Explore the vast range of titles available.

The paper presents a methodology for training neural networks for vision tasks on synthesized data on the example of steel defect recognition in automated production control systems. The article describes the process of dataset procedural generation of steel slab defects with a symmetrical distribution. The results of training two neural networks Unet and Xception on a generated data grid and testing them on real data are presented. The performance of these neural networks was assessed using real data from the Severstal: Steel Defect Detection set. In both cases, the neural networks showed good results in the classification and segmentation of surface defects of steel workpieces in the image. Dice score on synthetic data reaches 0.62, and accuracy—0.81.

Surface defects are a common issue that affects product quality in the industrial manufacturing process. Many companies put a lot of effort into developing automated inspection systems to handle this issue. In this work, we propose a novel deep learning–based surface defect inspection system called the forceful steel defect detector (FDD), especially for steel surface defect detection. Our model adopts the state-of-the-art cascade R-CNN as the baseline architecture and improves it with the deformable convolution and the deformable RoI pooling to adapt to the geometric shape of defects. Besides, our model adopts the guided anchoring region proposal to generate bounding boxes with higher accuracies. Moreover, to enrich the point of view of input images, we propose the random scaling and the ultimate scaling techniques in the training and inference process, respectively. The experimental studies on the Severstal steel dataset, NEU steel dataset, and DAGM dataset demonstrate that our proposed model effectively improved the detection accuracy in terms of the average recall (AR) and the mean average precision (mAP) compared to state-of-the-art defect detection methods. We expect our innovation to accelerate the automation of industrial manufacturing process by increasing the productivity and by sustaining high product qualities.

To address the challenges of steel surface defect detection—characterized by fine-grained textures, substantial scale variations, and complex background interference—conventional lightweight detectors often struggle to balance real-time navigation requirements with high-precision spatial localization on mobile inspection platforms. In this work, we propose KDM-YOLO, a lightweight visual localization and detection method built upon YOLOv10n, designed to provide an efficient perception engine for autonomous inspection robots. The proposed approach enhances the baseline through three key perspectives: feature extraction, context modeling, and multi-scale fusion. Specifically, KWConv is introduced to strengthen the representation of fine-grained texture and edge cues; C2f-DRB is employed to enlarge the effective receptive field and improve long-range dependency perception to reduce missed detections; and a multi-scale attention fusion (MSAF) module is inserted before the detection head to adaptively integrate spatial details with semantic context while suppressing redundant background responses. Ablation studies confirm that each module contributes to performance gains, and their combination yields the best overall results. Comparative experiments further demonstrate that KDM-YOLO significantly improves detection performance while retaining a compact model size and high inference speed. Compared with the YOLOv10n baseline, Precision, Recall and mAP@50 are increased to 91.0%, 93.9%, and 95.4%, respectively, with a parameter count of 3.29 M and an inference speed of 155.6 f/s. These results indicate that KDM-YOLO achieves an ideal balance between the accuracy and computational efficiency required for embedded navigation platforms, providing an effective solution for online autonomous inspection and real-time localization of steel surface defects.

Scratches and cracks in steel severely affect its service life and performance. However, owing to the irregular shapes and sizes of steel surface defects, defects within the same class may be different, whereas defects between classes may be similar. Existing methods focus only on spatial information, resulting in low detection accuracy. To alleviate these problems, this paper proposes the ECDY (EIFEM CARAFE DyHead) network to enhance the detection capability of steel defects. We first design a feature extraction module that focuses on the edge information of feature contours. This module uses the Sobel operator to extract the edge information of a feature and fuses it with the overall spatial information so that richer semantic information can be obtained. The module has improved accuracy in the YOLOv5, YOLOv8, and YOLOv10 versions, and uses fewer parameters and calculations. In particular, in YOLOv8x, mAP@0.5 increased by 2.5%, and the number of parameters was reduced by 12.4 M. Second, to retain the detailed information in the feature pyramid, and to better reconstruct features, we choose the content-aware reassembly feature method (CARAFE) as the upsampling method. Finally, the detection head was replaced with a dynamic unified detection head (DyHead) to adapt to different defect sizes and different task requirements. Compared with YOLOv8s, the proposed method improves precision by 1.6%, recall by 4%, and mAP@0.5 by 4%. This value is 4.2% higher than the mAP@0.5 of the current SOTA model RT-DETR-L in the field of object detection and has 23.2 M fewer parameters.

With the increasing popularity of deep learning, enterprises are replacing traditional inefficient and non-robust defect detection methods with intelligent recognition technology. This paper utilizes TL (transfer learning) to enhance the model’s recognition performance by integrating the Adam optimizer and a learning rate decay strategy. By comparing the TL-ResNet50 model with other classic CNN models (ResNet50, VGG19, and AlexNet), the superiority of the model used in this paper was fully demonstrated. To address the current lack of understanding regarding the internal mechanisms of CNN models, we employed an interpretable algorithm to analyze pre-trained models and visualize the learned semantic features of defects across various models. This further confirms the efficacy and reliability of CNN models in accurately recognizing different types of defects. Results showed that the TL-ResNet50 model achieved an overall accuracy of 99.4% on the testing set and demonstrated good identification ability for defect features.

