土木工程与管理学报2025，Vol.42Issue(6)：38-46,9.DOI:10.13579/j.cnki.2095-0985.2025.20240515
基于计算机视觉的钢筋交叉点绑扎状态识别方法研究
Research on the Recognition Method of Rebar Crosspoint Binding Status Based on Computer Vision
摘要
Abstract
In the context of the construction industry's transition towards digitalization and intelligent transformation,the application of rebar binding robots in intelligent construction has garnered increas-ing attention.At present,rebar binding robots rely mainly on computer vision technology for detecting and recognizing rebar intersections.Compared with the traditional physical sensor recognition,com-puter vision offers faster recognition speeds,stronger generalization capability,and higher accuracy.This paper reviews existing methods of detecting and recognizing the binding status of rebar intersec-tions using computer vision.Based on different image processing approaches,it details three primary methods:image processing,bag-of-features model,and convolutional neural network.The analysis highlights the advantages and disadvantages of current computer vision-based methods for recognizing the binding status of rebar crosspoints.Considering the differences between construction site environ-ments and laboratory settings,corresponding research suggestions are proposed to address these chal-lenges.关键词
钢筋交叉点检测识别/计算机视觉/图像处理/特征词袋/卷积神经网络Key words
rebar crosspoint detection and recognition/computer version/image process/bag of fea-tures/convolutional neural network分类
信息技术与安全科学引用本文复制引用
周燕,柯美翔,张晨阳,文世峰,周诚..基于计算机视觉的钢筋交叉点绑扎状态识别方法研究[J].土木工程与管理学报,2025,42(6):38-46,9.基金项目
"十四五"国家重点研发计划项目(2023YFC3806900) （2023YFC3806900）
