Anomaly detection method of substation equipment based on CenterNet

Peng Zhang; Quansen Sun

doi:10.1117/12.3010298

20 October 2023 Anomaly detection method of substation equipment based on CenterNet

Peng Zhang, Quansen Sun

Author Affiliations +

Proceedings Volume 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023); 128141W (2023) https://doi.org/10.1117/12.3010298
Event: Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 2023, Chongqing, China

Abstract

Deep learning has significantly advanced object detection algorithms in the area of abnormal detection for substation equipment. However, there are still several challenges and problems that hinder its effective application in this domain. Firstly, the scale of data acquisition is limited, which may affect the performance of deep learning models. Secondly, there are various types and shapes of substation equipment with complex backgrounds, which will also affect the detection performance of deep learning models. This paper proposes an improved CenterNet-based method for target detection of substation equipment. Firstly, we embedded a Non-Local attention mechanism in the backbone network ResNet50. The integration of the attention mechanism can effectively improve the model's ability to prioritize significant targets and minimize the negative impact of intricate backgrounds on its overall performance. Then, the DIOU loss is used to replace the center point offset loss and width-height loss to accelerate network convergence and improve model performance. The proposed method exhibits superior performance in substation equipment anomaly detection, as evidenced by experimental results, with a detection accuracy increase of 10.8% over the original method and outperforming existing techniques like Faster R-CNN.

(2023) Published by SPIE. Downloading of the abstract is permitted for personal use only.

Citation Download Citation

Peng Zhang and Quansen Sun "Anomaly detection method of substation equipment based on CenterNet", Proc. SPIE 12814, Third International Conference on Green Communication, Network, and Internet of Things (CNIoT 2023), 128141W (20 October 2023); https://doi.org/10.1117/12.3010298

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

;

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available

Members: $17.00

Non-members: $21.00 ADD TO CART

PROCEEDINGS
6 PAGES

DOWNLOAD PAPER SAVE TO MY LIBRARY

GET CITATION

RIGHTS & PERMISSIONS

Get copyright permission Get copyright permission on Copyright Marketplace

KEYWORDS

Object detection

Detection and tracking algorithms

Data modeling

Performance modeling

Deep learning

Image processing

Instrument modeling

Show All Keywords

Keywords/Phrases

Search In:

Publication Years