Defect-aware transformer network for intelligent visual surface defect detection

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 55, P. 101882 - 101882

Published: Jan. 1, 2023

Language: Английский

---

Reconstruction-based anomaly detection for multivariate time series using contrastive generative adversarial networks

Information Processing & Management, Journal Year: 2023, Volume and Issue: 61(1), P. 103569 - 103569

Published: Nov. 4, 2023

Language: Английский

---

Hierarchical spatial attention-based cross-scale detection network for Digital Works Supervision System (DWSS)

Computers & Industrial Engineering, Journal Year: 2024, Volume and Issue: 192, P. 110220 - 110220

Published: May 12, 2024

Language: Английский

---

Construction safety management in the data-rich era: A hybrid review based upon three perspectives of nature of dataset, machine learning approach, and research topic

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 58, P. 102144 - 102144

Published: Aug. 17, 2023

Language: Английский

---

Vision-Based Detection Method for Construction Site Monitoring by Integrating Data Augmentation and Semisupervised Learning

Journal of Construction Engineering and Management, Journal Year: 2024, Volume and Issue: 150(5)

Published: Feb. 26, 2024

Training deep learning models for vision-based monitoring of construction sites usually requires a large amount labeled data. Semisupervised methods can efficiently obtain unlabeled data with substantial cost savings. Thus, this paper proposes semisupervised object detection method site monitoring. Weather as well strong and weak augmentation are integrated to cope the complex conditions (weather changes, camera view shifts, so on) by integrating leverage valid information in images. To validate its effectiveness, proposed was tested on Alberta Construction Image Data Set (ACID), public set research community. The experimental results revealed that achieves an average accuracy [mean precision (mAP)] 81.1% when trained only 3% This study helps significantly reduce development sites.

Language: Английский

---

Self-supervised contrastive video representation learning for construction equipment activity recognition on limited dataset

Automation in Construction, Journal Year: 2023, Volume and Issue: 154, P. 105001 - 105001

Published: June 21, 2023

Language: Английский

---

Vision method based on deep learning for detecting concrete vibration quality

Case Studies in Construction Materials, Journal Year: 2023, Volume and Issue: 18, P. e02132 - e02132

Published: May 9, 2023

The vibration quality of concrete is crucial to ensure the long-term safe operation structural components. In recent years, computer vision technology based on deep learning has achieved excellent results in field inspection. However, when used for assisting construction inspection, this relies heavily large-scale, labeled, high-quality image data. To solve drawback, study proposes a vision-based method that integrates semi-supervised and data augmentation detecting quality. Initially, StyleGAN2 was adopted as strategy improve diversity dataset. Then, SE-ResNet50, model couples an attention mechanism module residual network, employed classifier accurately extracting information contained images. Subsequently, order reduce workload annotation, novel (Co-MixMatch) proposed train by coupling MixMatch with co-training. Finally, trained deployed mobile devices assist onsite workers vibration. A real-world dam dataset verify method. Based experimental results, improves accuracy baseline 3.62% average. Additionally, achieves 0.9600, which only 0.67% lower than supervised learning, while requiring 20% labeled Therefore, great application prospects can further promote intelligent development

Language: Английский

---

Generalized vision-based framework for construction productivity analysis using a standard classification system

Automation in Construction, Journal Year: 2024, Volume and Issue: 165, P. 105504 - 105504

Published: June 11, 2024

Language: Английский

---

Automated daily report generation from construction videos using ChatGPT and computer vision

Automation in Construction, Journal Year: 2024, Volume and Issue: 168, P. 105874 - 105874

Published: Nov. 21, 2024

Language: Английский

---

A Hybrid Framework for Predicting Crash Severity in Construction Work Zones Using Knowledge Distillation and Conditional GANs

Journal of Computing in Civil Engineering, Journal Year: 2025, Volume and Issue: 39(2)

Published: Jan. 13, 2025

Language: Английский

---