Feature Decoupling-Guided Annotation Framework for Surface Defects on Steel Strips

Surface defect detection is a core component for ensuring the quality of multi-wedge belt products, yet traditional detection methods based on Convolutional Neural Networks (CNNs) suffer from weak global feature extraction and limited multi-scale feature fusion, severely impacting detection efficiency. Meanwhile, the constrained computational resources in industrial scenarios pose critical challenges to balancing model accuracy and speed. To address these issues, we propose a lightweight defect detection model, LCT-YOLO (Lightweight CNN-Transformer YOLO), which integrates the technical advantages of CNN and Transformer while meeting the lightweight deployment requirements of industrial applications. Specifically, a dynamic channel selection feature extraction network is constructed by introducing the partial channel selection module CSP_PTB to enhance the model’s capability of capturing long-distance semantic information. A lightweight multi-scale feature fusion network is developed by designing a multi-scale path aggregation network and integrating the dynamic upsampling module Dysample in the model’s neck, effectively addressing challenges from feature scale variations. A lightweight shared convolution detection head is introduced to tackle the high computational cost of existing model detection heads, optimizing performance and reducing model complexity through parameter sharing strategies. Finally, model pruning is employed to remove redundant parameters for compression, and knowledge distillation is utilized to compensate for accuracy loss. Experimental results show that the algorithm achieves an mAP50 of 94.3% on the dataset, with 0.8 M parameters and 2.0 GFLOPs computation, outperforming the baseline model by 0.9% in mAP50, reducing parameters by 74.2%, computation by 75.3%, and featuring a model size of only 1.9 M, making it highly suitable for edge device deployment. The algorithm surpasses other advanced methods in both accuracy and speed, not only meeting the dual requirements of detection precision and real-time performance in industrial scenarios but also being successfully applied to practical production, providing robust technical support for defect detection and product quality control.

Fabric surface defects can significantly affect the appearance and performance of textile products, often resulting in reduced product quality and economic loss. As a key step in non-destructive quality control, fabric defect detection requires both high accuracy and real-time performance under complex industrial environments. In this study, we present a visual inspection framework based on an improved YOLOv11 architecture for the non-destructive detection of fabric surface defects. A custom image acquisition system, integrating a high-resolution industrial camera and ring-shaped LED lighting, is designed to enhance defect edge contrast and provide high-quality input data. The proposed network incorporates a C3k2-DSConv module with dynamic snake convolution to better capture irregular defect geometries at multiple scales, and introduces a lightweight Slim-Neck structure with GSConv and VoV-GSCSP for efficient feature fusion. Additionally, the WIoU loss function is adopted to stabilise training and improve localisation accuracy. Experimental results on the Cloth_dataset demonstrate a mAP of 96.1%, outperforming existing detectors in both speed and accuracy. This approach provides a practical and efficient solution for real-time, non-destructive defect detection in textile manufacturing lines.

Production defects caused by irresistible factors such as process design problems or differences in steel properties in strip production affect the economic benefits of the enterprise and threaten production safety. Traditional defect detection methods are difficult to achieve real-time and high-precision detection, so developing surface defect detection methods based on deep learning is of great significance for strip production. In order to effectively improve the accuracy of the deep learning model in detecting surface defects on hot-rolled strip, in this work we propose a real-time detection model for surface defects on strip steel based on the YOLOv8n model. Firstly, the newly convolutional layer Con5v is designed to replace the original convolutional layer in the neck, and an attention mechanism is added in front of each Con5v to improve the algorithm’s ability to extract small target information. Secondly, an additional set of upsampled feature extraction units is added to the neck in order to enhance the spatial information of the feature map. Subsequently, a set of feature fusion units is incorporated and the convolutional layers in it are improved to provide better feature maps. Thirdly, the number of decoupling detection heads is increased to receive more high-quality features. The final experimental results show that YOLOv8n-GAM (YOLOv8 Nano Model with Global Attention Mechanism) achieves 81.4mAP and 82.0FPS on the NEU-DET dataset and 71.2mAP and 55.0FPS on the GC10-DET dataset, which are 5.7% and 6.9% higher than those of YOLOv8n, respectively. The model proposed in this paper achieves a comprehensive performance improvement in strip steel.

