An End-to-End Steel Surface Defect Detection Approach via Fusing Multiple Hierarchical Features
Open accessThe DOI matches its Crossref record; 43 of the 53 references it cites were found in a registry or an index.
Abstract
A complete defect detection task aims to achieve the specific class and precise location of each defect in an image, which makes it still challenging for applying this task in practice. The defect detection is a composite task of classification and location, leading to related methods is often hard to take into account the accuracy of both. The implementation of defect detection depends on a special detection data set that contains expensive manual annotations. In this paper, we proposed a novel defect detection system based on deep learning and focused on a practical industrial application: steel plate defect inspection. In order to achieve strong classification ability, this system employs a baseline convolution neural network (CNN) to generate feature maps at each stage, and then the proposed multilevel feature fusion network (MFN) combines multiple hierarchical features into one feature, which can include more location details of defects. Based on these multilevel features, a region proposal network (RPN) is adopted to generate regions of interest (ROIs). For each ROI, a detector, consisting of a classifier and a bounding box regressor, produces the final detection results. Finally, we set up a defect detection data set NEU-DET for training and evaluating our method. On the NEU-DET, our method achieves 74.8/82.3 mAP with baseline networks ResNet34/50 by using 300 proposals. In addition, by using only 50 proposals, our method can detect at 20 ft/s on a single GPU and reach…
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“There are some shortcomings in manual testing, such as low confidence and high labor intensity.”
From A New Steel Defect Detection Algorithm Based on Deep Learning (2021) · cites this paper as [5]
In the citing paper
“most classification tasks can only find the defects with the highest confidence level in the defect category, resulting in imprecise classification results [3].”
From Steel Surface Defect Recognition: A Survey (2022) · cites this paper as [3]
In the citing paper
“there are many types of steel surface defects and some of these defects may overlap”
From Steel Surface Defect Recognition: A Survey (2022) · cites this paper as [3]
In the citing paper
“Using manual inspection of steel surface defects, it is difficult to distinguish the type of defect and accurately detect the location of the defect [2].”
From A deep learning model for steel surface defect detection (2024) · cites this paper as [2]
In the citing paper
“There are many categories of steel surface defects and the size of each defect category varies widely. The average size of each category in the NEU-DET [34] dataset is shown in Table1.”
From A deep learning model for steel surface defect detection (2024) · cites this paper as [34]
1 more claim about this paper
Create a free account to see every claim other papers make about this one.
Citations
Cites
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 1×
“The third type is ResNet using residual blocks to make networks deeper without overfitting [23].”
MethodVery Deep Convolutional Networks for Large-Scale Image Recognition
Simonyan, K., & Zisserman, A. (2014) · arXiv · doi:10.48550/arxiv.1409.1556 · Cited by 114,148
ImageNet classification with deep convolutional neural networks
Krizhevsky, A., Sutskever, I., & Hinton, G. E. (2017) · Communications of the ACM · doi:10.1145/3065386 · Cited by 112,155 · cited 1×
“cessful networks are based on the sequential pipeline architecture [25], which establish the basic structure of CNN and prove the importance of depth of networks.”
BackgroundLeCun, Y., Bengio, Y., & Hinton, G. (2015) · Nature · doi:10.1038/nature14539 · Cited by 84,126
Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2017) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2016.2577031 · Cited by 73,472 · cited 6×
“Next, RPN [14] is employed to generate region proposals”
BackgroundMethodInfluentialFaster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
Ren, S., He, K., Girshick, R., & Sun, J. (2015) · arXiv · doi:10.48550/arxiv.1506.01497 · Cited by 70,812 · cited 6×
“Next, RPN [14] is employed to generate region proposals”
BackgroundMethodInfluentialImageNet: A large-scale hierarchical image database
Deng, J., Dong, W., Socher, R., Li, L.-J., Li, K., & Fei-Fei, L. (2009) · 2009 IEEE Conference on Computer Vision and Pattern Recognition · doi:10.1109/cvpr.2009.5206848 · Cited by 65,333 · cited 1×
“As current research in transfer learning [15], the key to drive large networks is pretraining on ImageNet [22].”
BackgroundGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 1×
“combined? There are two essential conditions: nonadjacent, because adjacent layers have highly local correlation [32], and”
BackgroundYou Only Look Once: Unified, Real-Time Object Detection
Redmon, J., Divvala, S., Girshick, R., & Farhadi, A. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.91 · Cited by 53,803 · cited 1×
“The representative direct regression methods are YOLO [29] and SSD [30].”
MethodFully convolutional networks for semantic segmentation
Long, J., Shelhamer, E., & Darrell, T. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298965 · Cited by 38,834 · cited 2×
“The most famous region-based detectors are the “R-CNN family” [27] [28] [14].”
BackgroundMethod
Cited by
Items 201–300 of 1,338
Ibrahim, A. A. M. S., & Tapamo, J. (2024) · Scientific African · doi:10.1016/j.sciaf.2024.e02066 · Cited by 51
Liu, P., Yuan, X., Han, Q., et al. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.108075 · Cited by 32
Liu, Y., Xu, M., Xiao, T., Tang, H., Hu, Y., & Nie, L. (2024) · IEEE Transactions on Geoscience and Remote Sensing · doi:10.1109/tgrs.2024.3439401 · Cited by 55
“DigitalObjectIdentifier 10.1109/TGRS.2024.3439401 defect detection [9], [10], image quality assessment [11], [12], and visual tracking [13].”
BackgroundInfluentialA Novel Pixel-Wise Defect Inspection Method Based on Stable Background Reconstruction
Lv, C., Shen, F., Zhang, Z., Xu, D., & He, Y. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3038413 · Cited by 40
“based method, it has been widely adopted in the field of object detection [5], [6].”
MethodSelf-citationSMD Anomaly Detection: A Self-Supervised Texture–Structure Anomaly Detection Framework
Luo, J., Lin, J., Yang, Z., & Liu, H. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3194920 · Cited by 45
Steel Surface Defect Detection Algorithm Based on Improved YOLOv8n
Zhang, T., Pan, P., Zhang, J., & Zhang, X. (2024) · Applied Sciences · doi:10.3390/app14125325 · Cited by 33
“He [22] suggested that an MFN network can effectively improve the ability to obtain information about steel surface defects, and its features are able to integrate information from various levels to detect steel surface defects.”
BackgroundWang, K., Zhang, J., Ni, H., & Ren, F. (2021) · Electronics · doi:10.3390/electronics10161986 · Cited by 31
“RoI [34,35].”
BackgroundCASPPNet: a chained atrous spatial pyramid pooling network for steel defect detection
Zheng, Z., Hu, Y., Zhang, Y., Yang, H., Qiao, Y., Qu, Z., & Huang, Y. (2022) · Measurement Science and Technology · doi:10.1088/1361-6501/ac68d2 · Cited by 31
“Subsequently, a multilevel feature fusion network was designed to integrate multiple hierarchical features into one for rich feature information, which achieved a mAP of 82.3% on NEU-DET datasets [15].”
InfluentialThe Role of Deep Learning in Manufacturing Applications: Challenges and Opportunities
Malhan, R. K., & Gupta, S. K. (2023) · Journal of Computing and Information Science in Engineering · doi:10.1115/1.4062939 · Cited by 38
Deep learning-based automated steel surface defect segmentation: a comparative experimental study
Sime, D. M., Wang, G., Zeng, Z. P., & Peng, B. (2023) · Multimedia Tools and Applications · doi:10.1007/s11042-023-15307-y · Cited by 38
“Automated inspection can be categorized as image-level classification [13, 15, 48], region-based defect detection [3, 16, 18, 23, 26], and defect segmentation [8, 10, 20, 35, 43, 56].”
An Efficient Anchor-Free Defect Detector With Dynamic Receptive Field and Task Alignment
Zuo, F., Liu, J., Fu, M., Wang, L., & Zhao, Z. M. (2024) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2024.3371982 · Cited by 35
“They have been applied to defect detection tasks to alleviate the drawbacks of traditional methods [19], [20], [21].”
