Development of an ai-enabled video capturing device for bullet trajectory analysis and ballistic research
Abstract
A ballistic experts’ discipline is the ability to compare the characteristic marks found on the surface of different fired bullets to determine whether they were fired from the same gun. These tool marks become a “ballistic fingerprint” that examiners can use to identify specific characteristics of the firearm that discharged the bullet. One such tool mark is the striation marks left on the bullet, identical to scratch marks. Manually done, a comparison microscope is used in this process, where the testing bullet is rotated until a well-defined land or groove comes into view. The sample bullet is then rotated in search of a matching region. But in this process opinions are given through only the manual experimental process and not through an automated system. The proposed solution was to develop a cost-effective automated system that captures the video of the bullet in one go. Also, the focus was to develop a lighting arrangement independent of the environment, so that the device can be efficiently used in any environment.
Downloads
References
Crime in India, National Crime Records Bureau. MoHA,” 2019. [Online]. Available: https://ncrb.gov.in/sites/default/files/CII 2019 Volume 1.pdf.
Ministry of Home Affairs, Directorate of Forensic Science Services, Government of India,” 2021. http://www.dfs.nic.in/.
P. Changmai, K. Bora, R. Suresh, N. Deb, and L. B. Mahanta, “On the Study of Automated Identification of Firearms Through Associated Striations,” Dec. 2019, doi: 10.12783/ballistics2019/33156.
B. Barua, G.Chyrmang, K. Bora, R. Suresh, L.B. Mahanta, "2D Image Stitching to Generate a Film
of Fired Bullets", Springer Nature Singapore Pte Ltd. 2021. https://doi.org/10.1007/978-981-33-4084-8_31
Dutta, S., Saikia, S., Barman, A., Roy, R., Bora, K., Mahanta, L. and Suresh, R., 2021, " Study on enhanced deep learning approaches for value-added identification and segmentation of striation marks in bullets for precise firearm classification", Applied Soft Computing, 112, p.107789.
C. L. Smith, “Fireball: a forensic ballistics imaging system,” in Proceedings IEEE 31st Annual 1997 International Carnahan Conference on Security Technology, pp. 64–70, doi: 10.1109/CCST.1997.626240.
P. Pisantanaroj et al., “Automated Firearm Classification From Bullet Markings Using Deep Learning,” IEEE Access, vol. 8, pp. 78236–78251, 2020, doi: 10.1109/ACCESS.2020.2989673.
Copyright (c) 2024 ITEGAM-JETIA
This work is licensed under a Creative Commons Attribution 4.0 International License.