There are no reviews yet. Be the first to send feedback to the community and the maintainers!
Repository Details
The aim of project is the development of an automated text recognition system with the support of the Ukrainian language, which focuses on the pre-processing of images and processing of recognized text. Two models of machine learning have been developed: to determine the angle at which the image is rotated and to determine the type of document. Algorithms for image enhancement (binarization, noise removal, contrast adjustment) and an algorithm for correcting errors in text using fuzzy string logic and Levenshtein distance have also been developed. The system is presented as a desktop application. The server part was developed using the C# programming language and ASP.NET Core, ML.NET frameworks, client part - WPF (Windows Presentation Foundation), text recognition - Tesseract OCR.