Surface defects during steel production can severely impact product quality and safety, making defect detection crucial. To improve the precision and performance of conventional approaches, we introduce FMV-YOLO, a model for detecting steel surface defects, built upon YOLOv11n. First, we substitute the C2PSA attention module in the backbone network with an Adaptive Fine-Grained Channel Attention (FCA) module, which improves defect type identification while reducing the parameter count. Next, we incorporate a new Multi-Scale Attention Fusion module (MSAF) to strengthen feature representation and refine the loss function using Normalized Wasserstein Distance (NWD) loss, thereby improving the localization accuracy of small defects. Finally, we integrate the VoV-GSCSP module within the neck network to achieve lightweighting, facilitating real-world deployment. Extensive experiments on the GC10DET and NEU-DET datasets demonstrate that the model effectively balances detection accuracy, parameter count, and computational load. With 2.6M parameters and 5.7G FLOPs, the model attains an mAP@0.5 of 73.4% on GC10DET and 80.2% on NEU-DET. Additionally, the method achieves 99% detection accuracy on a self-constructed industrial dataset, proving its effectiveness in industrial defect detection.

Steel surface defect detection is challenging because it contains various atypical defects. Many studies have attempted to detect metal surface defects using deep learning and had success in applying deep learning. Despite many previous studies to solve the steel surface defect detection, it remains a difficult problem. To resolve the atypical defects problem, we introduce a hierarchical approach for the classification and detection of defects on the steel surface. The proposed approach has a hierarchical structure of the binary classifier at the first stage and the object detection and semantic segmentation algorithms at the second stage. It shows 98.6% accuracy in scratch and other types of defect classification and 77.12% mean average precision (mAP) in defect detection using the Northeastern University (NEU) surface defect detection dataset. A comparative analysis with the previous studies shows that the proposed approach achieves excellent results on the NEU dataset.

In a complex industrial environment, it is difficult to obtain hot rolled strip steel surface defect images. Moreover, there is a lack of effective identification methods. In response to this, this paper implements accurate classification of strip steel surface defects based on generative adversarial network and attention mechanism. Firstly, a novel WGAN model is proposed to generate new surface defect images from random noises. By expanding the number of samples from 1360 to 3773, the generated images can be further used for training classification algorithm. Secondly, a Multi-SE-ResNet34 model integrating attention mechanism is proposed to identify defects. The accuracy rate on the test set is 99.20%, which is 6.71%, 4.56%, 1.88%, 0.54% and 1.34% higher than AlexNet, VGG16, ShuffleNet v2 1×, ResNet34, and ResNet50, respectively. Finally, a visual comparison of the features extracted by different models using Grad-CAM reveals that the proposed model is more calibrated for feature extraction. Therefore, it can be concluded that the proposed methods provide a significant reference for data augmentation and classification of strip steel surface defects.

Recognizing defects on the steel plate surface has great application potential in the steel manufacturing process. However, it is still challenging to accurately recognize surface defects since most defects only occupy a small area of the whole image and have high similarities to the surrounding backgrounds. To solve the above issues, we propose an attention multi-hierarchical feature fusion network (AMHNet) to recognize defects. First, to better fuse the features from different levels, we propose a skipping attention module to selectively transfer informative features in low-level layers into high-level layers based on the convolutional block attention mechanism. Second, to dynamically fuse multi-hierarchical features, we propose a feature dynamic aggregation gate by gating mechanism to enhance defect-relevant features and suppress useless features. Finally, to verify the effectiveness and advantages of our model, we also collect a new challenging defect recognition dataset called NPU-DRD. Extensive experiments on dataset NPU-DRD show that our AMHNet achieves an accuracy of 97.58% and an AUC score of 97.23%, which are the new state-of-the-art results among existing methods. Our new dataset and source codes are available at https://github.com/Heisenberg828/AMHNet .

Detecting surface defects in steel is essential for ensuring structural safety and manufacturing efficiency. However, existing detection systems often struggle to accurately identify small flaws, handle complex surface conditions, and maintain real-time performance. To overcome these challenges, this study proposes ASFRW-YOLO, an enhanced version of the YOLOv5 framework. The model integrates a multi-scale ASF module to improve sensitivity to minute defects, replaces the conventional C3 module with RepNCSPELAN4 to strengthen feature representation, and adopts the Wise-IoU loss with adaptive weighting to refine bounding box regression. Experiments were conducted on the NEU-DET dataset, which was divided into training, validation, and testing sets with an 8:1:1 ratio. The proposed method achieved a mean Average Precision of 83.2% at an IoU threshold of 0.5 and 46.4% across IoU thresholds from 0.5 to 0.95, representing an improvement of approximately seven percentage points compared with YOLOv5s. Moreover, the model maintains a lightweight design with only 6.20 million parameters and processes 640 × 640 input images at about 125 frames per second on an RTX 4060 Laptop GPU (8 GB VRAM). These results demonstrate that ASFRW-YOLO effectively balances detection accuracy, computational efficiency, and model compactness, making it highly suitable for real-time industrial defect inspection.