• Home
  • Search
  • Semi-Supervised Surface Defect Segmentation via Contrastive Learning
  • https://doi.org/10.1109/tim.2025.3612555Copy DOI Icon

Semi-Supervised Surface Defect Segmentation via Contrastive Learning

Show More
  • Abstract
  • Literature Map
  • References
  • Similar Papers
Abstract

Quality inspection of industrial products is a critical task to ensure that products meet the required standards. Although deep learning-based detection models have achieved significant progress in recent years, the limited availability of samples in real industrial environments still constrains the generalizability of these models. In this paper, we propose a novel semi-supervised defect segmentation framework that leverages contrastive learning to handle a large number of unlabeled samples. The core of this framework is the contrastive consistency regularization loss and class affinity loss, which together enhance the model’s ability to learn from unlabeled data. Our proposed contrastive consistency regularization loss incorporates pixel-level features from the entire batch into a cross-image consistency objective, substantially enhancing the model’s robustness to appearance variations. Complementarily, the class affinity loss leverages a compact memory bank of high-quality prototypes to directly refine the feature space—pulling unlabeled pixels toward their probable class centroids and pushing them away from alternative classes—thereby tightening intra-class compactness and sharpening inter-class boundaries for superior segmentation accuracy. The experimental results indicate that in the three different available labeled data experiments on NEU-SEG dataset, our proposed method improves the mIoU of the supervised learning paradigm by 4.4%, 3.7%, and 3.8%, respectively, surpassing the previous state-of-the-art method by 0.6%, 0.7%, and 0.5%. Finally, the ablation studies validate the effectiveness of the proposed method.

Similar Papers
  • Research Article
  • Citations29

Deep Semisupervised Teacher–Student Model Based on Label Propagation for Sea Ice Classification

  • Jan 01, 2021
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Salman Khaleghian +4
  • Research Article
  • Citations52

A Covert Electricity-Theft Cyberattack Against Machine Learning-Based Detection Models

  • Nov 01, 2022
  • IEEE Transactions on Industrial Informatics
  • Lei Cui +6
  • Research Article
  • Citations17

Efficient Hierarchical Performance Modeling for Analog and Mixed-Signal Circuits via Bayesian Co-Learning

  • Dec 01, 2018
  • IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems
  • Mohamed Baker Alawieh +2
  • Research Article
  • Citations7

Semi-supervised adaptive anti-noise meta-learning for few-shot industrial gearbox fault diagnosis

  • Aug 02, 2024
  • Measurement Science and Technology
  • Junwei Hu +1
  • Research Article
  • Citations77

Positive-unlabeled learning in bioinformatics and computational biology: a brief review.

  • Nov 03, 2021
  • Briefings in Bioinformatics
  • Fuyi Li +13
  • Research Article

Semi-Supervised Regression by Preserving Ranking Relationships Between Close Unlabeled Samples

  • Mar 14, 2026
  • X Li +4
  • Conference Article
  • Citations14

Scalable Semi-Supervised SVM via Triply Stochastic Gradients

  • Aug 01, 2019
  • Xiang Geng +5
  • Conference Article

Tibetan-Chinese Cross-Lingual Sentiment Classification Based on Co-Training

  • Sep 01, 2021
  • Tingting Zhang +2
  • Research Article
  • Citations2

An Analysis Model of Sports Human Body Based on Computer Vision Tracking Technology

  • Sep 11, 2017
  • DEStech Transactions on Social Science, Education and Human Science
  • Mingzhu Yuan
  • Research Article
  • Citations181

Active learning for regression using greedy sampling

  • Sep 26, 2018
  • Information Sciences
  • Dongrui Wu +2
  • PDF
  • Research Article
  • Citations20

Semisupervised Variational Generative Adversarial Networks for Hyperspectral Image Classification

  • Jan 01, 2020
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Chao Tao +3
  • Research Article
  • Citations42

A new dynamic security assessment framework based on semi-supervised learning and data editing

  • Mar 28, 2019
  • Electric Power Systems Research
  • Ruidong Liu +2
  • Research Article
  • Citations2

Semi-supervised interlayer intelligent recognition method

  • May 26, 2023
  • Earth Science Informatics
  • Shixiang Jiao +6
  • Research Article
  • Citations8

Combining active learning and transductive support vector machines for sea ice detection

  • May 15, 2018
  • Journal of Applied Remote Sensing
  • Yanling Han +5
  • Preprint Article

DCSSM: Enhancing Medical Organ Segmentation with Positional Correlation and Feature Alignment

  • Mar 19, 2025
  • Preprints.org
  • Linjiang Liu +4
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.