Welding is a key technical means to ensure the safe and stable connection between pipeline systems. The intelligent welding defect assessment based on X-ray inspection helps to improve the efficiency and accuracy of pipeline welding fault diagnosis. However, this work is highly challenging and systematic, and includes multiple critical subtasks, such as weld area extraction and defect detection. The existing methods have fallen into the quagmire of various task flow concatenation, resulting in unsatisfactory accuracy and efficiency. To overcome this dilemma, a multiexpert inspection method based on X-ray is proposed to perform welding defect assessment. First, a preprocessing method based on window width and window level adjustment is designed to calibrate original data to ensure high-quality input of the subsequent neural network. Second, a basic network layer with universal representation ability initialization is designed to extract discriminative multitask feature representations. Next, a novel task-specific experts is presented to perform weld area extraction and defect detection in a divide and conquer manner. Finally, a rule-based defect postprocessing is designed to evaluate the level of weld defects. In the experiment, the performance of defect detection expert and weld seam segmentation expert based on hardware testing platform is systematically verified and discussed. In practical defect assessment application cases, we obtained competitive results (recall: 98.5%, precision: 95.2%, accuracy: 97.4%). Hence, the proposed intelligent multiexpert method has shown great potential in actual pipeline weld maintenance and significantly saves operators time and costs.

At present, the method based on deep learning performs well in public object detection tasks. However, there are still two problems to be solved for industrial defect detection: 1) Industrial scenes requires real-time and lightweight; 2) Lightweight network accuracy is limited. In order to tackle these issues, based on knowledge distillation, this paper proposes an effective lightweight defect detection network (KD-LightNet) suitable for edge scene. First of all, a lightweight network (LightNet) is designed based on structure reparameterization, which can sufficiently improve the capability of network feature extraction and reduce the complexity of model inferring. Moreover, a well prepared self-distillation strategy is proposed, which utilize the pre-trained LightNet network as a teacher model to transfer knowledge in the same structure. Then, in order to fully utilize the logits predicted by teacher model, an improved KL divergence loss is proposed to enhance the accuracy of the student model. Finally, in the experiments, three industrial datasets (PKU-Market-PCB, NEU-DET and Magnetic tile defect dataset) were used to validate the proposed model performance. The KD-LightNet detection accuracy (mAP) is improved by an average of 6.87%, while the average detection speed reaches 72 FPS @3070Ti (Params 4.7M), which meets the requirements of industrial defect detection accuracy and real-time.

In the welding process of long-distance pipelines, cracks pose significant risks and can cause severe damages. Intelligent detection of hazardous defects in pipeline welds, with a focus on complex crack features with different shapes and scales, is a necessary but challenging task. Current deep learning-based visual detection models exhibit weak generalization and transfer capabilities when it comes to complex defect features. To address these challenges, this paper proposes a spatial transformation-based multi-scale attention network (STMA-Net) for weld defect detection with multi-scale and shape-variant features in X-ray images. First, a novel spatial transformation attention network (STAN) is designed to capture the complex deformation features, which enhances the generalization ability of the feature extraction network. Then, in order to better utilize the discriminative features, a multi-level attention feature fusion network (MAFFN) is designed to improve the prediction accuracy of multi-scale defects. Furthermore, a multi-level deep supervision is proposed to better update network parameters and enhance the inference ability of multi-level detection heads. Finally, three types of X-ray weld crack datasets in the real world are collected to guide a series of experiments and analyses. The results show that the generalization and transfer capabilities of the proposed method outperforms other detectors in the detection task of cracks and other defects (AP increased by 6.5%).

Defect detection aims to locate and classify defects in images, which is a necessary yet challenging task in industrial product quality monitoring. The current anchor-based detectors have weak generalization performance due to their inability to consider numerous scale priors. Moreover, the basic networks lack the ability to dynamically capture and utilize multiscale feature representations, resulting in low accuracy in industrial defect detection. To counter these challenges, an efficient anchor-free detector with dynamic receptive field assignment (DRFA) and task alignment is proposed. First, a feature pyramid structure with DRFA is innovatively designed to sufficiently extract multiscale feature representation and flexibly adjust the receptive field to detect diverse defects. Second, a task decoupling prediction mechanism is proposed to improve localization and classification prediction capabilities by introducing feature reassembly and task-specific information enhancers. Next, an anchor-free-based deep supervision with task-aligned is presented to encourage both to make accurate and consistent predictions, thereby effectively improving the overall detection performance. Finally, three industrial defect datasets (NEU-DET, PCB, WELD) are employed for experiments. The results show that the proposed method achieves 5.3% higher average AP than other state-of-the-art detectors.

This article proposes a progressive frequency domain-guided depth model with adaptive preprocessing to solve the problem of defect detection with weak features based on X-ray images. In distinct intuitive surface defect detection tasks, non-destructive testing of castings using X-rays presents more complex and weak defect features, leading to lower accuracy and insufficient robustness on the part of current casting defect detection methods. To address these challenges, the proposed method establishes four specialized mechanisms to improve model accuracy. First, an adaptive image contrast enhancement method is proposed to enhance the features of defects in casting images to promote subsequent feature extraction and prediction. Second, a subtle clue mining module based on frequency domain attention is proposed to fully extract the discriminative features of casting defects. Third, a feature refinement module based on progressive learning is proposed to achieve a balance between feature resolution and semantic information. Finally, a refined deep regression supervision mechanism is designed to improve defect detection accuracy under strict intersection-to-union ratio standards. We established extensive ablation studies using casting defect images in GDXray, conducted detailed comparative experiments with other methods, and performed experiments to analyze the robustness of the resulting models. Compared with other X-ray defect detection methods, our framework achieves an average +4.6 AP. Compared to the baseline, our proposed refined deep regression supervision mechanism results in an improvement of 5.3 AP.

Automatic welding technology has been widely applied in the field of industrial production. In promoting high-quality intelligent manufacturing processes for the consumer market, precise detection of welding defects is a key link in enhancing product reliability. Considering the high demand for domain knowledge in welding defect inspection tasks, existing methods have shortcomings in fine-grained recognition. To overcome these weaknesses, this article proposes a dual-expert detection method for automatic welding defect inspection, inspired by the human multi-expert complementary evaluation process. The method process is mainly reflected in the following aspects. Firstly, a high-precision robot welding defect inspection system that integrates application platform based data collection, model training, and deployment has been effectively implemented. Then, diverse detection experts are constructed using complementary regression paradigms to obtain predictions from multiple perspectives. Finally, a dual-expert fusion scheme based on probabilistic ensembling is designed to fully integrate prediction information and obtain refined results. In the experiment, a database rich in diverse types of welding defects is applied for training and systematic testing, and multiple sets of refined indicators are evaluated. In addition, we also discussed edge deployment, failure case analysis, and the application of in consumer PCB defect detection. These practical application results indicate that the proposed method can achieve product level implementation, effectively reducing the threshold for industrial testing and enabling end-users to gain benefits.