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 701–800 of 1,338
Dong, F., Zhang, S., chen, X., Mu, D., Yu, X., & Shen, L. (2025) · IEEE Internet of Things Journal · doi:10.1109/jiot.2025.3632285 · Cited by 4
“To verify the e ffi cacy and ideal detection capabilities of the investigated Vanillanet-BGA network, this study conducts strip steel defects detection experiments on NEU-DET [49] and GC10-DET [50] datasets.”
Mohanty, S., Su, E., & Ho, C.-C. (2024) · Journal of Electronic Imaging · doi:10.1117/1.jei.33.1.013021 · Cited by 4
Vengaloor, R. P., & Muralidhar, R. (2023) · The International Arab Journal of Information Technology · doi:10.34028/iajit/21/1/9 · Cited by 10
An effective method for small objects detection based on MDFFAM and LKSPP
Xu, Z., Xu, Y., & Wang, M. (2024) · Scientific Reports · doi:10.1038/s41598-024-60745-9 · Cited by 8
Multiclass small target detection algorithm for surface defects of chemicals special steel
Wang, Y., Yan, S., Abdullahi, H. S., et al. (2024) · Frontiers in Physics · doi:10.3389/fphy.2024.1451165 · Cited by 4
“Given the special complexity of the application scenarios of chemical special steel and the strict requirements for safe production, it is particularly critical to achieve high accuracy in the detection of various minor defects [6].”
A Pipeline Leakage Aperture Identification Method Based on CNN-SVM Considering Sample Granularity
Han, X., Li, C., Cui, X., et al. (2024) · IEEE Sensors Journal · doi:10.1109/jsen.2024.3454269 · Cited by 4
“The convolutional neural network (CNN) stands out as a quintessential technique within the realm of deep learning that can automatically extract features from raw data, and it has been widely used in fault detection and classification [14], [15], [16].”
BackgroundLi, X., Xu, C., Li, J., Li, Y., & Zhou, X. (2024) · Ironmaking & Steelmaking Processes Products and Applications · doi:10.1177/03019233241293880 · Cited by 5
Investigating the generation of synthetic data for surface defect detection: A comparative analysis
Monnet, J., Petrović, O., & Herfs, W. (2024) · Procedia CIRP · doi:10.1016/j.procir.2024.10.162 · Cited by 12
EMC-YOLO: a feature enhancement and fusion based surface defect detection for hot rolled strip steel
Zhu, X., Wan, X., & Zhang, M. (2025) · Engineering Research Express · doi:10.1088/2631-8695/ada7c4 · Cited by 7
Foreign object detection in the inspection of cloud server center using separable self-attention
Jiang, Y., Li, Z., Zhao, B., Zhang, X., & Dong, X. (2025) · Intelligent Service Robotics · doi:10.1007/s11370-024-00580-z · Cited by 4
“The results 1–2 in Table 6 were obtained through self-experimentation, while 3 were obtained from [36], 4-5 from [38], 6 from[33], and 7 from [34].”
FFDDNet: Flexible Focused Defect Detection Network
Lin, Z., Li, Z., Yu, J., Hu, M., & Wang, X. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3551459 · Cited by 10
“He et al. [36] focus on practical industrial applications, a defect detection dataset NEU-DET was established, and a defect detection network (DDN) was proposed to merge multilayer features with a multilayer feature fusion network (MFN) to amplify the location details of defects.”
InfluentialZhang, J., Pan, J., Qu, H., et al. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3561319 · Cited by 4
“The ResNet34 network model was employed for crack detection in spiral pipelines [48], [49].”
Surface defect detection on rolling mill rolls: a transfer learning approach with improved YOLOv5
Hu, Y., Wen, P., Chen, S., Liu, J., Li, X., Sun, J., & Zhang, D. (2025) · Discover Applied Sciences · doi:10.1007/s42452-025-06920-7 · Cited by 4
“The steel strip dataset is sourced from NEU-DET [20], while the roll surface defect dataset is collected from real-world steel production environments.”
Huang, F., Zheng, L.-M., Wen, H., Dai, M., & Zhang, Z.-S. (2025) · Journal of Intelligent Manufacturing · doi:10.1007/s10845-025-02658-6 · Cited by 6
DC-Mamba for surface micro defect classification on large aperture optics with multi axis attention
Zhao, D., Sun, R., Tong, T., et al. (2025) · Scientific Reports · doi:10.1038/s41598-025-21756-2 · Cited by 3
“The multilevel feature fusion network 11 integrates multiple levels of features into one, enabling accurate classification and localization of defects in each image.”
