Attention-guided few-shot learning for metal surface defect classification

2025/10/17 by Xiaoyu Liu, Yue Zhang, Qin Wang +3
Engineering · #Industrial Vision Systems and Defect Detection #Integrated Circuits and Semiconductor Failure Analysis #Surface Roughness and Optical Measurements

paper · pdf · doi:10.1007/s00530-025-02030-x

openalex created_date 2025/10/17 · openalex publication_date 2025/10/17 · openalex updated_date 2026/07/30

Abstract

Abstract In the era of industrial transformation and intelligent manufacturing, ensuring the surface quality of metal materials is crucial for maintaining component performance and system reliability. Although deep learning-based visual inspection has shown promise in automating defect detection, existing methods heavily rely on large-scale labeled data, which are often scarce and costly to obtain in industrial settings. Furthermore, the high inter-class similarity and variability of metal surface textures pose additional challenges to robust defect classification under low-data regimes. To address these issues, we propose an attention-guided few-shot learning framework specifically tailored for metal surface defect classification. Our method comprises two key components. First, we design a Dual-Branch Attention Module to enhance feature extraction by explicitly modeling both channel-wise dependencies and spatial saliency. This module leverages lightweight convolutional operations to highlight discriminative regions and mitigate feature degradation due to limited training data. Second, we introduce a Cross-Set Guided Attention mechanism to improve semantic alignment between support and query samples. By employing scaled dot-product attention, the model dynamically adjusts feature representations based on cross-sample correlations, thereby enabling fine-grained discrimination of visually similar defect types. Extensive experiments conducted on benchmark metal defect datasets demonstrate that our framework significantly outperforms existing few-shot learning baselines in both classification accuracy and generalization capability. The proposed method provides a practical and efficient solution for real-time industrial quality inspection in data-scarce scenarios.

Citations

• A survey of the recent architectures of deep convolutional neural networks
• ECA-Net: Efficient Channel Attention for Deep Convolutional Neural Networks
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Fundamental tenis : resep meraih kemenangan / Tony Mottram
• Coordinate Attention for Efficient Mobile Network Design
• Defect Image Sample Generation With Diffusion Prior for Steel Surface Defect Recognition
• Relational Embedding for Few-Shot Classification
• Multi-level Metric Learning for Few-shot Image Recognition
• Coordinate Attention for Efficient Mobile Network Design
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• Finding Task-Relevant Features for Few-Shot Learning by Category Traversal
• GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond
• Few-Shot Learning via Embedding Adaptation with Set-to-Set Functions
• Dual Attention Network for Scene Segmentation
• CBAM: Convolutional Block Attention Module
• Non-local Neural Networks
• Learning to Compare: Relation Network for Few-Shot Learning
• Squeeze-and-Excitation Networks
• Attention Is All You Need
• Prototypical Networks for Few-shot Learning
• Matching Networks for One Shot Learning
• Swin Transformer: Hierarchical Vision Transformer using Shifted Windows
• Dual Student: Breaking the Limits of the Teacher in Semi-supervised Learning

Related

• TL-SDD: A Transfer Learning-Based Method for Surface Defect Detection with Few Samples
• LERENet: Eliminating Intra-class Differences for Metal Surface Defect Few-shot Semantic Segmentation
• Advancing Metallic Surface Defect Detection via Anomaly-Guided Pretraining on a Large Industrial Dataset
• Human-machine knowledge hybrid augmentation method for surface defect detection based few-data learning
• DDPM-MoCo: Advancing Industrial Surface Defect Generation and Detection with Generative and Contrastive Learning