• Home
  • Search
  • A Novel Adversarial Deep Learning Method for Substation Defect Image Generation.
  • Cite Icon3
  • https://doi.org/10.3390/s24144512Copy DOI Icon

A Novel Adversarial Deep Learning Method for Substation Defect Image Generation.

Show More
  • Abstract
  • Highlights & Summary
  • PDF
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

The presence of defects in substation equipment is a major factor affecting the safety of power transmission. Therefore, timely and accurate detection of these defects is crucial. As intelligent inspection robots advance, using mainstream object detection models to diagnose surface defects in substation equipment has become a focal point of current research. However, the lack of defect image data is one of the main factors affecting the accuracy of supervised deep learning-based defect detection models. To address the issue of insufficient training data for defect images with complex backgrounds, such as rust and surface oil leakage in substation equipment, which leads to the poor performance of detection models, this paper proposes a novel adversarial deep learning model for substation defect image generation: the Abnormal Defect Detection Generative Adversarial Network (ADD-GAN). Unlike existing generative adversarial networks, this model generates defect images based on effectively segmented local areas of substation equipment images, avoiding image distortion caused by global style changes. Additionally, the model uses a joint discriminator for both overall images and defect images to address the issue of low attention to local defect areas, thereby reducing the loss of image features. This approach enhances the overall quality of generated images as well as locally generated defect images, ultimately improving image realism. Experimental results demonstrate that the YOLOV7 object detection model trained on the dataset generated using the ADD-GAN method achieves a mean average precision (mAP) of 81.5% on the test dataset, and outperforms other image data augmentation and generation methods. This confirms that the ADD-GAN method can generate a high-fidelity image dataset of substation equipment defects.

Loading PDF

Similar Papers
  • Conference Article
  • Citations146

Rob-GAN: Generator, Discriminator, and Adversarial Attacker

  • Jun 01, 2019
  • Xuanqing Liu +1
  • Research Article
  • Citations43

Deep Adversarial Metric Learning.

  • Oct 25, 2019
  • IEEE Transactions on Image Processing
  • Yueqi Duan +3
  • Research Article
  • Citations19

Thermal Fault Diagnosis of Electrical Equipment in Substations Based on Image Fusion

  • Aug 31, 2021
  • Traitement du Signal
  • Mingshu Lu +2
  • Research Article
  • Citations4

Deep learning-based image classification of sea turtles using object detection and instance segmentation models

  • Nov 25, 2024
  • PLOS ONE
  • Jong-Won Baek +2
  • Research Article
  • Citations5

SAITI-DCGAN: Self-Attention Based Deep Convolutional Generative Adversarial Networks for Data Augmentation of Infrared Thermal Images

  • Dec 06, 2024
  • Applied Sciences
  • Zhichao Wu +3
  • Research Article

DCGAN-based synthetic image generation of denim jeans defects

  • Sep 29, 2025
  • PeerJ Computer Science
  • Muhammad Naeem +6
  • Research Article

Efficient Detection Framework Adaptation for Edge Computing: A Plug-and-play Neural Network Toolbox Enabling Edge Deployment

  • Jan 01, 2025
  • IEEE Transactions on Mobile Computing
  • Jiaqi Wu +9
  • Research Article
  • Citations113

Adversarial Deep Learning in EEG Biometrics.

  • Mar 27, 2019
  • IEEE Signal Processing Letters
  • Ozan Ozdenizci +3
  • Conference Article
  • Citations5

Bimodal-based Object Detection and Instance Segmentation Models for Substation Equipments

  • Oct 18, 2020
  • Nannan Yan +4
  • PDF
  • Supplementary Content
  • Citations26

Generative Adversarial Network (Generative Artificial Intelligence) in Pediatric Radiology: A Systematic Review

  • Aug 10, 2023
  • Children
  • Curtise K C Ng
  • Supplementary Content

Phenology-based learning framework for yield estimation and harvest forecasting of raspberry fruits

  • Feb 20, 2026
  • Research Square
  • Peyman Jafary +6
  • Conference Article
  • Citations1

Application of a YOLO Object Detection Model for Blind Review Mode in Image Content Recognition of Response Documents in the Power Industry

  • Nov 28, 2025
  • Jianwei Li +4
  • Research Article
  • Citations70

An adversarial attack on DNN-based black-box object detectors

  • Mar 29, 2020
  • Journal of Network and Computer Applications
  • Yajie Wang +4
  • Research Article
  • Citations43

DCFGAN: An adversarial deep reinforcement learning framework with improved negative sampling for session-based recommender systems

  • Mar 07, 2022
  • Information Sciences
  • Jianli Zhao +6
  • Research Article
  • Citations2

Aeroengine state prediction based on generative adversarial networks and deep learning

  • Jan 01, 2021
  • International Journal of Computer Aided Engineering and Technology
  • Qiang Fu +1
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.