Citation: Zheng, J.; Wu, H.; Zhang,
H.; Wang, Z.; Xu, W. Insulator-Defect
Detection Algorithm Based on
Improved YOLOv7. Sensors 2022, 22,
8801. https://doi.org/10.3390/
s22228801
Academic Editor: Yuan Yao
Received: 17 October 2022
Accepted: 8 November 2022
Published: 14 November 2022
Publisher’s Note: MDPI stays neutral
with regard to jurisdictional claims in
published maps and institutional affil-
iations.
Copyright: © 2022 by the authors.
Licensee MDPI, Basel, Switzerland.
This article is an open access article
distributed under the terms and
conditions of the Creative Commons
Attribution (CC BY) license (https://
creativecommons.org/licenses/by/
4.0/).
Article
Insulator-Defect Detection Algorithm Based on Improved YOLOv7
Jianfeng Zheng
1,2
, Hang Wu
1
, Han Zhang
3
, Zhaoqi Wang
1
and Weiyue Xu
1,2
*
1
School of Mechanical Engineering and Rail Transit, Changzhou University, Changzhou 213164, China
2
Jiangsu Province Engineering Research Center of High-Level Energy and Power Equipment,
Changzhou University, Changzhou 213164, China
3
Key Laboratory of Noise and Vibration, Institute of Acoustics, Chinese Academy of Sciences,
Beijing 100190, China
* Correspondence: wyxu@cczu.edu.cn
Abstract:
Existing detection methods face a huge challenge in identifying insulators with minor
defects when targeting transmission line images with complex backgrounds. To ensure the safe
operation of transmission lines, an improved YOLOv7 model is proposed to improve detection
results. Firstly, the target boxes of the insulator dataset are clustered based on K-means++ to generate
more suitable anchor boxes for detecting insulator-defect targets. Secondly, the Coordinate Attention
(CoordAtt) module and HorBlock module are added to the network. Then, in the channel and spatial
domains, the network can enhance the effective features of the feature-extraction process and weaken
the ineffective features. Finally, the SCYLLA-IoU (SIoU) and focal loss functions are used to accelerate
the convergence of the model and solve the imbalance of positive and negative samples. Furthermore,
to optimize the overall performance of the model, the method of non-maximum suppression (NMS)
is improved to reduce accidental deletion and false detection of defect targets. The experimental
results show that the mean average precision of our model is 93.8%, higher than the Faster R-CNN
model, the YOLOv7 model, and YOLOv5s model by 7.6%, 3.7%, and 4%, respectively. The proposed
YOLOv7 model can effectively realize the accurate detection of small objects in complex backgrounds.
Keywords: YOLOv7; insulator-defect detection; attention mechanism; HorBlock; SIoU
1. Introduction
An insulator is a special kind of insulation control and is essential electrical equipment
in overhead transmission lines. In early years, insulators were mainly used for power poles
and gradually developed into disc insulators, usually made of glass or ceramics, hung on
one end of the connection tower of high-voltage power lines. They fill the role of electrical
insulation and mechanical fixation in transmission lines [
1
]. In addition, by hanging
insulators on transmission lines, the transmission distance can be increased and capacitive
reactance between transmission lines can be reduced. However, due to the long-term
influence of factors such as strong electric fields and harsh environments, insulators have
many defects, such as self-explosion, damage, pollution flashover, and current leakage [
2
,
3
].
Among these defects, the most common faults are insulator damage and surface defects
caused by pollution flashovers. The damage is mainly caused by insulator manufacturing
defects, the combined action of various stresses, and other reasons. Surface defects are
mainly caused by partial discharge in pollution flashovers. Pollution flashover means
that pollutants are attached to the insulation surface, and soluble substances gradually
dissolve in water under wet conditions, forming a conductive film on the insulation surface
that reduces the insulation strength and increases the leakage current, resulting in partial
discharge [
4
]. According to the research, power system paralysis caused by insulator defects
accounts for more than half of power grid system failures. Therefore, research on the rapid
detection and identification of insulators and their defects has excellent application value
for maintenance and repair personnel [5].
Sensors 2022, 22, 8801. https://doi.org/10.3390/s22228801 https://www.mdpi.com/journal/sensors