Internal defects detection and classification in hollow cylindrical surfaces using single shot detection and MobileNet

The detection of defects on automotive coated surfaces is of paramount importance to ensure the quality of automotive appearance. However, due to the challenges posed by the lack of sufficient samples of product-specific coated defects, the uneven number of species, and the difficulty of detecting defects in small targets due to the complex background of the body surface. The detection of defects on automotive coated surfaces still relies on experienced manual labor to a significant extent. To address these issues, Poisson fusion is employed to enhance the data for generating defect images, thereby mitigating the limitations of insufficient coated defect samples and an imbalanced distribution of defect species. Based on this, a new YOLOv8-SC defect detection model is proposed, in which the C2f-Star is used to replace the original C2f structure. This aims to improve the model’s ability to extract defects of varying sizes in complex backgrounds, while reducing the number of model parameters and the amount of computation. The content-guided attention fusion (CGAFusion) module, situated before each detection head of the model, has been designed to enhance the model’s capacity to extract defects of varying sizes and complexity. This is achieved through the adaptive fusion of low-level and high-level features. The proposed model’s superiority in terms of detection metrics has been demonstrated through the use of example validation. The experiments were conducted on the enterprise dataset CPD-DET and the public dataset NEU-DET. The results demonstrated that the defective image generation and data enhancement method could significantly improve detection performance and have good generalization. The proposed YOLOv8-SC reduces the model parameters by 12.2% compared to the normal model. Additionally, the mean average precision (mAP) at 50% and 50%–95% thresholds are enhanced by 4.7% and 2.3%, respectively. Furthermore, the accuracy of the model exhibits a discernible positive growth trend with an increase in the sample set size. The design of an automotive coated surface defect detection system is presented at last. When deployed in industrial settings, this system can facilitate the intelligent detection of coated defects.

Surface defect inspection methods have proven effective in addressing casting quality control tasks. However, traditional inspection methods often struggle to achieve high-precision detection of surface defects in castings with similar characteristics and minor scales. The study introduces DES-YOLO, a novel real-time method for detecting castings' surface defects. In the DES-YOLO model, we incorporate the DSC-Darknet backbone network and global attention mechanism (GAM) module to enhance the identification of defect target features. These additions are essential for overcoming the challenge posed by the high similarity among defect characteristics, such as shrinkage holes and slag holes, which can result in decreased detection accuracy. An enhanced pyramid pooling module is also introduced to improve feature representation for small defective parts through multi-layer pooling. We integrate Slim-Neck and SIoU bounding box regression loss functions for real-time detection in actual production scenarios. These functions reduce memory overhead and enable real-time detection of surface defects in castings. Experimental findings demonstrate that the DES-YOLO model achieves a mean average precision (mAP) of 92.6% on the CSD-DET dataset and a single-image inference speed of 3.9 milliseconds. The proposed method proves capable of swiftly and accurately accomplishing real-time detection of surface defects in castings.

Deep learning methods have made remarkable strides in surface defect detection. But, they heavily rely on large amount of training data, which can be a costly endeavor, especially for specific applications like steel strip surface defect detection, where acquiring and labeling large-scale data is impractical due to the rarity of certain defective categories in production environment. Hence, realistic defect image synthesis can greatly alleviate this issue. However, training image generation networks also demand substantial data, making image data augmentation merely an auxiliary effort. In this work, we propose a Generative Adversarial Network (GAN)-based image synthesis framework. We selectively extract the defect edges of the original image as well as the background texture information, and use them as network input through the spatially-adaptive (de)normalization (SPADE) module. This enriches the input information, thus significantly reducing the amount of training data for GAN network in image generation, and enhancing the background details as well as the defect boundaries in the generated images. Additionally, we introduce a novel generator loss term that balances the similarity and perceptual fidelity between synthetic and real images by constraining high-level features at different feature levels. This provides more valuable information for data augmentation in training object detection models using synthetic images. Our experimental results demonstrate the sophistication of the proposed image synthesis method and its effectiveness in data augmentation for steel strip surface defect detection tasks.

