Semester

Fall

Date of Graduation

2020

Document Type

Thesis

Degree Type

MS

College

Eberly College of Arts and Sciences

Department

Forensic and Investigative Science

Committee Chair

Keith Morris

Committee Member

Jacqueline Speir

Committee Member

Debra Ayers

Abstract

Of the many biometric traits recognized today, fingerprints are the most prevalent and familiar. The analysis of fingerprints involves level 1, level 2, and/or level 3 detail in the identification of a potential match. Traditionally, fingerprint matching was completely performed by hand, utilizing the ACE-V method. Thanks to the development of rapidly evolving technology, fingerprint matching has become an automated procedure through the use of fingerprint matching algorithms. In the literature, there has been an increase in the interest of developing Automatic Fingerprint Identification System (AFIS) algorithms that include level 3 details in the matching process. These studies have utilized live scanned and/or inked fingerprints, rather than latent fingerprints. However, practical use of AFIS algorithms involves unknown fingerprints, such as those collected at crime scenes, which are often latent in nature. In addition, research has also found that there is a wide variety in size and shape of pore structure, making automatic detection of pores difficult. The resultant quality of latent fingerprints is subject to various factors at the time of deposition, such as the deposition surface, environmental conditions, and composition of the fingerprint itself. Consequently, these factors, in addition to the inherent variance in pore structure, may very well affect the observance and use of level 3 details within a fingerprint. If the prevalence of pores proves to be unreliable and inconsistent in latent fingerprints, the push for including level 3 detail in the AFIS matching process may all be for nothing. For this reason, the effects of latent fingerprint deposition factors on pore identification needs to be considered and currently appears to be greatly under studied. In effort to begin to fill this gap in the current research, newly deposited latent fingerprints were collected and developed using both black fingerprint powder and cyanoacrylate fuming. Developed fingerprints were subsequently imaged via digital scan or digital camera, and enhanced using either Image J or Adobe\textsuperscript{\textregistered} Photoshop\textsuperscript{\textregistered}. Following image enhancement, pores were manually identified and marked using the Federal Bureau of Investigation (FBI) developed Universal Latent Workstation (ULW) software.

Qualitative assessment of the 633 fingerprints collected resulted in 380 usable fingerprints for the remainder of the study. Observations regarding pore count within the replicate fingerprint sets indicated that total pore count/presence was not consistent. The Mann Whitney U test indicated that neither development method, black fingerprint powder nor cyanoacrylate fuming, produced pore data any better or worse than the other. Lastly, assessment of pore location resulted in a greater number of similarity scores being lower than the established threshold, indicating that pore location is not as easily assessed nor interpreted as hoped.

Download

Share

COinS