Revisiting overlapped fingerprint separation methods and their key requirements for forensic systems

At crime scenes, overlapped fingerprints are frequently found on commonly used surfaces, requiring separation methods for reliable person identification with forensic systems. While various methods for separating overlapped fingerprints have been developed in recent years, their practical usefulness has not been adequately evaluated due to a lack of practical requirements. This paper presents a comprehensive overview of overlapped fingerprint separation methods, based on a meta-study analyzing 33 methods published between 2010 and 2024 (i.e., 15 years). Additionally, we identified ten key requirements to strengthen the transfer of scientific methods to forensic practice. Our analysis shows a significant trend towards hybrid and AI-based approaches for fingerprint separation, offering both new opportunities (e.g., efficiency, accuracy) and challenges (e.g., reliability, explainability). Unfortunately, we found that current separation methods cannot yet be satisfactorily evaluated in terms of their practical usability for forensic systems. This limitation results from the fact that most existing studies only address some of the identified requirements or only to a certain extent, which significantly hinders their practical implementation. The study emphasizes the need for a more practice-oriented approach in evaluating fingerprint separation methods. To enhance their real-world applicability, we strongly encourage researchers to address the ten identified requirements, contributing to a facilitated transfer of theoretical methods into real-world forensic systems. • We present a meta-study reviewing 33 overlapped fingerprint separation methods (2010–2024). • We define ten key requirements for evaluating fingerprint separation in forensic systems. • We analyze to what extent existing methods address these requirements. • We identify research trends from traditional to hybrid and AI-based separation methods. • We highlight shortcomings and propose directions for practice-oriented evaluations.