HyperDefect-YOLO: Enhance YOLO with HyperGraph Computation for Industrial Defect Detection

TL;DR

HD-YOLO addresses industrial defect detection under varied defect sizes and complex backgrounds by embedding hypergraph computation into the YOLO framework. It introduces Defect Aware Module (DAM), Mixed Graph Network (MGNet), HyperGraph Aggregation Network (HGANet) for multi-scale feature fusion with Distance-Based Attention, Cross-Scale Fusion (CSF) for scale-aware fusion, and a Semantic Aware Module (SAM) in the neck for enhanced semantic reasoning. Across HRIPCB, NEU-DET, and MINILED, HD-YOLO achieves state-of-the-art $Pre$ and $mAP_{0.5}$ with strong efficiency, demonstrating practical viability for real-world industrial inspection. The work shows significant improvements in tiny defect detection, robustness to cluttered backgrounds, and generalization across diverse products and datasets.

Abstract

In the manufacturing industry, defect detection is an essential but challenging task aiming to detect defects generated in the process of production. Though traditional YOLO models presents a good performance in defect detection, they still have limitations in capturing high-order feature interrelationships, which hurdles defect detection in the complex scenarios and across the scales. To this end, we introduce hypergraph computation into YOLO framework, dubbed HyperDefect-YOLO (HD-YOLO), to improve representative ability and semantic exploitation. HD-YOLO consists of Defect Aware Module (DAM) and Mixed Graph Network (MGNet) in the backbone, which specialize for perception and extraction of defect features. To effectively aggregate multi-scale features, we propose HyperGraph Aggregation Network (HGANet) which combines hypergraph and attention mechanism to aggregate multi-scale features. Cross-Scale Fusion (CSF) is proposed to adaptively fuse and handle features instead of simple concatenation and convolution. Finally, we propose Semantic Aware Module (SAM) in the neck to enhance semantic exploitation for accurately localizing defects with different sizes in the disturbed background. HD-YOLO undergoes rigorous evaluation on public HRIPCB and NEU-DET datasets with significant improvements compared to state-of-the-art methods. We also evaluate HD-YOLO on self-built MINILED dataset collected in real industrial scenarios to demonstrate the effectiveness of the proposed method. The source codes are at https://github.com/Jay-zzcoder/HD-YOLO.

Figures (13)

• Figure 1: Examples and data distribution of different datasets.
• Figure 2: The process of hypergraph construction.
• Figure 3: Framework of the proposed HD-YOLO.
• Figure 4: Illustration of the proposed Mixed Graph Network (MGNet).
• Figure 5: The structure of HyperGraph Aggregation Network (HGANet).