基于改进Mask R-CNN与双目视觉的智能配筋检测
Intelligent rebar inspection based on improved Mask R-CNN and stereo vision
魏翠婷 1赵唯坚 2孙博超 2刘芸怡1
作者信息
- 1. 浙江大学建筑工程学院,浙江杭州 310058
- 2. 浙江大学建筑工程学院,浙江杭州 310058;浙江大学平衡建筑研究中心,浙江杭州 310028
- 折叠
摘要
为了提高配筋检测的智能化水平,提出基于改进掩膜区域卷积神经网络(Mask R-CNN)模型与双目视觉技术的配筋检测方法.通过在Mask R-CNN中加入自下而上的注意力机制路径,形成了带通道注意力和空间注意力的掩膜区域卷积神经网络(Mask R-CNN+CA-SA)改进模型.结合双目视觉技术进行坐标转换,获取钢筋直径与间距,实现智能配筋检测.在自建的包含 3 450 张钢筋图片的数据集上进行训练,结果表明,改进模型的F1 分数和全类平均精确率(mAP)相较于Mask R-CNN基础网络分别提高了 2.54%和 2.47%.通过钢筋网验证试验和复杂背景测试,钢筋直径的绝对误差和相对误差基本小于 1.7 mm和 10%,钢筋间距的绝对误差和相对误差分别小于 4 mm和 3.2%,所提方法在实际应用中具有较强的可操作性.智能配筋检测技术在保证足够的检测精度的同时,能够大大提升工效,降低人工成本.
Abstract
A rebar inspection method based on improved mask region with convolutional neural network(Mask R-CNN)model and stereo vision technology was proposed in order to promote the transformation of reinforcement inspection to intelligence.The improved model Mask R-CNN with channel attention and spatial attention(Mask R-CNN+CA-SA)was formed by adding a bottom-up path with attention mechanism in Mask R-CNN.The diameter and spacing of rebar can be obtained by combining stereo vision technology for coordinate transformation,thereby achieving intelligent rebar inspection.The training was conducted on a self-built dataset containing 3450 rebar pictures.Results showed that the Mask R-CNN+CA-SA model increased the F1 score and mean average precision(mAP)by 2.54%and 2.47%compared with the basic network of Mask R-CNN,respectively.The rebar mesh verification test and complex background test showed that the absolute error and relative error of rebar diameter were basically controlled within 1.7 mm and 10%,and the absolute error and relative error of rebar spacing were controlled within 4 mm and 3.2%respectively.The proposed method is highly operable in practical applications.The intelligent rebar inspection technology can greatly improve work efficiency and reduce labor costs while ensuring sufficient inspection accuracy.
关键词
配筋质量检测/Mask/R-CNN/注意力机制/深度学习/双目视觉技术Key words
rebar quality inspection/Mask R-CNN/attention mechanism/deep learning/stereo vision techno-logy引用本文复制引用
基金项目
国家自然科学基金(52108254)
国家自然科学基金(52208215)
出版年
2024
NETL
NSTL
维普
万方数据