Automated Neural Architecture Design for Industrial Defect Detection

2025-10-08

AI Summary

Industrial surface defect detection faces the core challenges of large intra-class variability and high inter-class similarity. To address these, this paper proposes AutoNAD, a fully automated neural architecture search (NAS) framework that jointly optimizes convolutional, Transformer, and MLP structures—enabling synergistic modeling of local details and global semantics. Key innovations include a cross-weight-sharing strategy to accelerate supernet training, a searchable multi-level feature aggregation module for enhanced multi-scale representation learning, and latency-aware architectural priors to optimize inference efficiency. Evaluated on three standard benchmarks—MVTec AD, NEU-CLS, and DAGM—AutoNAD consistently outperforms state-of-the-art methods, achieving superior trade-offs between detection accuracy and inference speed. The framework has been successfully deployed in an industrial defect imaging and inspection platform.

Technology Category

• Machine Learning: Deep Neural Architectures and Foundation Models
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Industrial surface defect detection (SDD) is critical for ensuring product quality and manufacturing reliability. Due to the diverse shapes and sizes of surface defects, SDD faces two main challenges: intraclass difference and interclass similarity. Existing methods primarily utilize manually designed models, which require extensive trial and error and often struggle to address both challenges effectively. To overcome this, we propose AutoNAD, an automated neural architecture design framework for SDD that jointly searches over convolutions, transformers, and multi-layer perceptrons. This hybrid design enables the model to capture both fine-grained local variations and long-range semantic context, addressing the two key challenges while reducing the cost of manual network design. To support efficient training of such a diverse search space, AutoNAD introduces a cross weight sharing strategy, which accelerates supernet convergence and improves subnet performance. Additionally, a searchable multi-level feature aggregation module (MFAM) is integrated to enhance multi-scale feature learning. Beyond detection accuracy, runtime efficiency is essential for industrial deployment. To this end, AutoNAD incorporates a latency-aware prior to guide the selection of efficient architectures. The effectiveness of AutoNAD is validated on three industrial defect datasets and further applied within a defect imaging and detection platform. Code will be available at https://github.com/Yuxi104/AutoNAD.

Problem

• Automating neural architecture design for industrial defect detection
• Addressing intraclass difference and interclass similarity challenges
• Improving detection accuracy while maintaining runtime efficiency

Innovation

• Automated neural architecture design for defect detection
• Hybrid search combining convolutions, transformers, and MLPs
• Cross weight sharing strategy for efficient training