WRITE - IT assisted search and comparison of handwritings
IT based solution which allows for the search of similar handwritings to expedite the identification of unknown writers by handwriting experts.
Law enforcement agencies possess an extensive collection of handwritten documents.
This includes for example documents belonging to open cases and reference samples from suspects and prisoners. Yet, these collections of documents can only be utilized to a limited extend, since for an identification of an unknown writer all documents have to be compared manually by handwriting experts.
WRITE aims to solve this problem by developing an IT based solution which allows for the search of similar handwritings. In this way, the identification of unknown writers by handwriting experts can be expedited, since only a small number of documents with similar handwritings have to be compared manually. In contrast to currently available commercial solutions, with the proposed system no manual classification of the handwritings needs to be performed beforehand. This is on the one hand time consuming and on the other hand error prone, since it depends on the subjective opinion of the individual expert. Ideally, it is only required to digitize the handwritten documents using an image scanner to utilize the proposed system. This also provides an effortless way to utilize already existing databases with handwritten documents and, as a consequence, these databases would gain a significant preventive character. The scientific-technical goal of WRITE is the development of a retrieval system based on artificial intelligence/machine learning that is accurate enough to list documents from the same writer in the top 2% of the retrieved sorted list in 80% of the cases. The Referat Urkunden- & Handschriftenuntersuchung im Büro für Kriminaltechnik des Bundeskriminalamts in the Federal Ministry of the Interior (BMI) possesses a comprehensive collection of handwritten documents essential for such a development. These data will also provide the basis for the research on the individuality and variability of handwriting that will be conducted by the GSK partner, which will in turn support the development of the methodology.
Univ.Prof. DI Dr. Robert Sablatnig
TU Wien, Computer Vision Lab, Institute of Visual Computing & Human-Centered Technology
A-1040 Vienna, Austria
Bundesministerium für Inneres, Bundeskriminalamt, Abt 6, Büro für Kriminaltechnik
cogvis software und consulting GmbH
Elisabeth Charkow, Dipl. Graphologin ÖGS und Schriftsachverständige