Development of a Dimensional Analysis Approach in Gunshot Residue Images Using Computerized Image Processing

Abstract

Computer image processing is a method that uses artificial intelligence and machine learning-based general learning algorithms. With this method, objects in digital images (photos or videos) can be grouped by being perceived and detected. Computerized image processing method can be applied to almost all kinds of digital data produced with the developing technology. Nowadays, the identification and detection of gunshot residues (GSR) can be done manually by experts from the acquired images. In this study, computerized image processing method was used for the identification and dimensional analysis of gunshot residues (GSR). In this new proposed method, a dataset of 18500 digital image samples obtained from three different caliber cartridges (MKE, Gecco and S&B brands) was used. From the results of the study, it has been shown that the Computer Vision Method is a successful method in the automatic dimensional classification of GSRs.