An algorithm for ink analysis based on machine learning has been created at St. Petersburg State University (SPbSU). The innovative technology will effectively and inexpensively identify document forgeries, according to the university's press service.
Chemists from the Center for Artificial Intelligence and Data Science at St. Petersburg State University, together with colleagues from Germany and Denmark, have introduced machine learning methods into ink analysis. The unique DCA ML method they developed provides high accuracy in detecting forgeries.
Ink analysis in forensics is a complex process that is crucial for verifying the authenticity of documents. Existing research methods have their drawbacks: high cost and risk of sample damage. One of the most promising areas is digital color analysis (DCA), which is actively being developed by scientists at St. Petersburg State University.
Digital color analysis is an innovative approach that uses images and scans of documents to study ink without physically affecting the original. Scientists have improved this method by combining it with machine learning, which has increased its accuracy.
The new approach has many advantages. It has a minimal impact on the structure of the original document and does not require the use of aggressive chemicals. In addition, the method can be adapted to analyze antique papers, signatures, and works of art.
Read more on the topic:
AI operating on the principle of the brain, created in NNSU
Universal machine vision system created by Russian scientists
AI-based medical assistant created by specialists from the Institute of Artificial Intelligence