Convolutional Neural Networks for Defect Detection on LV cables
Gelan, T. M. (2023) · Journal of Applied Artificial Intelligence · doi:10.48185/jaai.v3i2.620 · Cited by 4
Cross-Domain Transfer Learning for Galvanized Steel Strips Defect Detection and Recognition
Chen, H., Lin, H., Xu, Q., Li, Y., Zheng, Y., Fei, J., Yang, K., Fan, W., & Nie, Z. (2023) · Journal of Computing and Information Science in Engineering · doi:10.1115/1.4063102 · Cited by 4
Wu, Z., Huang, H., Zhang, R., Liu, J., & Tang, J. (2023) · Measurement Science and Technology · doi:10.1088/1361-6501/acfbee · Cited by 4
“He et al used computer vision to propose a deep learning based defect detection method for steel surfaces, which improved the performance of defect classification and location [26].”
SSDD-Net: A Lightweight and Efficient Deep Learning Model for Steel Surface Defect Detection
Li, Z., Wei, X., & Jiang, X. (2023) · Lecture notes in computer science · doi:10.1007/978-981-99-8549-4_20 · Cited by 4
Vision-based Robotic Arm in Defect Detection and Object Classification Applications
Lin, C., Jhang, J., Gao, Y.-J., & Huang, H. (2024) · Sensors and Materials · doi:10.18494/sam4683 · Cited by 4
He, X., Jin, J., Yu, F., Zhao, S., Chen, D., & Gao, X. Y. (2024) · Measurement · doi:10.1016/j.measurement.2024.115753 · Cited by 5
CFF-YOLO: cross-space feature fusion based YOLO model for screw detection in vehicle chassis
Xu, H., Ding, F., Zhou, W., et al. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03474-w · Cited by 4
“And insulator defect localization in unmanned aerial vehicle (UAV) power inspections [18], steel strip surface defect detection [19] achieved high mAP.”
A Multi-Scale Feature Fusion Network for Chip Surface Defect Detection
Ren, H., Tian, M., Zhang, G., & Zhou, W. (2024) · Proceedings - International Conference on Image Processing · doi:10.1109/icip51287.2024.10647644 · Cited by 7
A Fast Surface-Defect Detection Method Based on Dense-Yolo Network
Gao, F., Zhu, Q., Shao, G., Su, Y., Yang, J., & Yu, X. (2022) · CAAI Transactions on Intelligence Technology · doi:10.1049/cit2.12407 · Cited by 8
Xu, P., Li, Y., Gou, F., et al. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3541702 · Cited by 4
“He et al. [19] proposed an end-to-end steel surface defect detection method that achieved precise defect localization and specific category recognition through multilevel feature fusion.”
Luo, S., Liu, B., Chen, H., Wang, Z., Yang, R., & Ying, H. (2025) · IEEE Sensors Journal · doi:10.1109/jsen.2025.3549521 · Cited by 4
“NEU-DET: To validate the effectiveness and generalizability of our proposed method, we conducted experiments using NEU-DET [48], a widely studied steel defect detection dataset.”
Brake disc defect detection method based on an improved multi-scale template matching algorithm
Yang, Y., Guo, J. H., Liu, C., Qian, H., & Gu, P. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-04093-9 · Cited by 4
“In this regard, He et al. [9] proposed the Multilevel Feature Fusion Network (MFN), which utilizes CNNs to generate feature maps at different stages.”
Co-training encoder-based transformer method for steel surface defect detection
Wang, Y., Cheng, H., Shi, J., Du, X., & Wang, P. (2025) · Journal of Electronic Imaging · doi:10.1117/1.jei.34.2.023058 · Cited by 4
“To validate the efficacy of our proposed method, we employed the NEU-DET dataset 38 for training.”
Semi-SSDDet: An adaptive semi-supervised steel surface defect detection
Liu, G., Chu, M., & Gong, R. (2025) · Signal Image and Video Processing · doi:10.1007/s11760-025-04416-w · Cited by 4
“In this section, we conduct a series of experiments on the NET-DET [17] dataset and GC10-DET [49] dataset to validate the effectiveness of our proposed Semi-SSDDet.”
