,
Mamun University Uzbekistan
,
Jizzakh State Pedagogical University , Jizzakh , Uzbekistan
,
Tashkent State University of Oriental Studies , Tashkent , Uzbekistan
,
The Institute of History of The Academy of Sciences of The Republic of Uzbekistan , Shahrisabz , Uzbekistan
,
Tashkent State University of Oriental Studies , Tashkent , Uzbekistan
,
Tashkent University of Information Technology , Tashkent , Uzbekistan
,
Namangan State Pedagogical Institute Uzbekistan
Uzbek National Pedagogical University Uzbekistan
Preserving and interpreting historical engineering documents aids in appreciating the nature of scientific reasoning as well as technological advances. Digitization and detailed analysis are highly challenging for many of such documents which are handwritten, eroded and in delicate conditions. This paper presents research on reconstruction of historical engineering documents by employing Optical Character Recognition (OCR) techniques along with Artificial Intelligence (AI) driven machine learning and natural language processing (NLP). With deep learning-based OCR models trained on historical scripts pertaining to specific fields, complex texts, diagrams, and advanced engineering annotations can be extracted, deciphered, and reconstructed accurately. Besides, the engineering text recognition models built in this work utilize contextual understanding that requires the structure of embedded text and documents to improve the accuracy of recognition and creation of engineering document metadata improving retrieval within the archives of documents. Critical analysis of engineering documents of the 18th and 19th centuries demonstrates marked growth in both efficiency and accuracy of transcription as well as speed of processing over conventional OCR techniques. With this research, cultural heritage can be preserved using advanced AI technologies which provide easier access, understanding, and repurposing of ancient information on engineering captured in these documents.
This is an open access article distributed under the Creative Commons Attribution Non-Commercial License (CC BY-NC) License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
The statements, opinions and data contained in the journal are solely those of the individual authors and contributors and not of the publisher and the editor(s). We stay neutral with regard to jurisdictional claims in published maps and institutional affiliations.