MADet: A Multi-Dimensional Feature Fusion Model for Detecting Typical Defects in Weld Radiographs

TL;DR

MADet is a new model for detecting weld defects in X-ray images that improves accuracy and reduces errors compared to existing methods.

Contribution

MADet introduces a multi-branch deep fusion network with attention modules and a feature-selective detection head for weld defect detection.

Findings

MADet outperformed state-of-the-art YOLO variants by 7.41% in mAP@0.5.

The model effectively handles challenges like noise, low contrast, and small defect detection in weld radiographs.

Experiments on public and proprietary datasets confirmed its strong industrial applicability.

Abstract

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.

Figures