using different kinds of systems and techniques for the identification of missing and deceased individuals. Researchers from West Virginia University, Michigan State University, and University of Miami proposed an Automated Dental Identification System (ADIS) which utilizes dental features for the post mortem identification process. A web-based ADIS, namely Web-ADIS, was developed. The initial testing shows that the system can be used efficiently with a small database.;A new and improved system design has been proposed in this research enabling practical use of the ADIS with a large database. We designate this new system as High Performance Web Automated Dental Identification System Grid (HP Web-ADIS Grid). The ADIS tool uses a set of default databases provided by FBI and other law enforcement agencies for identification and matching. However, certain situations like a plane crash, earthquakes, and tsunamis etc. require special reference databases. The proposed system design provides users with the ability to upload their own reference databases for matching purposes. Analysis of the previous work done on ADIS tool revealed the need for performance enhancement. We observed that the Image Comparison component is very slow and time consuming. To improve system performance, a parallel processing grid based system is proposed. The HP Web-ADIS Grid design utilizes multiple machines to process requests resulting in significant reduction in the processing time. The performance testing of HP Web-ADIS was conducted by comparing a system utilizing a single node with that of the proposed grid based configuration. Results obtained show an exponential improvement in the performance of ADIS significantly reducing the overall time for performing record to record comparisons. Forensic scientists would benefit greatly from this improved and highly efficient system enabling faster retrieval of results.

Download

Share

COinS