TRDM: A Two-Stage Real-Time Discrimination Method for Spiral Weld Defects Under Dynamic Distorted Imaging

Non-destructive testing of weld based on radiographic images is crucial for quality control of spiral steel pipes. However, accurately discriminating defect images from high-resolution and low-contrast weld images remains a challenging task. Moreover, the real-time imaging of radiography in industrial settings introduces dynamic distorted interferences to weld images, which has received limited attention in existing research. To address these issues, we propose a two-stage real-time discrimination method (TRDM) consisting of two carefully designed networks: an extremely lightweight weld region segmentation network and a semantic reconstruction-based discrimination network. Our method aims to adaptively reduce interference from invalid regions in the weld images to better exploit defect features from a semantic perspective and reconstruct them as defect-free regions as much as possible. To overcome the scarcity of weld defect image data, we construct two publicly available industrial weld datasets, including an image dataset and a video dataset. TRDM achieves competitive performance with state-of-the-art techniques, demonstrating its effectiveness in discriminating weld images. Furthermore, we integrate the TRDM into developed detection software for defect assessment, validating its efficacy in real welding scenarios. The source code and dataset are publicly available at https://github.com/Weld-det/TRDM Note to Practitioners—The motivation of this work is to address the detection challenges posed by dynamic distorted imaging of spiral welds in real-time operating conditions. Currently, manual judgment by professionals remains the predominant approach in spiral welding nondestructive testing processes, but it suffers from various limitations. This study aims to assist professionals in making real-time decisions for weld defect identification, thereby reducing missed detections and false alarms. To the best of our knowledge, there has been limited research attention given to this area, which represents a significant challenge. Therefore, this paper constructs a two-stage real-time discrimination network for screening spiral weld defect images. TRDM only requires training with normal defect-free weld images, which is particularly applicable to spiral welding scenarios with extremely scarce defect images. The trained model demonstrates outstanding real-time detection performance on both static weld images and dynamic distorted weld images, showcasing significant potential for practical applications in spiral welding inspection systems.