EUVIMEDEUVIMED entdecken
Crossref

ALGORITHM AND STAGES OF TRAINING A TEXT RECOGNITION MODEL FOR THE TURKMEN LANGUAGE USING TESSERACT OCR

Maksud Sharipov, Khikmat J. Rakhimboev, Ikhtiyor Avezmatov

Scientific works/Elmi eserler · 22.07.2026 · 512

DOI: 10.58225/mpmma.2026.512 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Scientific works/Elmi eserler, online erschienen am 22.06.2026. Verfasst von Maksud Sharipov und 2 weiteren Autor:innen. Die Arbeit nennt 40 Literaturangaben.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Enhancing the accuracy of Optical Character Recognition (OCR) models for languages with specific characters, such as the Turkmen language, is of great importance within the field of digital text processing. This paper proposes a complete algorithm and the practical stages for developing and fine-tuning a high-accuracy text recognition model for the Turkmen language using the Tesseract OCR engine. During the training process, the syntactic and morphological rules (langdata, unicharset) of the Turkmen language were formalized on the basis of the related Turkish language, and 6078 clean text–image pairs were prepared and trained within the Tesseract (tesstrain) architecture. The research findings demonstrate that the new model, trained using the Transfer Learning approach on LSTM neural networks, can be effectively applied with high accuracy (97.8%) and a low Character Error Rate (CER) in document digitization, automated translation, and Natural Language Processing (NLP) systems for the Turkmen language. Keywords: Turkmen language · Optical Character Recognition (OCR) · Tesseract OCR · Model fine-tuning · LSTM neural networks · Transfer Learning · Natural Language Processing (NLP).

Zitieren

APA

Sharipov, M., Rakhimboev, K. J., & Avezmatov, I. (2026). ALGORITHM AND STAGES OF TRAINING A TEXT RECOGNITION MODEL FOR THE TURKMEN LANGUAGE USING TESSERACT OCR. Scientific works/Elmi eserler, 512. https://doi.org/10.58225/mpmma.2026.512

Vancouver

Sharipov M, Rakhimboev KJ, Avezmatov I. ALGORITHM AND STAGES OF TRAINING A TEXT RECOGNITION MODEL FOR THE TURKMEN LANGUAGE USING TESSERACT OCR. Scientific works/Elmi eserler. 2026:512. doi:10.58225/mpmma.2026.512

BibTeX
@article{Sharipov2026,
  author = {Sharipov, Maksud and Rakhimboev, Khikmat J. and Avezmatov, Ikhtiyor},
  title = {ALGORITHM AND STAGES OF TRAINING A TEXT RECOGNITION MODEL FOR THE TURKMEN LANGUAGE USING TESSERACT OCR},
  journal = {Scientific works/Elmi eserler},
  year = {2026},
  pages = {512},
  doi = {10.58225/mpmma.2026.512}
}
Kostenlos starten →