Xue Shuai, Xu Wei, Xiong Zhu, Zhang Jing, Liang Yanyan
School of Computer Science and Engineering, Faculty of Innovation Engineering, Macau University of Science and Technology, Macau 999078, China.
School of Applied Science and Civil Engineering, Beijing Institute of Technology, Zhuhai 519000, China.
Materials (Basel). 2025 Aug 3;18(15):3646. doi: 10.3390/ma18153646.
Accurate weld defect detection is critical for ensuring structural safety and evaluating welding quality in industrial applications. Manual inspection methods have inherent limitations, including inefficiency and inadequate sensitivity to subtle defects. Existing detection models, primarily designed for natural images, struggle to adapt to the characteristic challenges of weld X-ray images, such as high noise, low contrast, and inter-defect similarity, particularly leading to missed detections and false positives for small defects. To address these challenges, a multi-dimensional feature fusion model (MADet), which is a multi-branch deep fusion network for weld defect detection, was proposed. The framework incorporates two key innovations: (1) A multi-scale feature fusion network integrated with lightweight attention residual modules to enhance the perception of fine-grained defect features by leveraging low-level texture information. (2) An anchor-based feature-selective detection head was used to improve the discrimination and localization accuracy for five typical defect categories. Extensive experiments on both public and proprietary weld defect datasets demonstrated that MADet achieved significant improvements over the state-of-the-art YOLO variants. Specifically, it surpassed the suboptimal model by 7.41% in mAP@0.5, indicating strong industrial applicability.
准确的焊缝缺陷检测对于确保工业应用中的结构安全和评估焊接质量至关重要。人工检测方法存在固有局限性，包括效率低下以及对细微缺陷的敏感度不足。现有的检测模型主要针对自然图像设计，难以适应焊缝X射线图像的特征挑战，如高噪声、低对比度以及缺陷间的相似性，尤其容易导致对小缺陷的漏检和误报。为应对这些挑战，提出了一种多维度特征融合模型（MADet），它是用于焊缝缺陷检测的多分支深度融合网络。该框架包含两项关键创新：（1）一个多尺度特征融合网络，集成了轻量级注意力残差模块，通过利用低级纹理信息增强对细粒度缺陷特征的感知。（2）一个基于锚点的特征选择检测头，用于提高对五种典型缺陷类别的辨别和定位精度。在公共和专有焊缝缺陷数据集上进行的大量实验表明，MADet比最先进的YOLO变体有显著改进。具体而言，它在mAP@0.5上比次优模型高出7.41%，表明具有很强的工业适用性。
