Researchers at Italy's Central Bank have introduced a new artificial intelligence tool designed to strengthen quality control in banknote production. The study, published on June 4, focused on a method called "Siamese networks" to distinguish defective banknotes.
This approach is based on two structurally identical subnetworks that evaluate two images together rather than learning from individual samples. Both networks share the same weights and parameters; the system learns comparatively whether the images belong to the same class.
The study aims to identify defects in banknotes passing through the production line more quickly and consistently. This provides a technical framework that can support human oversight during the quality control process.
Quality control in banknote production requires distinguishing not only visible errors but also non-standard printing and surface defects. The method used by the researchers stands out as a model aimed at automating this differentiation.
AI-based systems like this are seen as tools that can reduce the inspection burden, especially in high-volume production processes. However, the framework presented in the study indicates that the method is considered an approach to quality control.