Deep learning-based construction equipment operators’ mental fatigue classification using wearable EEG sensor data

Advanced Engineering Informatics, Journal Year: 2023, Volume and Issue: 56, P. 101978 - 101978

Published: April 1, 2023

Language: Английский

---

Small and overlapping worker detection at construction sites

Automation in Construction, Journal Year: 2023, Volume and Issue: 151, P. 104856 - 104856

Published: April 12, 2023

Although there has been study on worker detection using computer vision (CV) for the safety of construction sites, it is still challenging to identify employees who are obstructed or have poor vision. To solve these problems, we propose a method small and overlapping target (worker) at complex site named SOC-YOLO. The based YOLOv5 utilizes distance intersection over union (DIoU) non-maximum suppression (NMS), incorporating weighted triplet attention, expansion feature-level, Soft-pool. Workers can be captured with overlap, particularly in large-scale DIoU-based loss function, NMS contributed accuracy improvement. Next, weighted-triplet attention mechanism that extract feature information from space more effectively channel when learning object networks, simple average approach same weight between existing attention. model adds additional predictive heads residual connections address workers photographed long distances. A low-level map containing regarding targets used by extending level. Finally, Softpool-spatial pyramid pooling fast (Softpool-SPPF) proposed problem inconsistent input image sizes. Softpool-SPPF performs an spatial (SPP) function while preserving functional accurate detection. Experiments were conducted published datasets handmade datasets, results showed increase 81.26% 84.63% precision (AP) objects, 67.52% 73.88% mAP minute 74.56% to77.57% objects. expected useful monitoring applying tracking model.

Language: Английский

---

Integrating text parsing and object detection for automated monitoring of finishing works in construction projects

Automation in Construction, Journal Year: 2025, Volume and Issue: 174, P. 106139 - 106139

Published: March 23, 2025

Language: Английский

---

Robot-assisted mobile scanning for automated 3D reconstruction and point cloud semantic segmentation of building interiors

Automation in Construction, Journal Year: 2023, Volume and Issue: 152, P. 104949 - 104949

Published: May 29, 2023

Language: Английский

---

Construction Instance Segmentation (CIS) Dataset for Deep Learning-Based Computer Vision

Automation in Construction, Journal Year: 2023, Volume and Issue: 156, P. 105083 - 105083

Published: Sept. 9, 2023

Language: Английский

---

YOLO and Faster R-CNN object detection for smart Industry 4.0 and Industry 5.0: applications, challenges, and opportunities

SSRN Electronic Journal, Journal Year: 2023, Volume and Issue: unknown

Published: Jan. 1, 2023

The rise of Industry 4.0 and the emerging paradigm 5.0 have driven unprecedented technological progress in various fields. Central to this transformation are real-time object detection technologies, notably You Only Look Once (YOLO) Faster Region Convolutional Neural Network (Faster R-CNN) algorithms. This study thoroughly examines applications, challenges, prospects YOLO R-CNN diverse industrial domains. In realm automation, these algorithms redefined efficiency safety standards by enabling rapid precise recognition, thus enhancing overall production workflows. Furthermore, construction industry has experienced significant advancements project management site safety, thanks accurate identification materials equipment. healthcare, revolutionized patient care facilitating medical instruments anomalies, thereby improving diagnostics treatment processes. integration into autonomous vehicles substantially enhanced their capabilities, ensuring superior road navigation. Additionally, precision agriculture, streamlined crop management, leading increased agricultural productivity sustainability. Moreover, retail e-commerce sectors undergone a shift with personalized customer experiences efficient inventory all powered technologies. Despite remarkable advancements, paper explores challenges such as data privacy concerns, computational complexity, ethical considerations. Addressing opens unique avenues for further research innovation. Lastly, environmental monitoring also benefited from algorithms, tracking analysis changes informed decision-making towards sustainable future. illuminates transformative potential detection, paving way ongoing upcoming 5.0. These technologies shaping smarter, more connected, future across sectors.

Language: Английский

---

Deep learning-based automated productivity monitoring for on-site module installation in off-site construction

Developments in the Built Environment, Journal Year: 2024, Volume and Issue: 18, P. 100382 - 100382

Published: March 12, 2024

Effectively monitoring and analyzing on-site module installation for modular integrated construction (MiC) is essential to properly coordinating the MiC process. In this study, authors propose an automated productivity framework operations consisting of three modules: object detection, activity classification, analysis. The detection detects mobile cranes modules interacting with cranes, classification classifies activities into five different by considering spatiotemporal relationship between detected objects. Finally, analysis analyzes process utilizing accumulated results over image frames. proposed model achieves average accuracy 89% (hooking: 85.71%, lifting: 84.44%, positioning: 94.90%, returning: 83.09%, idling: 96.87%) in classifying activities. developed enables practitioners measure automatically. addition, data stored from diverse sites contribute identifying progress-impeding factors improving entire

Language: Английский

---

From data to knowledge: Construction process analysis through continuous image capturing, object detection, and knowledge graph creation

Automation in Construction, Journal Year: 2024, Volume and Issue: 164, P. 105451 - 105451

Published: May 9, 2024

Persistent issues of schedule deviations and cost overruns within large construction projects aggravate the industry's global productivity concerns. However, how holistic, data-oriented methods can effectively be leveraged for investigating project performance identifying potential bottlenecks during phase remains unanswered. Our research addresses this issue with a novel approach encompassing data acquisition, object detection, geometric projection, graph-based linking. Image data, continuously captured by crane-camera systems, gets transformed into higher-level information using an end-to-end deep learning-based pipeline that covers detection specific on-site objects integrates it in knowledge graph. The graph facilitates extracting precise metrics, spatiotemporal irregularities, like work hotspots characterized high activity intensive concentrations, but also phases low activity. proposed method improves learning from past aiding stakeholders inspiring further real-time monitoring, predictive analytics, data-integrated decision-making systems to reshape practices.

Language: Английский

---

Automatic identification of integrated construction elements using open-set object detection based on image and text modality fusion

Advanced Engineering Informatics, Journal Year: 2025, Volume and Issue: 64, P. 103075 - 103075

Published: Jan. 6, 2025

Language: Английский

---

A Comprehensive Survey of Deep Learning Approaches in Image Processing

Sensors, Journal Year: 2025, Volume and Issue: 25(2), P. 531 - 531

Published: Jan. 17, 2025

The integration of deep learning (DL) into image processing has driven transformative advancements, enabling capabilities far beyond the reach traditional methodologies. This survey offers an in-depth exploration DL approaches that have redefined processing, tracing their evolution from early innovations to latest state-of-the-art developments. It also analyzes progression architectural designs and paradigms significantly enhanced ability process interpret complex visual data. Key such as techniques improving model efficiency, generalization, robustness, are examined, showcasing DL's address increasingly sophisticated image-processing tasks across diverse domains. Metrics used for rigorous evaluation discussed, underscoring importance performance assessment in varied application contexts. impact is highlighted through its tackle challenges generate actionable insights. Finally, this identifies potential future directions, including emerging technologies like quantum computing neuromorphic architectures efficiency federated privacy-preserving training. Additionally, it highlights combining with edge explainable artificial intelligence (AI) scalability interpretability challenges. These advancements positioned further extend applications DL, driving innovation processing.

Language: Английский

---