Online Program Home
My Program

Abstract Details

Activity Number: 582 - Nonparametric Methods for Statistical Inference
Type: Contributed
Date/Time: Wednesday, July 31, 2019 : 2:00 PM to 3:50 PM
Sponsor: Section on Nonparametric Statistics
Abstract #307343
Title: Non-Parametric Test and Similarity Measure for Matching Bullets
Author(s): Ganesh Krishnan* and Heike Hofmann
Companies: Center for Statistics and Applications in Forensic Evidence (CSAFE) and Iowa State University and Iowa State University
Keywords: forensic science; signatures; cross-correlation; Mann-whitney U statistic; land engraved areas (LEAs); same-source problem

The same-source problem remains a major challenge in forensic toolmark and firearm examination. Here, we investigate the applicability of the Chumbley method(1)(10), developed for screwdriver markings, for same-source identification of striations on bullet Land Engraved Areas (LEAs). The Hamby datasets 44 and 252 measured by NIST and CSAFE (high-resolution scans) are used here. We provide methods to identify parameters that minimize error rates for matching of LEAs, and a remedial algorithm to alleviate the problem of failed tests, while increasing the power of the test and reducing error rates. For 85,491 land-to-land comparisons (84,235 known non-matches and 1256 known matches), the adapted test does not provide a result in 176 situations (originally more than 500). The Type I and Type II error rates are 7.2% (6105 out of 84235) and 21.4% (271 out of 1256) respectively. This puts the proposed method on similar footing as other single feature matching approaches in the literature. We also increase the power of the test and method by combining land-to-land comparisons associated with respective bullets, to get a bullet-to-bullet score.

Authors who are presenting talks have a * after their name.

Back to the full JSM 2019 program