The design and study of pulsed eddy current sensors for detecting surface defects in small-diameter rods are highly significant. Accurate detection and identification of surface defects in small-diameter rods may be attained by the ongoing optimization of sensor design and enhancement of detection technologies. This article presents the construction of a non-coaxial differential eddy current sensor (Tx-Rx sensor) and examines the detection of surface defects in a small diameter bar. A COMSOL 3D model is developed to examine the variations in eddy current distribution and defect signal characteristics between the plate and rod components. The position of the excitation coil on the bar and the eddy current disruption around the defect are examined. Additionally, a Tx-Rx sensor has been developed and enhanced concerning coil dimensions, coil separation, and elevation height. An experimental system is established to detect bar structures with surface defects of varying depths, and a model correlating differential signal attenuation with defect depth is proposed, achieving a quantitative relative error of less than 5%, thereby offering a reference for the quantitative detection of bar surface defects.

Detection of surface defects in strip steel is crucial in industrial automation to ensure steel quality. Recently, learning-based detection methods have gained prominence among researchers for their robust capabilities. However, accurately delineating clear area boundaries remains a challenge, particularly in the complex backgrounds of steel surfaces. In this paper, we propose the edge-aware Interactive Refinement Network (EAIRNet) to address this challenge. EAIRNet comprises three core modules: a Multiscale Feature Extractor, a Sal-Edge Interactive Module, and a regional attention guided module (RAGM). Firstly, our network extracts multiscale features using ResNet from the steel surface environment. Subsequently, features from encoder’s stage 3–5 are integrated to derive edge features. The SEIM facilitates interaction between salient and edge features in the decoder stage, thereby enhancing the network’s focus on defect area boundaries and refining detection precision. Additionally, we introduce the RAGM to improve the focus of shallow decoder layers on defect areas. Our EAIRNet’s effectiveness is proven through comprehensive experimental results on the public dataset named SD-Saliency-900, in comparison to state-of-the-art methods.

Carbon fiber reinforced polymer (CFRP) has become an essential structural material in aerospace, new energy, rail transportation, and other fields due to its lightweight and high-strength properties. However, its complex manufacturing processes and demanding service environments can lead to various types of damage on both the surface and inside the material, significantly compromising safety. Existing detection techniques often adopt a step-by-step inspection mode, which results in complex processes and makes it difficult to achieve cross-dimensional correlation analysis of defects, leading to blind spots in structural health assessment. To address these issues, this paper proposes a method for simultaneous detection of surface and subsurface defects in CFRP based on infrared polarization imaging. First, we developed an infrared polarization imaging-based defect detection system and collected a CFRP defect dataset of samples containing various surface and subsurface defects. Then, we proposed a surface and subsurface defect detection network, which integrates extracted thermal diffusion features and polarization features to decouple and enhance surface and subsurface features. Through the constraints of multi-task defect detection, simultaneous detection of surface and subsurface defects is achieved. The experiments on the collected CFRP defect dataset demonstrate the effectiveness of the proposed method.

For the detection of wood surface defects, a convolutional neural network has a low detection efficiency and insufficient generalization ability, so it does not meet the requirements of online detection. Aiming to solve the above problems, the YOLOv3 baseline model, which has the advantage of multi-objective dynamic detection, was improved and applied to the online detection of wood surface defects. To solve the problem of the poor generalization ability of the network, GridMask was used to enhance the data and improve the robustness of the network. In order to solve the problem of the considerable amount of network parameter calculations and insufficient real-time performance, the residual block of the backbone network was changed to a Ghost block structure to achieve a lightweight model. Finally, the confidence loss function of the network was improved to reduce the influence of simple samples and negative samples on model convergence. The experimental results showed that, compared with the original network, the improved algorithm increased the mean average precision by 5.73% and the detection speed was increased to 28 frames per second (an increase of 11), which met the requirements for real-time industrial detection.

