Sep 2026· University of Thi-Qar journal for engineering sciences· Vol 16, pp. 112-129· 0 citations· 25 references
Abstract
This paper provides a literature survey of AI techniques used in rigid pavement monitoring, including the acquisition of image and sensor data, machine learning, and deep learning models, and mobile/IoT applications to resource-poor areas. Some of the most important AI applications are convolutional neural networks (CNNs) to identify crack segments, object detectors to identify distress in a specific location, and machine learning with accelerating sensors in smartphones to determine roughness. We address the model measures of evaluation (accuracy, F-score, IoU), existing datasets, and explore the issue of implementation. Lastly, we describe future directions which include multi-sensor fusion, edge devices and lightweight models and self-supervised learning to enhance generalization of a variety of pavement conditions.
The findings from the review indicate that YOLO and CNN were extensively employed by researchers, however in recent times, vision transformers gained popularity among researchers and pavement engineers, leading to safer, more resilient, and sustainable transportation infrastructure.
This research introduces a machine-learning algorithm to automate the identification and classification of airport pavement cracks. Utilizing drones equipped with high-definition cameras, the study implements a deep learning neural network based on the U-net framework, along with a specialized module for precise quanti...
Zefeng Lyu, C. Starr, A. Yu· IISE Annual Conference &...· 0 citations
Railway sleeper degradation is a major problem for track safety and maintenance planning, especially in real–world scenarios where ballast occlusion, variable lighting, and weather–induced surface changes make visual inspection difficult. Surface–based image analysis can also be affected by complex backgrounds, while c...
This study highlights the importance of early detection of tire cracks to improve driving safety. Conventional inspection methods remain manual, time-consuming, and prone to human error, thereby making them less effective. Therefore, this research aims to analyze the application of big data and predictive analytics for...
Shofa Shofiah Hilabi, Savina Savina, Rieke Retnosary· Applied Information System a...· 0 citations
A focused review of recent advancements in vision-based concrete crack detection and measurement is presented, synthesizing findings from 27 representative and influential studies selected through an expert-driven, non-exhaustive screening process rather than a fully systematic protocol.
The HPA (Hybrid Pooling Attention) module in the YOLOv8n Neck stage is introduced, which combines average pooling and max pooling with cross space learning to enhance the transmission and fusion of multi-scale crack features and alleviate information attenuation in feature propagation.
Gui-Zhuo Ai, Lang Ma· Academic Journal of Architec...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
