Adapting the Chumbley Score to Match Striae on Land Engraved Areas (LEAs) of Bullets,

Abstract

The same‐source problem remains a major challenge in forensic toolmark and firearm examination. Here, we investigate the applicability of the Chumbley method (J Forensic Sci, 2018, 63, 849; J Forensic Sci, 2010, 55, 953) (10,12), developed for screwdriver markings, for same‐source identification of striations on bullet 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 nonmatches 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 84,235) 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.

Accuracy of comparison decisions by forensic firearms examiners

“…Since the publication of the PCAST report, a number of firearms examiners and independent researchers have conducted additional investigations dealing with various aspects of comparative examinations. These include the estimation of examiner error rates [ 10 , 11 , 12 , 13 , 14 , 15 , 16 , 17 , 18 ], statistical evaluation methods in the Identification of toolmarks [ 19 , 20 ], and efforts to produce either automated or computer‐based objective determinations [ 15 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 29 , 30 , 31 ]. Additionally, several compilations contain general discussions and document research efforts as applied to firearms and toolmark examinations [ 32 , 33 , 34 , 35 , 36 , 37 ].…”

Hierarchical Decision Ensembles- An inferential framework for uncertain Human-AI collaboration in forensic examinations

“…In their work they show a standard procedure of moving from data to model to random forest scores to bullet scores. Krishnan and Hofmann [22] later employed a similar pipeline to come up with a non-parametric score of comparison. The standard steps of this bullet matching pipeline can be seen in the Figure 3.…”

“…Thus, for each of the 6 LEAs a suitable cross-section location is chosen and the marking corresponding to this cross-cut is extracted. We call this extracted marking a 'profile' [22] [18]. Profiles reflect the curvature and other specific traits of an LEA.…”