Thursday, September 15, 2016

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



Authors: Jennifer Mnookin, Philip J. Kellman, Itiel Dror, Gennady Erlikhman, Patrick Garrigan, Tandra Ghose, Everett Metler, Dave Charlton

Abstract:
The comparison of forensic fingerprint images for purposes of identification is a complex task that, despite advances in image processing, still requires highly trained human examiners to achieve adequate levels of performance.

This project aims to determine more 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 degree, novices.

While it is likely that well-trained, experienced examiners are highly accurate when making positive identifications, it is also clear that errors still occur.

These experiments showed that experts have substantial, albeit imperfect, subjective knowledge about the difficulty of print pairs. Experiments also showed that novices perform very poorly and showed no consistent pattern of feature use.

Results indicate the plausibility of using objective fingerprint image metrics to predict expert performance and subjective assessment of difficulty in fingerprint comparisons.
While further research is necessary, this research provides strong support for the plausible but previously untested assumption that for expert fingerprint analysis, difficulty is in significant part a function of measurable, visual dimensions of print comparison pairs.

At present, there are not validated, objective metrics for any meaningful step in the comparison process, from the determination of whether there is sufficient information to warrant a comparison to the final judgment of match or non-match.

Overall, a more sophisticated understanding of the relationship between error rate and difficulty should also be extremely important for the courts in weighing fingerprint evidence.

Further research is necessary to build on these results, but this research provides significant steps forwards for helping to establish that error rates are related to difficulty; for beginning to provide validated evidence for what visual dimensions of fingerprint comparison pairs are associated with difficulty; and for helping to tease out both examiner’s metacognitive abilities and the substantial degree of examiner expertise in this domain.

No comments: