Error Rates for Latent Fingerprinting as a Function of Visual Complexity and Cognitive Difficulty

• NCJ Number: 249890
• Author(s): Jennifer Mnookin; Philip J. Kellman; Itiel Dror; Gennady Erlikhman; Patrick Garrigan; Tandra Ghose; Everett Metler; Dave Charlton
• Date Published: May 2016
• Length: 59 pages
• Grant Number(s): 2009-DN-BX-K225
• Sponsoring Agency: National Institute of Justice (NIJ)
• Sale Source: US Dept of Justice NIJ Pub
• Publication Format: Document; Document (Online)
• Publication Type: Research (Applied/Empirical); Report (Study/Research); Report (Grant Sponsored)
• Language: English
• Country: United States of America

Annotation

The purpose of this research was to increase knowledge about the relationship between the measurable, visual dimensions of fingerprint pairs and the level of comparison difficulty for human examiners, both experts and, to a lesser extent, novices.

Abstract

The results show the plausibility of using objective fingerprint image metrics to predict expert performance and subjective assessment of the difficulty of a particular fingerprint comparison. Among the factors related to difficulty in making comparisons were variables related to image quality metrics, such as intensity and contrast information, as well as measures of information quantity, such as the total fingerprint area. Configurable features were also noted by the experts, such as the presence and clarity of global features and fingerprint ridges. The study also examined the extension of these findings to settings that better approximate real-world fingerprint examiner scenarios. The researchers' regression model continued to provide significant explanatory value for a substantial portion of the prints. Although additional research is needed, the current research provides support for the plausible but previously untested assumption that for expert fingerprint analysis, difficulty and error rate are significantly linked to measurable, visual dimensions of print comparison pairs. In addition, the experiments showed that novice examiners performed poorly and showed no consistent pattern of feature use. The project assembled a new database of latent fingerprints, matching ten-prints, and close, non-matching ten-prints. Using this database, researchers measured expert examiner performance and judgments of difficulty and confidence in a variety of settings. For the experts, the project developed a number of quantitative measures of image characteristics and used multiple regression techniques to make predictions of error, perceived difficulty, and confidence in judgments. 10 figures, 6 tables, and 72 references