Likelihood Ratios for Deep Neural Networks in Face Comparison. Issue 4 (12th May 2020)

• Record Type: Journal Article
• Authors: Macarulla Rodriguez, Andrea; Geradts, Zeno; Worring, Marcel
• Is Part Of: Journal of forensic sciences. Volume 65:Issue 4(2020)
• Page Start: 1169
• Page End: 1183
• Publication Date: 2020-05-12
• Subjects: digital forensic science -- face recognition -- face verification likelihood ratio -- deep learning -- ENFSI proficiency test; Medical jurisprudence -- Periodicals; Forensic sciences -- Periodicals; Forensic Medicine -- Periodicals; Gerechtelijke geneeskunde; Gerechtelijke chemie; Gerechtelijke psychiatrie; 363.2505
• DOI: 10.1111/1556-4029.14324
• Languages: English
• ISSNs: 0022-1198

Abstract

Abstract: In this study, we aim to compare the performance of systems and forensic facial comparison experts in terms of likelihood ratio computation to assess the potential of the machine to support the human expert in the courtroom. In forensics, transparency in the methods is essential. Consequently, state‐of‐the‐art free software was preferred over commercial software. Three different open‐source automated systems chosen for their availability and clarity were as follows: OpenFace, SeetaFace, and FaceNet; all three based on convolutional neural networks that return a distance (OpenFace, FaceNet) or similarity (SeetaFace). The returned distance or similarity is converted to a likelihood ratio using three different distribution fits: parametric fit Weibull distribution, nonparametric fit kernel density estimation, and isotonic regression with pool adjacent violators algorithm. The results show that with low‐quality frontal images, automated systems have better performance to detect nonmatches than investigators: 100% of precision and specificity in confusion matrix against 89% and 86% obtained by investigators, but with good quality images forensic experts have better results. The rank correlation between investigators and software is around 80%. We conclude that the software can assist in reporting officers as it can do faster and more reliable comparisons with full‐frontal images, which can help the forensic expert in casework.