Scholarlink

<p>Wood surface crack detection is crucial for automated quality inspection in wood processing, yet remains challenging in real production lines due to thin and low-contrast crack patterns, strong interference from repetitive wood-grain textures, and pronounced appearance shifts across wood species and illumination conditions. To address these issues, we propose TDDC-YOLO, a robust crack detection framework built upon YOLOv8. The core idea is to explicitly disentangle texture and defect representations via an orthogonality-constrained decoupling component, thereby suppressing texture-induced false activations. Meanwhile, we introduce a lightweight crack-geometry guidance branch with readily constructible pseudo labels to enhance the structural consistency of slender cracks, and further improve robustness under domain shifts using a frequency-aware mixing strategy with matched-positive consistency regularization. Experiments on the evaluated wood-surface crack dataset show that TDDC-YOLO improves detection accuracy with limited additional overhead, reaching <InlineEquation ID="IEq1"><EquationSource Format="TEX">\(\textrm{mAP}_{50}\)</EquationSource></InlineEquation> = 0.938 and <InlineEquation ID="IEq2"><EquationSource Format="TEX">\(\textrm{mAP}_{50:95}\)</EquationSource></InlineEquation> = 0.673 on the test set while maintaining real-time inference on our test platform (3.9 ms/image, 256 FPS). Moreover, cross-wood evaluations on two species (Ash and Bubinga) indicate improved robustness to wood-species induced texture shifts (e.g., Ash-to-Bubinga <InlineEquation ID="IEq3"><EquationSource Format="TEX">\(\textrm{mAP}_{50:95}\)</EquationSource></InlineEquation> from 0.575 to 0.615), suggesting its potential for more stable detection under the evaluated cross-wood setting.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

TDDC-YOLO: texture–defect disentanglement for robust wood surface crack detection under domain shifts

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月25日

查看全文
  • Yan Chen

摘要

Wood surface crack detection is crucial for automated quality inspection in wood processing, yet remains challenging in real production lines due to thin and low-contrast crack patterns, strong interference from repetitive wood-grain textures, and pronounced appearance shifts across wood species and illumination conditions. To address these issues, we propose TDDC-YOLO, a robust crack detection framework built upon YOLOv8. The core idea is to explicitly disentangle texture and defect representations via an orthogonality-constrained decoupling component, thereby suppressing texture-induced false activations. Meanwhile, we introduce a lightweight crack-geometry guidance branch with readily constructible pseudo labels to enhance the structural consistency of slender cracks, and further improve robustness under domain shifts using a frequency-aware mixing strategy with matched-positive consistency regularization. Experiments on the evaluated wood-surface crack dataset show that TDDC-YOLO improves detection accuracy with limited additional overhead, reaching \(\textrm{mAP}_{50}\) = 0.938 and \(\textrm{mAP}_{50:95}\) = 0.673 on the test set while maintaining real-time inference on our test platform (3.9 ms/image, 256 FPS). Moreover, cross-wood evaluations on two species (Ash and Bubinga) indicate improved robustness to wood-species induced texture shifts (e.g., Ash-to-Bubinga \(\textrm{mAP}_{50:95}\) from 0.575 to 0.615), suggesting its potential for more stable detection under the evaluated cross-wood setting.