MethodYOLO-ADS: An Improved YOLOv8 Algorithm for Metal Surface Defect Detection
Gui, Z., & Geng, J. (2024) · Electronics · doi:10.3390/electronics13163129 · Cited by 37
Wu, X., Duan, J., Yang, L., & Duan, S. (2024) · Computers in Industry · doi:10.1016/j.compind.2024.104146 · Cited by 35
A survey on deep learning fundamentals
Tian, C., Cheng, T., Peng, Z., Zuo, W., Tian, Y., Zhang, Q., Wang, F., & Zhang, D. D. (2025) · Artificial Intelligence Review · doi:10.1007/s10462-025-11368-7 · Cited by 53
Nahar, L., Awrangjeb, M., & Islam, M. S. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.104067 · Cited by 48
Automatic and Accurate Measurement of Microhardness Profile Based on Image Processing
Zhao, Y., Xu, W., Xi, C. Z., et al. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3067191 · Cited by 30
“The successful applications can include the measurements of particle sizes in abrasive machining [3], machined surface roughness [4], [5] and defects [6], [7], geometry measurement in both subtractive [8], and additive manufacturing [9], [10].”
Joint Attention-Guided Feature Fusion Network for Saliency Detection of Surface Defects
Jiang, X., Yan, F., Lu, Y., Wang, K., Guo, S., Zhang, T., Pang, Y., Niu, J., & Xu, M. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3218547 · Cited by 48
“challenges, researchers have designed CNN-based models that target surface defect inspection, such as the region-level defect inspection methods [15], [16], [17], [18], [19] and the pixel-level ones [20], [21], [22], [23], [24].”
BackgroundSingh, A. R., Bashford‐Rogers, T., Marnerides, D., Debattista, K., & Hazra, S. (2023) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-022-10763-6 · Cited by 37
“Results showed the proposed method surpassed Swin transformer and other CNN-based state-of-the-art models on SeverstalSteel [28] and NEU-DET [29] datasets.”
MethodLiu, Y., Gao, X., Wen, J. Z., & Luo, H. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3271754 · Cited by 50
“D EFECT detection [1] is one of the important technologies to ensure product quality, which aims to find the appearance defects of various industrial products.”
BackgroundCDDNet: Camouflaged Defect Detection Network for Steel Surface
Luo, Q., Li, B., Su, J., Yang, C., Gui, W., Sílven, O., & Liu, L. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3336452 · Cited by 45
“How to achieve accurate surface defect detection for an automatic visual inspection instrument under such a harsh environment is a key prerequisite to realize timely quality controlling of end steel products [2], [3], [4].”
BackgroundMethodZhong, H., Xiao, L., Wang, H. F., Zhang, X., Wan, C., Hu, Y., & Wu, B. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.112520 · Cited by 37
Zhang, N., Chen, X., Huang, X., Fan, X., Gan, M., Ji, Z., Sun, Z., & Peng, Z. (2022) · Measurement · doi:10.1016/j.measurement.2022.111849 · Cited by 29
An End-to-End Framework for Shipping Container Corrosion Defect Inspection
Bahrami, Z., Zhang, R., Wang, T., & Liu, Z. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3204091 · Cited by 32
A real-time and efficient surface defect detection method based on YOLOv4
Liu, J., Cui, G., & Xiao, C. (2023) · Journal of Real-Time Image Processing · doi:10.1007/s11554-023-01333-4 · Cited by 33
“[16] integrated the backbone network ResNet with multilevel feature fusion network (MFN) inspired by Faster R-CNN, which achieved a mean average precision (mAP) of 74.”
MethodInfluentialDLF-YOLOF: an improved YOLOF-based surface defect detection for steel plate
Liu, G., Chu, M., Gong, R., & Zheng, Z. (2023) · Journal of Iron and Steel Research International · doi:10.1007/s42243-023-01059-4 · Cited by 30
Lu, J., & Lee, S.-H. (2023) · Electronics · doi:10.3390/electronics12214388 · Cited by 32
SSN: Shift Suppression Network for Endogenous Shift of Photovoltaic Defect Detection
Zhao, S., Chen, H., Wang, C., & Zhang, Z. (2023) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2023.3327572 · Cited by 32
Hassan, S. A., Beliatis, M. J., Radziwon, A., Menciassi, A., & Oddo, C. M. (2024) · Electronics · doi:10.3390/electronics13214314 · Cited by 44
Liu, Q., Guo, Q., Wang, W., Zhang, Y., & Kang, Q. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3054627 · Cited by 35
Wang, H., Ji, Y., Song, K., Sun, M., Lv, P., & Zhang, T. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3122121 · Cited by 28
Self-citationMa, Z., Li, Y., Huang, M., & Deng, N. (2023) · Journal of Manufacturing Systems · doi:10.1016/j.jmsy.2023.06.016 · Cited by 30
A Kernel Two-Sample Test for Functional DataPreprint
Wynne, G., & Duncan, A. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2008.11095 · Cited by 39
“We perform the two-sample test on two classes from the North Eastern University steel defect dataset [67, 29, 18].”
