TY - GEN
T1 - Engineering Drawing Manager
T2 - 15th International Conference on Intelligent Robotics and Applications, ICIRA 2022
AU - Yang, Honglong
AU - Du, Yang
AU - Guo, Jingwei
AU - Wei, Shiyi
AU - Ma, Hongbin
N1 - Publisher Copyright:
© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.
PY - 2022
Y1 - 2022
N2 - Engineering drawings play an important role in fields such as architecture, industrial engineering, and electric engineering, within which tables contain essential data and structures. However, most engineering drawings exist in the form of scanned PDFs or images, which is inconvenient for data management and storage, especially for table information. Also, many industries are in urgent need of data management software for engineering drawings to improve the degree of digital preservation and management. To this end, a software, ED Manager which is based on the fusion of deep learning and traditional image processing, is presented to detect the position and structure of the table, split and recognize characters, and reconstruct the table in a digital form. Further, we extract crucial information and develop a user interface and database to construct a comprehensive model that fits most engineering drawings. Our software can accurately locate tables for various complex drawings, extract structured information from tables, and build a better data management software for engineering drawings.
AB - Engineering drawings play an important role in fields such as architecture, industrial engineering, and electric engineering, within which tables contain essential data and structures. However, most engineering drawings exist in the form of scanned PDFs or images, which is inconvenient for data management and storage, especially for table information. Also, many industries are in urgent need of data management software for engineering drawings to improve the degree of digital preservation and management. To this end, a software, ED Manager which is based on the fusion of deep learning and traditional image processing, is presented to detect the position and structure of the table, split and recognize characters, and reconstruct the table in a digital form. Further, we extract crucial information and develop a user interface and database to construct a comprehensive model that fits most engineering drawings. Our software can accurately locate tables for various complex drawings, extract structured information from tables, and build a better data management software for engineering drawings.
KW - Engineering drawings
KW - Optimal character recognition
KW - Table reconstruction
KW - Table structure recognition
UR - http://www.scopus.com/inward/record.url?scp=85136930619&partnerID=8YFLogxK
U2 - 10.1007/978-3-031-13841-6_22
DO - 10.1007/978-3-031-13841-6_22
M3 - Conference contribution
AN - SCOPUS:85136930619
SN - 9783031138409
T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
SP - 230
EP - 241
BT - Intelligent Robotics and Applications - 15th International Conference, ICIRA 2022, Proceedings
A2 - Liu, Honghai
A2 - Ren, Weihong
A2 - Yin, Zhouping
A2 - Liu, Lianqing
A2 - Jiang, Li
A2 - Gu, Guoying
A2 - Wu, Xinyu
PB - Springer Science and Business Media Deutschland GmbH
Y2 - 1 August 2022 through 3 August 2022
ER -