当前位置: X-MOL 学术Autom. Constr. › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
Investigation of steel frame damage based on computer vision and deep learning
Automation in Construction ( IF 10.3 ) Pub Date : 2021-09-09 , DOI: 10.1016/j.autcon.2021.103941
Bubryur Kim 1 , N. Yuvaraj 2 , Hee Won Park 3 , K.R. Sri Preethaa 4 , R. Arun Pandian 5 , Dong-Eun Lee 3, 6
Affiliation  

Visual damage inspection of steel frames by eyes alone is time-consuming and cumbersome; therefore, it produces inconsistent results. Existing computer vision-based methods for inspecting civil structures using deep learning algorithms have not reached full maturity in exactly locating the damage. This paper presents a deep convolutional neural network-based damage locating (DCNN-DL) method that classifies the steel frame images provided as inputs as damaged and undamaged. DenseNet, a DCNN architecture, was trained to classify the damage. The DenseNet output was upscaled and superimposed on the original image to locate the damaged part of the steel frame. The DCNN-DL method was validated using 144 training and 114 validation sets of steel frame images. DenseNet, with an accuracy of 99.3%, outperformed MobileNet and ResNet with accuracies of 96.2% and 95.4%, respectively. This case study confirms that the DCNN-DL method effectively facilitates the real-time inspection and location of steel frame damage.



中文翻译:

基于计算机视觉和深度学习的钢框架损伤调查

仅靠肉眼对钢架进行目视检查既费时又麻烦;因此,它会产生不一致的结果。现有的基于计算机视觉的使用深度学习算法检查土木结构的方法在准确定位损坏方面还没有完全成熟。本文提出了一种基于深度卷积神经网络的损伤定位 (DCNN-DL) 方法,该方法将作为输入提供的钢框架图像分类为损坏和未损坏。DenseNet 是一种 DCNN 架构,经过训练对损伤进行分类。DenseNet 输出被放大并叠加在原始图像上以定位钢架的损坏部分。DCNN-DL 方法使用 144 个训练集和 114 个验证集的钢架图像进行了验证。DenseNet 的准确率为 99.3%,优于 MobileNet 和 ResNet,准确率为 96。分别为 2% 和 95.4%。本案例研究证实,DCNN-DL 方法有效地促进了钢框架损伤的实时检查和定位。

更新日期:2021-09-10
down
wechat
bug