Are printed fingermarks realistic enough to be implemented in fingermark visualisation training and proficiency...

In recent years, artificial fingermarks have emerged as a promising solution to improve control over fingermark quality and to mitigate the inherent variability introduced by human donors. This is particularly relevant in applications such as proficiency testing and training, where standardization and reproducibility are critical. This study explored the use of inkjet printing to produce artificial fingermarks that visually resemble real marks. Using templates derived from actual latent and inked fingermarks, artificial marks were printed on paper, using varying concentrations of an amino acid mixture, and subsequently developed with either 1,2-indanedione/zinc or ninhydrin. To assess realism of the printed fingermarks, six evaluators were asked to classify a randomized set of 174 fingermarks as "real" or "printed" without prior knowledge of their nature. While most fingermarks were correctly identified (i.e., 74% of real and 73% of printed), a significant portion was misclassified or labelled as ambiguous, especially when artificial fingermarks were derived from lower-quality latent mark models, which tended to mislead evaluators. Unrealistic homogeneity, absence of pore detail, and occurrence of printing artifacts were commonly cited by assessors to identify printed marks. However, these indicators were not always reliable, leading to uncertainty and disagreement. The use of automated quality assessment algorithms confirmed that the quality of the printed fingermarks was comparable to the original marks. These findings demonstrate the feasibility of creating high-quality, realistic artificial fingermarks suitable for blind proficiency testing and forensic education, offering improved control and reproducibility compared to natural fingermarks deposited directly by human donors.