Improving the Understanding and the Reliability of the Concept of “Sufficiency” in Friction Ridge Examination
Cedric Neumann, Christophe Champod, Mina Yoo, Thibault Genessay, Glenn Langenburg
See also the instructional videos on "Fingerprint Comparison"
Executive Summary
This document reports on a 2 year research project sponsored by the National Institute of Justice of the Department of Justice of the United States of America under contract 2010-DN-BX-K267. The aim of the project was to study the concept of sufficiency associated with the decisions made by latent print examiners at the end of the various phases of the examination process. During this 2 years effort, a web-based interface was designed to capture the observations made by 146 latent print examiners and latent print trainees on a set of 15 pairs of latent/control prints. The variables of interest ranged from demographics data on the participants through to the type of features, their quality, their level of agreement between the latent and control prints, and their decisions at the end of each phase of the examination process. A statistical model was also developed to quantify the specificity of the configurations of minutiae annotated by the participants on the prints. Random Forest classifiers were used to measure the importance of the different variables on the decisions made by the participants. Random Forest classifiers were used as rational proxies of the decision-making process of human examiners based on the observations of the latent/control prints.
Two main findings resulted from our study:
- The concept of sufficiency is mainly driven by the number and spatial relationships between the minutiae observed on the latent and control prints.Our data indicate that demographics (training, certification, years of experience) or non-minutiae based features (such as level 3 features) do not play a major role in the making of decisions by examiners;
- Our results show a significant variability between the detection and interpretation of friction ridge features. This has been observed at all levels of details, as well as for factors potentially influencing the examination process, such as degradation, distortion, or influence of the background and the development technique. There is an urgent need for development of standards and training to ensure consistency in the definition, selection, interpretation and use of observations made on friction ridge impressions.