A Sewer Pipeline Defect Detection Method Based on Improved YOLOv5

Tong Wang; Yuhang Li; Yidi Zhai; Weihua Wang; Rongjie Huang

LAPSE

Living Archive for Process Systems Engineering

LAPSE:2024.0059

Published Article

LAPSE:2024.0059

A Sewer Pipeline Defect Detection Method Based on Improved YOLOv5

Tong Wang, Yuhang Li, Yidi Zhai, Weihua Wang, Rongjie Huang

January 12, 2024

To address the issues of strong subjectivity, low efficiency, and difficulty in on-site model deployment encountered in existing CCTV defect detection of pipelines, this article proposes an object detection model based on an improved YOLOv5s algorithm. Firstly, involution modules and GSConv simplified models are introduced into the backbone network and feature fusion network, respectively, to enhance the detection accuracy. Secondly, a CBAM attention mechanism is integrated to improve the detection accuracy of overlapping targets in complex backgrounds. Finally, knowledge distillation is performed on the improved model to further enhance its accuracy. Experimental results demonstrate that the improved YOLOv5s achieved an mAP@0.5 of 80.5%, which is a 2.4% increase over the baseline, and reduces the parameter and computation volume by 30.1% and 29.4%, respectively, with a detection speed of 75 FPS. This method offers good detection accuracy and robustness while ensuring real-time detection and can be employed in the on-site detection process of sewer pipeline defects.

Record ID

LAPSE:2024.0059

Keywords

attention mechanism, detection of sewer defects, GSConv, improved YOLOv5, involution, knowledge distillation

Subject

Process Monitoring

Suggested Citation

Wang T, Li Y, Zhai Y, Wang W, Huang R. A Sewer Pipeline Defect Detection Method Based on Improved YOLOv5. (2024). LAPSE:2024.0059

Author Affiliations

Wang T: Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, China; College of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China;
Li Y: Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, China; College of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China
Zhai Y: Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, China; College of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China
Wang W: China Special Equipment Inspection and Research Institute, Beijing 100029, China [ORCID]
Huang R: Henan Key Laboratory of Intelligent Manufacturing of Mechanical Equipment, Zhengzhou University of Light Industry, Zhengzhou 450002, China; College of Mechanical and Electrical Engineering, Zhengzhou University of Light Industry, Zhengzhou 450002, China;

Journal Name

Processes

Volume

11

Issue

8

First Page

2508

Year

2023

Publication Date

2023-08-21