Scientific and technical conference “System for processing and digitization of incoming documents”

On March 7, 2024, the conference “System for the processing and digitization of incoming documents” was held in the Faculty of Electrical Engineering and Computer Sciences  Council Room, jointly organized by the Lublin company Infinite and the Department of Computer Science of the Lublin University of Technology. The conference was held in a mixed mode: in person in the Faculty Council Room and remotely using MS Teams. Over 60 participants took part in the conference.

The conference was devoted to the results of the joint project Infinite and the Department of Computer Science of the Lublin University of Technology. The conference was opened by Prof. Tomasz Zientarski – head of the Department of Computer Science.

Representatives of Infinite and the Department of Computer Science presentations performed. Representatives of Infinite presented the entire project and its results, which are currently being commercialized.

In particular, Prof. Marek Miłosz (project manager on behalf of the Lublin University of Technology) presented the scientific activities of the Department of Computer Science team and its achievements in the form of scientific publications:

  • Recognition of handwritten Latin characters with diacritics using CNN / Edyta Łukasik, Małgorzata Charytanowicz, Marek Miłosz, Mikhail Tokovarov, Monika Kaczorowska, Dariusz Czerwiński and Tomasz Zientarski // Bulletin of the Polish Academy of Sciences Technical Sciences.- 2021, vol. 69, nr 1, pp. 1-12 [MEiN: 100]
  • Development of Extensive Polish Handwritten Characters Database for Text Recognition Research / Mikhail Tokovarov, Monika Kaczorowska, Marek Miłosz // Advances in Science and Technology Research Journal.- 2020, vol. 14, nr 3, pp. 30-38 [MEiN: 100]

and an extensive data set of images of 89 characters (uppercase and lowercase letters, numbers and special characters) handwritten from over 2,000 people and processed (normalized to a 32×32 px square) as part of the project. The database containing an average of 6,000 examples of classified sign images (over 530,000 images in total) is publicly available at: https://cs.pollub.pl/phcd/. This is probably the largest dataset of images of Polish handwritten signs. This dataset also contains an image of a fragment of the Invocation from Pan Tadeusz, handwritten by over 2,000 people of various professions and ages.

 

Conference organizing committee:

Marek Miłosz – chairman

Paweł Grzywna – vice-chairman

Małgorzata Charytanowicz

Edyta Łukasik

Elżbieta Miłosz

Mariusz Olejnik

Bogusław Oleksiejuk

Marek Radomski

Tomasz Zientarski