MethodSurface Defect Detection of Strip-Steel Based on an Improved PP-YOLOE-m Detection Network
Zhang, Y., Liu, X., Guo, J., & Zhou, P. (2022) · Electronics · doi:10.3390/electronics11162603 · Cited by 31
Sime, D. M., Wang, G., Zeng, Z., Wang, W., & Peng, B. (2022) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2022.3230785 · Cited by 26
“Defect detection, the region-based classification with a bounding box around a defective region in the image has also been fairly studied [8], [9].”
BackgroundYOLO-Xray: A Bubble Defect Detection Algorithm for Chip X-ray Images Based on Improved YOLOv5
Wang, J., Lin, B., Li, G., Zhou, Y., Zhong, L., Li, X., & Zhang, X. (2023) · Electronics · doi:10.3390/electronics12143060 · Cited by 31
High-Speed and Accurate Cascade Detection Method for Chip Surface Defects
Zhu, X., Wang, S., Su, J. L., Liu, F., & Zeng, L. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3351238 · Cited by 35
“Subsequently, this information is fused with high-level features to achieve focus and delivery of semantic and content information, improving the network’s capability to extract and integrate information across various levels [20].”
BackgroundFD-YOLO11: A Feature-Enhanced Deep Learning Model for Steel Surface Defect Detection
Dang, Z., & Wang, X. (2025) · IEEE Access · doi:10.1109/access.2025.3559733 · Cited by 35
“In this experiment, the NEU-DET [15] dataset is used for training and testing the proposed model.”
Combining Prior Knowledge With CNN for Weak Scratch Inspection of Optical Components
Hou, W. S., Tao, X., & Xu, D. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3011299 · Cited by 27
“He et al. [1] proposed a multilevel-feature fusion network and a region proposal network to detect steel surface defects.”
BackgroundSurface Defects Classification of Hot Rolled Strip Based on Improved Convolutional Neural Network
Wang, W., Lu, K., Wu, Z., Long, H., Zhang, J., Chen, P., & Wang, B. (2021) · ISIJ International · doi:10.2355/isijinternational.isijint-2020-451 · Cited by 28
AnoDFDNet: A Deep Feature Difference Network for Anomaly Detection
Wang, Z., Zhang, Y., Luo, L., & Wang, N. (2022) · Journal of Sensors · doi:10.1155/2022/3538541 · Cited by 33
A Wafer Surface Defect Detection Method Built on Generic Object Detection Network
Wang, X., Jia, X., Jiang, C., & Jiang, S. (2022) · Digital Signal Processing · doi:10.1016/j.dsp.2022.103718 · Cited by 32
Lu, H., Fang, M., Qiu, Y. N., & Xu, W. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3229728 · Cited by 35
“Notably, DDN and RDN [46] are the anchor-based detectors specially designed for NEU-DET.”
BackgroundMethodInfluentialTruong, V. D., Xia, J., Jeong, Y., & Yoon, J. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.106268 · Cited by 27
Lu, B., & Huang, B. (2023) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-023-02105-4 · Cited by 25
“He et al. (2019) employ a two-stage defect detection model embedded with the proposed multilevel feature fusion network that can integrate multiple hierarchical features for detection.”
Huang, J., Zhang, X., Jia, L., & Zhou, Y. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad6281 · Cited by 31
“The effectiveness of the proposed method was evaluated on the steel surface defect samples from the NEU-DET dataset [3] and GC10-DET dataset [25], respectively.”
A steel defect detection method based on edge feature extraction via the Sobel operator
Wang, Y., Yin, T., Chen, X., Hauwa, A. S., Deng, B., Zhu, Y., Gao, S., Zang, H., & Zhao, H. (2024) · Scientific Reports · doi:10.1038/s41598-024-79205-5 · Cited by 39
“To effectively apply these target detection algorithms to the field of steel defect detection, He et al. 12 proposed a multilevel feature fusion network (MFN) to fuse multiple layers of features.”