Detection of metal surface defects is of critical importance for the reliability and safety of components manufactured in manufacturing processes. In particular, the examination of bullet case mouth surfaces requires the detection of small defects that may affect ballistic performance or cause failures. This study proposes a fully automated image processing framework focused exclusively on the detection of circularity deviations in bullet case mouths. The core objective is to quantify structural irregularities and identify deviations from ideal circular geometry. The proposed methodology involves eight stages: image acquisition under controlled lighting, Otsu-based binarization for foreground segmentation, morphological filtering to remove noise and artifacts, and geometric feature extraction via region-based statistics. The methodology continues with a combination of classical and advanced shape analysis tools: Hough Circle Transform is used to detect inner and outer rings of the case mouth, followed by precise centroid deviation analysis to measure concentricity. The framework also incorporates radius difference checks and circular variance computation to assess uniformity of radial symmetry. The proposed system was applied on a dataset consisting of 200 images obtained under controlled illumination and 96.5% classification accuracy was achieved. By focusing on continuous shape metrics and model-free inference, the proposed approach offers a reliable foundation for integrating visual quality control into high-throughput manufacturing pipelines.

Additive manufacturing (AM) is widely used in industries such as aerospace, medical, and automotive. Within this domain, defect detection technology has emerged as a critical area of research focus in the quality inspection phase of AM. The main challenge lies in that under extreme lighting conditions, strong reflected light obscures defect feature information, leading to a significant decrease in the defect detection rate. This paper introduces a novel methodology for intelligent defect detection in AM components with reflective surfaces, leveraging virtual polarization filtering (IEVPF) and an improved YOLO V5-W model. The IEVPF algorithm is designed to enhance image quality through the virtual manipulation of light polarization, thereby improving defect visibility. The YOLO V5-W model, integrated with CBAM attention, DenseNet connections, and an EIoU loss function, demonstrates superior performance in defect identification across various lighting conditions. Experiments show a 40.3% reduction in loss, a 10.8% improvement in precision, a 10.3% improvement in recall, and a 13.7% improvement in mAP compared to the original YOLO V5 model. Our findings highlight the potential of combining virtual polarization filtering with advanced deep learning models for enhanced AM surface defect detection.

Sanitary ceramic products, such as toilet and wash basin, are widely used in our daily life. Sanitary ceramics are expected to have some excellent physical properties, such as corrosion resistance, easy cleaning, and low water absorption. However, surface defects in sanitary ceramics are inevitable due to complex production processes and changing production environment. Therefore, surface defect detection must be performed in the manufacturing process of sanitary ceramics. There are many types of surface defects in sanitary ceramics, and different types of defects have large differences in characteristics and scales. Traditional detection methods with artificially designed features and classifiers are difficult to apply in this context. In addition, there are few studies on surface defect detection methods of sanitary ceramics based on deep neural networks. In this article, a lightweight real-time defect detection network based on the lightweight backbone MobileNetV3 is presented. The proposed network achieves multi-scale detection of surface defects in sanitary ceramics with a multi-layer feature pyramid. Combining region proposal network and anchor-free method, real-time defect detection is achieved. Finally, a detection head with channel attention structure and a low-level mixed feature classification strategy is used to perform defect classification with higher accuracy. Experimental results show that the proposed approach achieves at least 22.9% detection speed improvement and 35.0% average precision improvement while reducing memory consumption by at least 8.4% compared with the classic one-stage SSD, YOLO V3 and two-stage Faster R-CNN methods.

Surface defect detection aims to classify and locate a certain defect that exists in the target surface area. It is an important part of industrial quality inspection. Most of the research on surface defect detection are currently based on convolutional neural networks (CNNs), which are more concerned with local information and lack global perception. Thus, CNNs are unable to effectively extract the defect features. In this paper, a defect detection method based on the Swin transformer is proposed. The structure of the Swin transformer has been fine-tuned so that it has five scales of output, making it more suitable for defect detection tasks with large variations in target size. A bi-directional feature pyramid network is used as the feature fusion part to efficiently fuse to the extracted features. The focal loss is used as a loss function to weight the hard- and easy-to-distinguish samples, potentially making the model fit the surface defect data better. To reduce the number of parameters in the model, a shared detection head was chosen for result prediction. Experiments were conducted on the flange surface defect dataset and the steel surface defect dataset, respectively. Compared with the classical CNNs target detection algorithm, our method improves the mean average precision (mAP) by about 15.4%, while the model volume and detection speed are essentially the same as those of the CNNs-based method. The experimental results show that our proposed method is more competitive compared with CNNs-based methods and has some generality for different types of defects.