WonJun Choi, JeSeol Yu, Journal of Internet Computing and Services, Vol. 26, No. 2, pp. 51-60, Apr. 2025
10.7472/jksii.2025.26.2.51, Full Text:  HTML
Keywords: forensic, scientific ciriminal investigation, activity level, Fingerprint, Deep Learning

Abstract

This study aimed to explore the possibility of obtaining activity level information from fingerprints and interpreting it. Activity level information refers to details about what occurred at the scene when the evidence was left behind, including how and what objects interacted to create the evidence. In contrast, source level information pertains to identifying who the evidence originated from. Currently, the forensic science field in South Korea recognizes the importance of interpreting the meaning of evidence, such as analyzing activity level information to maximize its value. However, in practice, the focus remains primarily on source level information, with insufficient education and research on activity level information. In line with the purpose of this study, deep learning was utilized to determine whether activity level information could be extracted from fingerprints. Approximately 70% of fingerprints tested were successful in the experiment. Based on these results, we examined the extent of success achieved by the successful fingerprints and analyzed the outcomes of the failed tests.

Statistics
Cite this article