A Deep-Learning-based 3D Defect Quantitative Inspection System in CC Products Surface
Zhao, L., Li, F., Zhang, Y., Xu, X., Xiao, H. Y., & Yang, F. (2020) · Sensors · doi:10.3390/s20040980 · Cited by 26
Surface defect detection of vehicle light guide plates based on an improved RetinaNet
Li, J., & Wang, H. (2021) · Measurement Science and Technology · doi:10.1088/1361-6501/ac4597 · Cited by 28
“He et al [15] combined multilevel feature maps in the backbone of the Faster R-CNN network with a multiscale feature map, and achieved 82.3% mean average precision (mAP) on the strip steel surface defect dataset NEU-DET.”
InfluentialKähler, F., Schmedemann, O., & Schüppstuhl, T. (2022) · Procedia CIRP · doi:10.1016/j.procir.2022.05.197 · Cited by 29
ELCNN: A Deep Neural Network for Small Object Defect Detection of Magnetic Tile
Liang, W., & Sun, Y. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3193175 · Cited by 28
Coarse-to-Fine Few-Shot Defect Recognition With Dynamic Weighting and Joint Metric
Song, Y., Liu, Z., Ling, S., Tang, R., Duan, G., & Tan, J. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3193204 · Cited by 28
Real-Time Terahertz Characterization of Minor Defects by the YOLOX-MSA Network
Wang, X., Zhang, Z., Xu, Y., Zhang, L., Yan, R., & Chen, X. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3201945 · Cited by 35
“Traditional defect detection 74 methods mainly rely on visual and subjective inspection by 75 quality inspection engineers, which requires high labor cost 76 and has low detection efficiency [6], [7].”
MethodD-CenterNet: An Anchor-Free Detector With Knowledge Distillation for Industrial Defect Detection
Liu, Z., Lyu, W., Wang, C., Guo, Q., Zhou, D., & Xu, W. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3204332 · Cited by 34
“In this section, we evaluate the performance of D-CenterNet 498 based on two public datasets collected by optical image sen-499 sors, Aliyun Tianchi fabric dataset [41] and NEU-DET [42].”
MethodPark, K.-B., & Lee, J. Y. (2022) · Journal of Computational Design and Engineering · doi:10.1093/jcde/qwac115 · Cited by 38
Ragab, K. (2023) · Sensors · doi:10.3390/s23010544 · Cited by 24
“The authors in [50] developed a defect detection network (DDN) that integrates the baseline ResNet34, ResNet50 [52] networks and Region proposal network (RPN) for precise defect detection and localization.”
BackgroundDeep Learning and Laser-Based 3-D Pixel-Level Rail Surface Defect Detection Method
Ye, J., Stewart, E., Chen, Q., Roberts, C., Hajiyavand, A. M., & Lei, Y. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3272033 · Cited by 37
“One is region proposal networks (RPNs), which work by taking a feature map from a CNN as input and output anchor boxes of multiple scales and aspect ratios, each of which has a score indicating whether a defect exists [1].”
BackgroundZheng, X., Liu, W., & Huang, Y. (2024) · IEEE Access · doi:10.1109/access.2024.3349628 · Cited by 26
“(2)Compared with the methods based on deep learning such as ResNet34 + MFN [21] and Decaf [21] methods, the proposed method still has the highest recognition accuracy.”
BackgroundMethodInfluentialQian, K., Zou, L., Wang, Z., & Wang, W. (2024) · Applied Soft Computing · doi:10.1016/j.asoc.2024.111589 · Cited by 31
Wang, Y., Zhou, Z., Tan, X., Pan, Y., Yuan, J., Qiu, Z., & Liu, C. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.127837 · Cited by 32
Mohandas, R., Mongan, P. G., & Hayes, M. J. (2024) · Sensors · doi:10.3390/s24206553 · Cited by 30
“Gao et al. [43] investigated real-time defect detection for industrial applications and tested their methodology on the defect datasets NEU-DET [44] and DeepPCB [45].”
