Engineering deep learning methods on automatic detection of damage in infrastructure due to extreme events

Abstract

This paper presents a few comprehensive experimental studies for automated Structural Damage Detection (SDD) in extreme events using deep learning methods for processing 2D images. In the first study, a 152-layer Residual network (ResNet) is utilized to classify multiple classes in eight SDD tasks, which include identification of scene levels, damage levels, and material types. The proposed ResNet achieved high accuracy for each task while the positions of the damage are not identifiable. In the second study, the existing ResNet and a segmentation network (U-Net) are combined into a new pipeline, cascaded networks, for categorizing and locating structural damage. The results show that the accuracy of damage detection is significantly improved compared to only using a segmentation network. In the third and fourth studies, end-to-end networks are developed and tested as a new solution to directly detect cracks and spalling in the image collections of recent large earthquakes. One of the proposed networks can achieve an accuracy above 67 .6% for all tested images at various scales and resolutions, and shows its robustness for these human-free detection tasks. As a preliminary field study, we applied the proposed method to detect damage in a concrete structure that was tested to study its progressive collapse performance. The experiments indicate that these solutions for automatic detection of structural damage using deep learning methods are feasible and promising. The training datasets and codes will be made available for the public upon the publication of this paper.

Artificial intelligence volleyball special test system based on deep learning and 3D imaging technology

“…The feature layer is from low to high, and its receptive field is from small to large. Different feature layers are helpful for detecting objects of different sizes 7 .…”

A novel transformer-based semantic segmentation framework for structural condition assessment

“…Rubio et al 24 evaluated FCNs for damage segmentation on a database of bridges in Niigata Prefecture. Besides, other CNNbased architectures, such as SegNet 25 and U-Net, 26 have also demonstrated great advances [27][28][29][30] in visionbased SHM. Narazaki et al 31 developed a vision-based automated bridge component recognition framework by exploring FCNs and SegNet.…”

Deep neural network-based structural health monitoring technique for real-time crack detection and localization using strain gauge sensors

“…Therefore, structural health monitoring (SHM) techniques have been developed to monitor the extraction of damage-sensitive features such as strain, acoustic emission (AE), vibration signals, and electromechanical impedance, recorded using various sensors installed on the structure and determine the current state of structural health through statistical analysis of the extracted features 1 – 6 . Recently, with the development of high-performance graphics processing units and parallel computing, convolutional neural networks (CNN)-based approach using the 2D images for detecting damage have been proposed 7 – 10 .…”