DEVELOPMENT OF A SIMPLE AND EFFECTIVE HANDWRITING RECOGNITION SYSTEM IN RUSSIAN AND KYRGYZ LANGUAGES
This article examines the current problem of digitization of handwritten data faced by educational and government institutions in Kyrgyzstan. Despite the development of artificial intelligence technologies, existing optical character recognition (OCR) systems often incorrectly process certain letters of the Kyrgyz alphabet (Ө, Ң, Ү) or require expensive licenses. The aim of the work was to develop an accessible and easy-to-use web-based system for automatic handwriting recognition in Russian and Kyrgyz. The proposed solution is based on a machine learning method with a teacher using a database of reference images. The article provides a detailed description of the system's algorithm, from converting handwritten text into a binary matrix (a 10x10 pixel grid) to comparing the resulting image with thousands of saved templates. The software implementation is made in Python using the Django framework, which allows you to use the application through a regular web browser without installing complex software. The main methods and algorithms of the pattern recognition theory were considered, as well as the possibilities of their practical application in the task of recognizing handwritten characters of the Kyrgyz alphabet. In the course of the work, it was found that the effectiveness of recognition does not depend on one single method, but on the correct organization of the entire data processing process.
Сведения
- Опубликовано
- 2026 · Т. 614, № 1005, С. 701-708
- Язык
- Английский
- ISSN
- 1694-7851 · 1694-8505
- Издатель
- Kyrgyz State University named after I. Arabaev
- Аффилиации
- Kyrgyz State University named after I. Arabaev