MethodWan, F., Zhang, G., & Li, Z. (2025) · Computers in Industry · doi:10.1016/j.compind.2025.104265 · Cited by 27
FCS-YOLO: an efficient algorithm for detecting steel surface defects
Zhao, C., Fan, Y., Tan, J., Li, Q., Lin, Z., Luo, S., & Chen, X. (2024) · Measurement Science and Technology · doi:10.1088/1361-6501/ad4410 · Cited by 26
“He et al [11] proposed a novel defect detection system based on deep learning, using baseline convolutional neural network to generate feature maps at each stage, and then proposed multilevel feature fusion network to combine multiple layered features into one feature, which can contain more…”
Surface defect identification of aluminium strips with non-subsampled shearlet transform
Liu, X., Xu, K., Zhou, P., Zhou, D., & Zhou, Y. (2020) · Optics and Lasers in Engineering · doi:10.1016/j.optlaseng.2019.105986 · Cited by 24
A Review and Benchmark on State-of-the-Art Steel Defects Detection
Chazhoor, A. A. P., Ho, E. S. L., Gao, B., & Woo, W. L. (2023) · SN Computer Science · doi:10.1007/s42979-023-02436-2 · Cited by 41
“He et al. [32] devised a new deep learning-based defect detection system with realistic commercial applicability by fusing multiple hierarchical features that can determine the class along with the detailed location of the defect.”
BackgroundCFE-YOLOv8s: Improved YOLOv8s for Steel Surface Defect Detection
Yang, S., Xie, Y., Wu, J., Huang, W., Yan, H., Wang, J., Bi, W., Yu, X., Wu, Q., & Xie, F. (2024) · Electronics · doi:10.3390/electronics13142771 · Cited by 29
Liang, F., Wang, Z., Ma, W., Liu, B., En, Q., Wang, D., & Duan, L. (2024) · Measurement · doi:10.1016/j.measurement.2024.116255 · Cited by 27
Feature optimization-guided high-precision and real-time metal surface defect detection network
Chan, S., Li, S., Zhang, H., Zhou, X., Mao, J., & Hong, F. (2024) · Scientific Reports · doi:10.1038/s41598-024-83430-3 · Cited by 31
FPGA-Based Acceleration on Additive Manufacturing Defects Inspection
Luo, Y., & Chen, Y. (2021) · Sensors · doi:10.3390/s21062123 · Cited by 26
“[12] proposed a defect detection network (DDN) and set up the NEUDET dataset.”
BackgroundWang, Y., Zhang, Y., Li, Z., Yin, L., Chen, J., & Lu, J. (2021) · Sensors · doi:10.3390/s21206773 · Cited by 27
Strip Steel Defect Classification Using the Improved GAN and EfficientNet
Guan, S., Chang, J., Shi, H., Xiao, X., Li, Z., Wang, X., & Wang, X. (2021) · Applied Artificial Intelligence · doi:10.1080/08839514.2021.1995231 · Cited by 25
Three-Channel Infrared Imaging for Object Detection in Haze
Yu, B., Chen, Y., Cao, S.-Y., Shen, H., & Li, J. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3164062 · Cited by 24
Chen, H., Song, M., Zhang, Z., & Liu, K. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3218111 · Cited by 26
“Deep CNN-based approaches have recently shown excellent performance because of the capacity to automatically extract depth abstract features [12], [13], [14], [15].”
BackgroundFerrite Beads Surface Defect Detection Based on Spatial Attention Under Weakly Supervised Learning
Li, Y., Wu, X., Li, P., & Liu, Y. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3246499 · Cited by 25
“And the cost of labeling will increase rapidly as the labeling level (image level [1], bounding box level [2], pixel level [3]) increase.”
BackgroundMethodA gated multi-hierarchical feature fusion network for recognizing steel plate surface defects
Tao, H., Lu, M., Hu, Z., & An, J. (2023) · Multimedia Systems · doi:10.1007/s00530-023-01066-1 · Cited by 25
“It can be observed that our model performs better than all the baseline methods including DDN [17], SPANet [48], IDNet [49], and ResnetYolo [50].”
