SDFormer: A Novel Transformer Neural Network for Structural Damage Identification by Segmenting the Strain Field Map

Li, Zhaoyang; Xu, Ping; Xing, Jie; Yang, Chengxing

doi:10.3390/s22062358

Cited by 7 publications

(4 citation statements)

References 33 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Developments in machine learning, such as convolutional and transformer neural networks [244], will support autonomous analysis and decision-making both for robots and for SHM. Exploration of how this dense information can be optimally combined and leveraged will require further integration of robotic sensing platforms and artificial intelligence with building information modeling [245].…”

Section: Advances In Science and Technology To Meet Challengesmentioning

confidence: 99%

Roadmap on measurement technologies for next generation structural health monitoring systems

Laflamme

Ubertini

Matteo

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

Structural Health Monitoring (SHM) is the automation of the condition assessment process of an engineered system. When applied to geometrically large components or structures, such as those found in civil and aerospace infrastructure and systems, a critical challenge is in designing the sensing solution that could yield actionable information. This is a difficult task to conduct cost-effectively, because of the large surfaces under consideration and the localized nature of typical defects and damages. There has been significant research efforts in empowering conventional measurement technologies for applications to SHM in order to improve performance of the condition assessment process. Yet, the field implementation of these SHM solutions is still in its infancy, attributable to various economic and technical challenges. The objective of this Roadmap publication is to discuss modern measurement technologies that were developed for SHM purposes, along with their associated challenges and opportunities, and to provide a path to research and development efforts that could yield impactful field applications.

show abstract

Section: Advances In Science and Technology To Meet Challengesmentioning

confidence: 99%

Roadmap on measurement technologies for next generation structural health monitoring systems

Laflamme

Ubertini

Matteo

et al. 2023

Meas. Sci. Technol.

View full text Add to dashboard Cite

show abstract

“…There exist many related methods [27], [28], [29], [30], [31], [32], [44] for designing spatiotemporal self-attention.…”

Section: ) Joint Spatiotemporal Self-attentionmentioning

confidence: 99%

Campus Abnormal Behavior Recognition With Temporal Segment Transformers

et al. 2023

View full text Add to dashboard Cite

The intelligent campus surveillance system is beneficial to improve safety in school. Abnormal behavior recognition, a field of action recognition in computer vision, plays an essential role in intelligent surveillance systems. Computer vision has been actively applied to action recognition systems based on Convolutional Neural Networks (CNNs). However, capturing sufficient motion sequence features from videos remains a significant challenge in action recognition. This work explores the challenges of video-based abnormal behavior recognition on campus. In addition, a novel framework is established on long-range temporal video structure modeling and a global sparse uniform sampling strategy that divides a video into three segments of identical durations and uniformly samples each snippet. The proposed method incorporates a consensus of three temporal segment transformers (TST) that globally connects patches and computes self-attention with joint spatiotemporal factorization. The proposed model is developed on the newly created campus abnormal behavior recognition (CABR50) dataset, which contains 50 human abnormal action classes with an average of over 700 clips per class. Experiments show that it is feasible to implement abnormal behavior recognition on campus and that the proposed method is competitive with other peer video recognition in terms of Top-1 and Top-5 recognition accuracy. The results suggest that TST-L+ can improve campus abnormal behavior recognition, corresponding to Top-1 and Top-5 accuracy results of 83.57% and 97.16%, respectively.

show abstract

“…With the advancements in techniques to measure full-field strain maps such as S 4 [32][33][34][35] and Digital Image Correlation [36], SSD detection methods based on strain measurement can be a viable option. There are a few studies on damage detection and localization using full-field strain information [37,38]. However, ref.…”

Section: Introductionmentioning

confidence: 99%

“…However, ref. [37] focuses on the surface damage of metallic structures, and ref. [38] uses strain mode shapes for localization on a global level which cannot provide a precise localization at the local level.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning-Based Subsurface Damage Localization Using Full-Field Surface Strains

Pal,

Meng,

Nagarajaiah

2023

Sensors

View full text Add to dashboard Cite

Structures in their service life are often damaged as a result of aging or extreme events such as earthquakes or storms. It is essential to detect damage in a timely fashion to ensure the safe operation of the structure. If left unchecked, subsurface damage (SSD) can cause significant internal damage and may result in premature structural failure. In this study, a Convolutional Neural Network (CNN) has been developed for SSD detection using surface strain measurements. The adopted network architecture is capable of pixel-level image segmentation, that is, it classifies each location of strain measurement as damaged or undamaged. The CNN which is fed full-field strain measurements as an input image of size 256 × 256 projects the SSD onto an output image of the same size. The data for network training is generated by numerical simulation of aluminum bars with different damage scenarios, including single damage and double damage cases at a random location, direction, length, and thickness. The trained network achieves an Intersection over Union (IoU) score of 0.790 for the validation set and 0.794 for the testing set. To check the applicability of the trained network on materials other than aluminum, testing is performed on a numerically generated steel dataset. The IoU score is 0.793, the same as the aluminum dataset, affirming the network’s capability to apply to materials exhibiting a similar stress–strain relationship. To check the generalization potential of the network, it is tested on triple damage cases; the IoU score is found to be 0.764, suggesting that the network works well for unseen damage patterns as well. The network was also found to provide accurate predictions for real experimental data obtained from Strain Sensing Smart Skin (S4). This proves the efficacy of the network to work in real-life scenarios utilizing the full potential of the novel full-field strain sensing methods such as S4. The performance of the proposed network affirms that it can be used as a non-destructive testing method for subsurface crack detection and localization.

show abstract

SDFormer: A Novel Transformer Neural Network for Structural Damage Identification by Segmenting the Strain Field Map

Cited by 7 publications

References 33 publications

Roadmap on measurement technologies for next generation structural health monitoring systems

Roadmap on measurement technologies for next generation structural health monitoring systems

Campus Abnormal Behavior Recognition With Temporal Segment Transformers

Deep Learning-Based Subsurface Damage Localization Using Full-Field Surface Strains

Contact Info

Product

Resources

About