Defense Date
2025
Document Type
Directed Research Project
First Advisor
Stephanie Walcott
Abstract
Shotgun pellet pattern analysis plays a critical role in forensic firearm investigations, particularly for estimating muzzle-to-target distances. However, conventional methods based on pattern diameter become unreliable when only partial patterns are recovered, as is often the case in forensic scenarios involving soft tissue or incomplete impact surfaces. This study aimed to validate ImageJ, an open-source image analysis software, as a standardized, accurate, and reproducible tool for measuring pellet-to-pellet distances in partial shotgun patterns. The methodology included calibration of ImageJ using a perforated poly board, comparison of manual versus digital measurements on 17 pre-made pellet patterns, and development of a calibration curve for distance estimation using 340 measurements from 17 test patterns at known distances. Two ImageJ measurement methods, the line tool and multipoint coordinate analysis were tested. Statistical comparison via paired t-tests showed no significant difference between manual and ImageJ measurements at identical points (p = 0.59), with a high Pearson correlation (r = 0.997) and a mean absolute error of 0.0605 inches. Intra-observer reliability testing revealed a small but statistically significant difference between repeated measurements from different points within the same pattern (p = 0.041), though results remained highly correlated (r = 0.979). A linear regression model generated from known patterns produced a calibration equation (y = 0.16x + 0.1398; R² = 0.9581), which was then successfully applied to estimate distances in three unknown test patterns. These findings support the use of ImageJ as a valid tool for the quantitative analysis of shotgun pellet patterns, particularly when complete dispersion data is unavailable. This methodology provides a scientifically rigorous and reproducible alternative for estimating firing distances in cases involving partial pattern recovery, thereby enhancing the objectivity and evidentiary strength of forensic firearm analysis.
Rights
© The Author(s)
Is Part Of
VCU Master of Science in Forensic Science Directed Research Projects
Date of Submission
5-12-2025