Deep Learning for Structural Health Monitoring: Data, Algorithms, Applications, Challenges, and Trends

Sensors, Journal Year: 2023, Volume and Issue: 23(21), P. 8824 - 8824

Published: Oct. 30, 2023

Environmental effects may lead to cracking, stiffness loss, brace damage, and other damages in bridges, frame structures, buildings, etc. Structural Health Monitoring (SHM) technology could prevent catastrophic events by detecting damage early. In recent years, Deep Learning (DL) has developed rapidly been applied SHM detect, localize, evaluate diverse through efficient feature extraction. This paper analyzes 337 articles a systematic literature review investigate the application of DL for operation maintenance phase facilities from three perspectives: data, algorithms, applications. Firstly, data types corresponding collection methods are summarized analyzed. The most common vibration signals images, accounting 80% studied. Secondly, popular algorithm areas reviewed, which CNN accounts 60%. Then, this article carefully specific functions based on facility’s characteristics. scrutinized study focused cracks, 30 percent research papers. Finally, challenges trends applying discussed. Among trends, Digital Twin (SHMDT) model framework is suggested response trend strong coupling between (DT), can advance digitalization, visualization, intelligent management SHM.

Language: Английский

---

Rock Image Classification Based on EfficientNet and Triplet Attention Mechanism

Applied Sciences, Journal Year: 2023, Volume and Issue: 13(5), P. 3180 - 3180

Published: March 1, 2023

Rock image classification is a fundamental and crucial task in the creation of geological surveys. Traditional rock methods mainly rely on manual operation, resulting high costs unstable accuracy. While existing based deep learning models have overcome limitations traditional achieved intelligent classification, they still suffer from low accuracy due to suboptimal network structures. In this study, model EfficientNet triplet attention mechanism proposed achieve accurate end-to-end classification. The was built EfficientNet, which boasts an efficient structure thanks NAS technology compound scaling method, thus achieving for Additionally, introduced address shortcoming feature expression enable fully capture channel spatial information images, further improving During training, transfer employed by loading pre-trained parameters into model, accelerated convergence reduced training time. results show that with 92.6% set 93.2% Top-1 test set, outperforming other mainstream demonstrating strong robustness generalization ability.

Language: Английский

---

Experiments on image data augmentation techniques for geological rock type classification with convolutional neural networks

Journal of Rock Mechanics and Geotechnical Engineering, Journal Year: 2024, Volume and Issue: unknown

Published: May 1, 2024

The integration of image analysis through deep learning (DL) into rock classification represents a significant leap forward in geological research. While traditional methods remain invaluable for their expertise and historical context, DL offers powerful complement by enhancing the speed, objectivity, precision process. This research explores significance data augmentation techniques optimizing performance convolutional neural networks (CNNs) analysis, particularly igneous, metamorphic, sedimentary types from thin section (RTS) images. study primarily focuses on classic evaluates impact model accuracy precision. Results demonstrate that like Equalize significantly enhance model's capabilities, achieving an F1-Score 0.9869 igneous rocks, 0.9884 metamorphic 0.9929 representing improvements compared to baseline original results. Moreover, weighted average across all classes is 0.9886, indicating enhancement. Conversely, Distort lead decreased F1-Score, with 0.949 0.954 0.9416 exacerbating baseline. underscores practicality advocates adoption this domain automation improved findings can benefit various fields, including remote sensing, mineral exploration, environmental monitoring, both scientific industrial applications.

Language: Английский

---

Convolutional neural network-based model for recognizing TBM rock chip gradation

Automation in Construction, Journal Year: 2024, Volume and Issue: 163, P. 105414 - 105414

Published: April 13, 2024

Language: Английский

---

Machine Learning Approaches for Real-Time Mineral Classification and Educational Applications

Applied Sciences, Journal Year: 2025, Volume and Issue: 15(4), P. 1871 - 1871

Published: Feb. 11, 2025

The main objective of the present study was to develop a real-time mineral classification system designed for multiple detection, which integrates classical computer vision techniques with advanced deep learning algorithms. employs three CNN architectures—VGG-16, Xception, and MobileNet V2—designed identify minerals within single frame output probabilities various types, including Pyrite, Aragonite, Quartz, Obsidian, Gypsum, Azurite, Hematite. Among these, V2 demonstrated exceptional performance, achieving highest accuracy (98.98%) lowest loss (0.0202), while Xception VGG-16 also performed competitively, excelling in feature extraction detailed analyses, respectively. Gradient-weighted Class Activation Mapping visualizations illustrated models’ ability capture distinctive features, enhancing interpretability. Furthermore, stacking ensemble approach achieved an impressive 99.71%, effectively leveraging complementary strengths individual models. Despite its robust method poses computational challenges, particularly applications on resource-constrained devices. application this methodology Mineral Quest, educational Python-based game, underscores practical potential geology education, mining, geological surveys, offering engaging accurate tool classification.

Language: Английский

---

Lithology Identification of Lithium Minerals Based on TL-FMix-MobileViT Model

Natural Resources Research, Journal Year: 2025, Volume and Issue: unknown

Published: March 11, 2025

Language: Английский

---

An intelligent lithology recognition system for continental shale by using digital coring images and convolutional neural networks

Geoenergy Science and Engineering, Journal Year: 2024, Volume and Issue: 239, P. 212909 - 212909

Published: May 10, 2024

Language: Английский

---

OCM: an intelligent recognition method of rock discontinuity based on optimal color mapping of 3D Point cloud via deep learning

Rock Mechanics and Rock Engineering, Journal Year: 2024, Volume and Issue: 57(7), P. 4873 - 4905

Published: March 18, 2024

Language: Английский

---

Review of Image Processing Methods for Surface and Tool Condition Assessments in Machining

Journal of Manufacturing and Materials Processing, Journal Year: 2024, Volume and Issue: 8(6), P. 244 - 244

Published: Oct. 31, 2024

This paper systematically explores the applications of image processing techniques in machined surface analysis, a critical area industries like manufacturing, aerospace, automotive, and healthcare. It examines integration traditional Computer Numerical Control (CNC) machining micromachining, focusing on its role tool wear workpiece detection, automatic CNC programming, defect inspection. With AI machine learning advancements, these technologies enhance texture predictive maintenance, quality optimization. The also discusses future advancements high resolutions, 3D imaging, augmented reality, Industry 4.0, highlighting their impact productivity, precision, challenges such as data privacy. In conclusion, remains vital to improving manufacturing efficiency control.

Language: Английский

---

Identification of maize and wheat seedlings and weeds based on deep learning

Frontiers in Earth Science, Journal Year: 2023, Volume and Issue: 11

Published: Feb. 10, 2023

Introduction: It is well-known that maize and wheat are main food crops in the world. Thus, promoting high quality abundant guarantees development of grain industry, which needed to support world hunger. Weeds seriously affect growing environment maize, wheat, their seedlings, resulting low crop yields poor seedling quality. This paper focuses on identification seedlings field weeds using deep learning. Methods: Maize research objects. A weed model based UNet network ViT classification algorithm proposed. The uses segment images. Python Imaging Library used green plant leaves from binary images, enhance feature extraction leaves. segmented image construct a model, improves recognition accuracy field. Results: average accuracy, recall, F1 score evaluate performance model. rate (for accurately identifying field) reaches 99.3%. Compared with Alexnet, VGG16, MobileNet V3 models, results show effect trained method presented this better than other existing models. Discussion: method, disambiguates can provide effective information for subsequent pesticide spraying mechanical weeding.

Language: Английский

---