BackgroundMethodInfluentialZhao, S., Chen, H., Wang, C., Zhou, Y., & Zhang, Z. (2023) · Expert Systems with Applications · doi:10.1016/j.eswa.2023.123034 · Cited by 24
Intelligent Recognition Model of Hot Rolling Strip Edge Defects Based on Deep Learning
Wang, D., Xu, Y., Duan, B., Wang, Y., Song, M., Yu, H., & Liu, H. (2021) · Metals · doi:10.3390/met11020223 · Cited by 27
Logit Inducing With Abnormality Capturing for Semi-Supervised Image Anomaly Detection
Wan, Q., Gao, L., & Li, X. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3205674 · Cited by 22
Context-Aware Adaptive Weighted Attention Network for Real-Time Surface Defect Segmentation
Zhang, G., Lu, Y., Jiang, X., Yan, F., & Xu, M. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3449979 · Cited by 36
“One commonly employed approach is the classification of surface defects, while another extensively used approach is region-based defect detection [1], [3], [13].”
MethodSteel Surface Defect Detection Method Based on Improved YOLOv9 Network
Zou, J., & Wang, H. (2024) · IEEE Access · doi:10.1109/access.2024.3453931 · Cited by 25
“The study utilizes the NEU-DET [6] dataset for detecting defects on steel surfaces, which includes six categories: craz-ing, inclusion, patches, pitted_surface, rolled-in_scale, and scratches.”
BackgroundMethodA high precision YOLO model for surface defect detection based on PyConv and CISBA
Ruan, S., Zhan, C., Liu, B., Wan, Q., & Song, K. (2025) · Scientific Reports · doi:10.1038/s41598-025-91930-z · Cited by 30
Self-citationLiu, T., Bao, J., Wang, J., & Wang, J. (2021) · International Journal of Computer Integrated Manufacturing · doi:10.1080/0951192x.2021.1901319 · Cited by 24
“He et al. 2020; Akram et al. 2020; Kang et al. 2018; Wei et al. 2019) are commonly used in the industrial field.”
BackgroundInfluentialClassification and Fast Few-Shot Learning of Steel Surface Defects with Randomized Network
Nagy, A. M., & Czúni, L. (2022) · Applied Sciences · doi:10.3390/app12083967 · Cited by 30
Dam Crack Image Detection Model on Feature Enhancement and Attention Mechanism
Xu, G., Han, X., Zhang, Y., et al. (2022) · Water · doi:10.3390/w15010064 · Cited by 22
Surface Defect Detection of Steel Strip with Double Pyramid Network
Zhou, X., Wei, M., Li, Q., Fu, Y., Gan, Y., Liu, H., Ruan, J., & Liang, J. (2023) · Applied Sciences · doi:10.3390/app13021054 · Cited by 27
Tian, Y., Meng, H., Yuan, F., Ling, Y., & Yuan, N. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3268455 · Cited by 27
“IN RECENT years, the successful applications of deep learning [1], [2], [3] have led to tremendous developments in visual surveillance ship recognition systems, maritime traffic management, and port security surveillance.”
BackgroundAdaptive-MAML: Few-shot metal surface defects diagnosis based on model-agnostic meta-learning
Pang, S., Zhang, L., Yuan, Y., Zhao, W., Wang, S., & Wang, S. (2023) · Measurement · doi:10.1016/j.measurement.2023.113612 · Cited by 26
Deep Gaussian Attention Network for Lumber Surface Defect Segmentation
Zhong, Y., Ling, Z., Liu, L., Zhang, S., & Wen, H. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3381269 · Cited by 28
“He et al. [5] proposed a novel end-to-end defect detection approach based on deep learning via fusing multiple hierarchical features for a practical industrial application.”
BackgroundYu, T., Luo, X., Li, Q., & Li, L. (2024) · International Journal of Computational Intelligence Systems · doi:10.1007/s44196-024-00559-9 · Cited by 28
“He et al. [10] proposed a strip surface defect detection network based on the improved Faster R-CNN, and multiple hierarchical features were combined into one feature in the Resnet backbone network, which made good use of global semantic information and reduced feature loss.”
Gas pipeline defect detection based on improved deep learning approach
Zhang, T., Ma, C., Liu, Z., Rehman, S. U., Li, Y., & Saraee, M. H. (2024) · Expert Systems with Applications · doi:10.1016/j.eswa.2024.126212 · Cited by 25
Zheng, X., Liu, W., & Huang, Y. (2025) · Measurement · doi:10.1016/j.measurement.2025.117393 · Cited by 26
SPS-Net: Self-Attention Photometric Stereo Network
Liu, H., Yan, Y., Song, K., & Yu, H. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3047917 · Cited by 21
“D sensing [9], object positioning [10], and defect inspection [11], [12].”
