Development of Synthetically Generated LEA Signatures to Generalize Probability of False Positive Identification Estimates

The purpose of this study is to develop computer models that will be able to quantify the probability that two bullets could originate from two different barrels associated with the same type of firearm. In other words, how likely is it that this bullet was fired out of this 9mm Glock rather than that 9mm Glock. In so doing, this study will introduce a rigor to ballistic analysis that has been heretofore lacking in this forensic discipline. It is well understood that no two gun barrels are alike. This is due to both minute imperfections resulting from the manufacturing process and wear over time as bullets are fired through them. Consequently, the striations imprinted on a bullet by the lands and grooves in a given barrel are unique. But no one has heretofore quantified how unique. This effort will do that, by developing computer models that will enable the virtual firing of thousands of rounds through the barrels associated with specific makes of guns, enabling the development of the statistically significant data needed to be able to provide that quantification.

ca/ncf