Optimizing Text Recognition in Mechanical Drawings: A Comprehensive Approach
Machines · 20.03.2025 · 13 (3) · 254
DOI: 10.3390/machines13030254 ↗Kurz gefasst
Fachartikel in der Zeitschrift Machines, online erschienen am 20.03.2025. Verfasst von Javier Villena Toro und 1 weiteren Autor:innen. Die Arbeit wurde bisher 18-mal zitiert (Crossref) und nennt 30 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Vinnova, iPROD project, DART project.
Abstract (Originalsprache)
The digitalization of engineering drawings is a pivotal step toward automating and improving the efficiency of product design and manufacturing systems (PDMSs). This study presents eDOCr2, a framework that combines traditional OCR and image processing to extract structured information from mechanical drawings. It segments drawings into key elements—such as information blocks, dimensions, and feature control frames—achieving a text recall of 93.75% and a character error rate (CER) below 1% in a benchmark with drawings from different sources. To improve semantic understanding and reasoning, eDOCr2 integrates Vision Language models (Qwen2-VL-7B and GPT-4o) after segmentation to verify, filter, or retrieve information. This integration enables PDMS applications such as automated design validation, quality control, or manufacturing assessment. The code is available on Github.
Förderung
- Vinnova
- iPROD project
- DART project
Zitieren
APA
Toro, J. V., & Tarkian, M. (2025). Optimizing Text Recognition in Mechanical Drawings: A Comprehensive Approach. Machines, 13(3), 254. https://doi.org/10.3390/machines13030254
Vancouver
Toro JV, Tarkian M. Optimizing Text Recognition in Mechanical Drawings: A Comprehensive Approach. Machines. 2025;13(3):254. doi:10.3390/machines13030254
BibTeX
@article{Toro2025,
author = {Toro, Javier Villena and Tarkian, Mehdi},
title = {Optimizing Text Recognition in Mechanical Drawings: A Comprehensive Approach},
journal = {Machines},
year = {2025},
volume = {13},
number = {3},
pages = {254},
doi = {10.3390/machines13030254}
}
EUVIMED