BackgroundSelf-citationHot‐Rolled Steel Strip Surface Inspection Based on Transfer Learning Model
Wu, H., & Lv, Q. (2021) · Journal of Sensors · doi:10.1155/2021/6637252 · Cited by 22
“The data set of surface defects of hot-rolled strips used in this experiment is from Northeast University [21], as shown in Figure 4.”
MethodAn explanation space to align user studies with the technical development of Explainable AI
Cabour, G., Morales‐Forero, A., Ledoux, É., & Bassetto, S. (2022) · AI & Society · doi:10.1007/s00146-022-01536-6 · Cited by 42
Jiang, L., Yuan, B., Ma, W., & Wang, Y. (2023) · Frontiers in Plant Science · doi:10.3389/fpls.2022.1108437 · Cited by 21
“He et al. (2020) proposed An end-to-end steel SDD approach via fusing multiple hierarchical features.”
MethodAutomatic pixel-level detection of tire defects based on a lightweight Transformer architecture
Sun, Y., Liu, X., Zhai, X., Sun, K., Zhao, M., Chang, Y., & Zhang, Y. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/acd5f2 · Cited by 22
“steel [1], fabric [2], tires [3], etc, which have greatly improved the quality inspection efficiency.”
ADNet: Anti-noise dual-branch network for road defect detection
Wan, B., Zhou, X., Sun, Y., Wang, T., Lv, C., Wang, S., Yin, H., & Yan, C. C. (2024) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2024.107963 · Cited by 25
Surface Defect Classification of Steel Strip with Few Samples Based on Dual‐Stream Neural Network
Zhang, J., Li, S., Yan, Y., Ni, Z., & Ni, H. (2021) · steel research international · doi:10.1002/srin.202100554 · Cited by 23
Self-citationA Survey on Surface Defect Inspection Based on Generative Models in Manufacturing
He, Y., Li, S., Wen, X., & Xu, J. (2024) · Applied Sciences · doi:10.3390/app14156774 · Cited by 35
Self-citationAdaptive Cross Transformer With Contrastive Learning for Surface Defect Detection
Huang, X., Li, Y., Bao, Y., & Zheng, W. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3470998 · Cited by 32
“He et al. [3] focused on a practical industrial application, establishing the NEU steel surface defect detection dataset for training and evaluating their method.”
MethodAn Efficient Deep Neural Network for Surface Defect Detection in Industrial Edge Sensing
Wang, J., Zou, H., Zhou, M., & Su, R. (2024) · IEEE Transactions on Industrial Informatics · doi:10.1109/tii.2024.3507954 · Cited by 30
“I NDUSTRIAL defect detection is important for industrial products, such as circuit board, chip, battery, steel, and material [1], [2], [3], [4].”
Metadata
- Authors
- He, Y., Song, K., Meng, Q., & Yan, Y.
- Affiliations
- Northeastern University (CN)
- Loughborough University (GB)
- Venue
- IEEE Transactions on Instrumentation and Measurement
- Published
- 2020-04
- Volume
- 69
- Issue
- 4
- Pages
- 1493-1504
- ISSN
- 0018-9456 (ISSN-L)
- 1557-9662
- OpenAlex
- W2944303778
- Semantic Scholar
- 46d4a2b7ccf38b80b09574cf17544ff2297a8fbf
- Cited by
- 1,341
- References
- 53
- Citations by year
- 2026: 188 · 2025: 304 · 2024: 269 · 2023: 238 · 2022: 186 · 2021: 111 · 2020: 34
- FWCI
- 65.18
- Type
- journal article
- Language
- en
- Open access
- green
- figshare.com
- Licence
- Locations
- IEEE Transactions on Instrumentation and Measurement (publishedVersion)
- Figshare (submittedVersion, cc-by)
- Funders
- National Natural Science Foundation of China (51805078, 51374063)
- National Basic Research Program of China (2017YFB0304200)
- Fundamental Research Funds for the Central Universities (N170304014, N150308001)
- China Scholarship Council (201806085007)
- Topics
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Physical Sciences › Computer Science › Computer Vision and Pattern Recognition › Advanced Neural Network Applications
- Keywords
- multi-level feature fusion, hierarchical features, region proposal network, bounding box regression, NEU-DET dataset, real-time defect detection, industrial visual inspection, object detection
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.