Intelligent Detection Technologies for Pre-Paving Asphalt Pavement Southeast University, School of Transportation

Format:

Book

Conference/Event

Author/Creator:

Hu, Yufan, author.

Contributor:

Fan, Jianwei

Ma, Tao

Tang, Fanlong

Conference Name:

2024 International Conference on Smart Transportation Interdisciplinary Studies (2024-12-13 : Nanjing, China)

Language:

English

Physical Description:

1 online resource cm

Place of Publication:

Warrendale, PA SAE International 2025

Summary:

This paper presents advanced intelligent monitoring methods aimed at enhancing the quality and durability of asphalt pavement construction. The study focuses on two critical tasks: foreign object detection and the uniform application of tack coat oil. For object recognition, the YOLOv5 algorithm is employed, which provides real-time detection capabilities essential for construction environments where timely decisions are crucial. A meticulously annotated dataset comprising 4,108 images, created with the LabelImg tool, ensures the accurate detection of foreign objects such as leaves and cigarette butts. By utilizing pre-trained weights during model training, the research achieved significant improvements in key performance metrics, including precision and recall rates.In addition to object detection, the study explores color space analysis through the HSV (Hue, Saturation, Value) model to effectively differentiate between coated and uncoated pavement areas following the application of tack coat oil. Statistical analyses, including mean and standard deviation calculations of HSV values, provide critical insights into color differences that inform the establishment of threshold settings for effective identification. The research also addresses various challenges posed by environmental factors, such as steam and smoke, which can interfere with visual recognition during construction operations. To mitigate these challenges, an innovative automated mechanical system was designed to stabilize the camera, ensuring consistent data acquisition and significantly enhancing the reliability of visual data for detection tasks. By significantly improving identification accuracy and overall pavement quality, this research contributes to the development of more efficient construction methodologies and maintenance procedures. The implications of this work suggest that the adoption of advanced technologies is vital for facilitating reliable and efficient construction processes, ultimately leading to better long-term performance of asphalt pavement surfaces. This study aims to establish a foundation for future research in intelligent monitoring, promoting the continuous improvement of construction practices within the industry

Notes:

Vendor supplied data

Publisher Number:

2025-01-7179

Access Restriction:

Restricted for use by site license