Forensic Analysis of Digital Dynamic Signatures: New Methods for Data Treatment and Feature Evaluation. Issue 2 (17th November 2016)

• Record Type: Journal Article
• Authors: Linden, Jacques; Marquis, Raymond; Mazzella, Williams
• Is Part Of: Journal of forensic sciences. Volume 62:Issue 2(2017)
• Page Start: 382
• Page End: 391
• Publication Date: 2016-11-17
• Subjects: forensic science -- forensic document examination -- dynamic signature -- data analysis methods -- variation study -- feature discrimination; Medical jurisprudence -- Periodicals; Forensic sciences -- Periodicals; Forensic Medicine -- Periodicals; Gerechtelijke geneeskunde; Gerechtelijke chemie; Gerechtelijke psychiatrie; 363.2505
• DOI: 10.1111/1556-4029.13288
• Languages: English
• ISSNs: 0022-1198

Abstract

Abstract: This study explored digital dynamic signatures containing quantifiable dynamic data. The change in data content and nature necessitates the development of new data treatment approaches. A SignPad Omega digitizing tablet was used to assess measurement reproducibility, as well as within‐writer variation and the occurrence of correctly simulated features. Measurement reproducibility was found to be high except for pressure information. Within‐writer variation was found to be higher between days than on a same day. Occurrence of correct simulation was low for features such as signature size, trajectory length, and total signature time. Feature discrimination factors combining within‐writer variability and the occurrence of correctly simulated features were computed and show that signature size, trajectory length, and signature time are the features that perform the best for discriminating genuine from simulated signatures. A final experiment indicates that dynamic information can be used to create connections between simulation cases.