Description of original award (Fiscal Year 2009, $866,674)
The goal of this project is to develop and validate a quantifiable metric of difficulty, and then to examine error rates as a function of difficulty. In this proposal, the researchers will develop difficulty measures that combine visual complexity measures and cognitive difficulty measures. They will first construct a database of prints and establish ground truth (Phase 1). Phase 2 will use a variety of approaches to develop a metric of visual complexity of the prints. The researchers will then test and validate the metrics by examining cognitive difficulty, using signal detection theory (SDT), human psychophysical methods, and qualitative expert-based difficulty rating measurements (Phase 3). These results combined will assist the researchers to develop a quantifiable and validated assessment tool, allowing a difficulty rating to be associated with any given latent print or comparison.
Finally, using this hierarchy of fingerprint comparison difficulty, the researchers will investigate the error rates, if any, associated with comparisons of varying degrees of difficulty, using ecologically valid studies with qualified and experienced latent print forensic examiners (Phase 4).
- Quantification of Tool Markings
- NIST/Office of Law Enforcement Standards funding for two projects associated with E-Crime technology portfolio, Computer Forensic Tool Testing and National Software Reference Library
- Illuminating the Dark: Molecular Networking as a Novel Psychoactive Substance Identification Strategy