Kolmogorov-Arnold networks for metal surface defect classification

2025-01-10

AI Summary

To address the challenges of few-shot learning, slow convergence, and poor interpretability in fine-grained classification of steel surface defects (e.g., cracks, inclusions, patches, pits, and scratches), this work pioneers the application of Kolmogorov–Arnold Networks (KANs) to industrial surface defect recognition. We propose an end-to-end image classification architecture featuring learnable B-spline activations, explicitly replacing conventional MLP weight matrices with interpretable spline functions and eliminating convolutional operations—thereby achieving parameter efficiency and structural transparency. Evaluated on a metal surface defect dataset, our method outperforms CNN baselines by +3.2% in accuracy, reduces model parameters by 67%, and accelerates training convergence by 2.1×. This work establishes a lightweight, highly interpretable, and high-performance paradigm for few-shot industrial vision tasks.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Knowledge Representation and Reasoning: Diagnosis and Abductive Reasoning
• Computer Vision: Representation Learning for Vision

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Web Mining and Content Analysis: Large pretrained models with web data
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

This paper presents the application of Kolmogorov-Arnold Networks (KAN) in classifying metal surface defects. Specifically, steel surfaces are analyzed to detect defects such as cracks, inclusions, patches, pitted surfaces, and scratches. Drawing on the Kolmogorov-Arnold theorem, KAN provides a novel approach compared to conventional multilayer perceptrons (MLPs), facilitating more efficient function approximation by utilizing spline functions. The results show that KAN networks can achieve better accuracy than convolutional neural networks (CNNs) with fewer parameters, resulting in faster convergence and improved performance in image classification.

Problem

• Kolmogorov-Arnold network
• metal surface
• flaw discrimination

Innovation

• Kolmogorov-Arnold Networks
• Metal Surface Defect Detection
• Efficiency vs. Convolutional Neural Networks