同济大学学报（自然科学版）2026，Vol.54Issue(9)：1355-1365,11.DOI:10.11908/j.issn.0253-374x.25207
建筑图纸到钢筋下料单的智能生成技术
Intelligent Generation Technology of Rebar Cutting Lists from Architectural Drawings Based on Deep Learning
庄健琨 1杨彬 1胡晋铭1
作者信息
- 1. 同济大学 土木工程学院,上海 200092
- 折叠
摘要
Abstract
This paper proposes a deep learning-based method for intelligent generation of cutting lists from architectural drawings.The approach integrates object detection and Transformer networks to achieve accurate table region localization and efficient information classification.Experiments were conducted on a dataset of 200 architectural drawings,which was divided into training,validation,and test sets in a ratio of 7∶2∶1.The results show that the proposed system achieves a table data extraction accuracy of 99.8%.In practical applications,it reduces cutting list preparation time by 99.0%and lowers the material waste rate to 7.43%,offering an effective technical solution for intelligent construction management.关键词
建筑图纸/表格识别/深度学习/钢筋下料Key words
architectural drawings/table recognition/deep learning/rebar cutting lists分类
建筑与水利引用本文复制引用
庄健琨,杨彬,胡晋铭..建筑图纸到钢筋下料单的智能生成技术[J].同济大学学报（自然科学版）,2026,54(9):1355-1365,11.