A nonparallel support vector machine with pinball loss for pattern classification
Liu, L., Chu, M., Gong, R., & Qi, X. (2020) · Journal of Intelligent & Fuzzy Systems · doi:10.3233/jifs-191845 · Cited by 5
“To demonstrate the effectiveness of our Pin-NPSVM, we apply the Pin-NPSVM, NPSVM, TSVM, Pin-SVM, SVM and DT algorithms to steel surface defects datasets [31, 32].”
Improving the Algorithm Study of YOLO in Steel Surface Defect Detection
Fu, S., & Wang, S. (2022) · International Journal of Materials · doi:10.46300/91018.2022.9.5 · Cited by 4
Evaluation of Additional Augmented Images for Steel Surface Defect Detection
Farady, I., Shashank, Sarkar, M. D., et al. (2022) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-taiwan55306.2022.9869272 · Cited by 4
Using Computer Vision to Recognize Defects on the Surface of Hot-rolled Steel
Vozmilov, A. G., Губаевич, У. В., & Анатольевич, Л. А. (2022) · International Ural Conference on Electrical Power Engineering (UralCon) · doi:10.1109/uralcon54942.2022.9906737 · Cited by 5
Nonparallel Support Vector Machine with L2-norm Loss and its DCD-type Solver
Liu, L., Li, P., Chu, M., & Gong, R. (2022) · Neural Processing Letters · doi:10.1007/s11063-022-11067-2 · Cited by 4
“The defect images were provided by the Northeastern University (NEU) surface defect database [36, 37].”
MethodYan, 周. Z., Jiangnan, 孟. M., Jia, 吴. W., Zhi, 罗. L., & Dongli, 王. W. (2023) · Laser & Optoelectronics Progress · doi:10.3788/lop213302 · Cited by 4
Metal Surface Defect Detection using Object Detection Models
Shah, H., & Patil, V. (2023) · International Conference on Intelligent and Innovative Technologies in Computing, Electrical and Electronics (IITCEE) · doi:10.1109/iitcee57236.2023.10090466 · Cited by 5
“[1] propose a new architecture combining CNN, MFN, and RPN to detect and localize defects on metal surfaces.”
BackgroundTexture and semantic convolutional auto‐encoder for anomaly detection and segmentation
Luo, J., Gao, C., Wan, D., & Shen, L. (2023) · IET Computer Vision · doi:10.1049/cvi2.12200 · Cited by 4
Steel surface defect detection based on Improved YOLOX Algorithm
Li, G., & Mokhtarzadeh, A. A. (2023) · Journal of Physics Conference Series · doi:10.1088/1742-6596/2467/1/012005 · Cited by 5
“[4] proposed a strip steel surface defect detection network based on improved Faster R-CNN.”
MethodA Two-Stage Focal Transformer for Human–Robot Collaboration-Based Surface Defect Inspection
Gao, Y., Gao, L. S., & Li, X. (2023) · Journal of Manufacturing Science and Engineering · doi:10.1115/1.4062860 · Cited by 4
A Novel Detection Method for Valve Damage of Nuclear Power Using Attention-Based U-Net (ABUNet)
Cao, S., Shen, S., Fan, S., Wang, X., Zhu, Z., Shi, Y., & Lou, J. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3303012 · Cited by 5
“site and assign it a class label [25], [26], [27].”
BackgroundGao, S., Xia, T., Hong, G., Zhu, Y., Chen, Z., Pan, E., & Xi, L. (2023) · Measurement · doi:10.1016/j.measurement.2023.113957 · Cited by 6
Hu, D. (2024) · Journal of Grid Computing · doi:10.1007/s10723-023-09738-y · Cited by 4
“The data set used for training did not reflect the manufacturing floor because it was not recorded in a warehouse setting, which raised concerns regarding the real applicability of the suggested design when looking more closely at the manufacturing process [28, 29].”
Research on surface defect detection model of steel strip based on MFFA‐YOLOv5
Chen, H., Qiu, J., Gao, D., Qian, L., & Li, X. (2024) · IET Image Processing · doi:10.1049/ipr2.13086 · Cited by 4
Zhang, H., Miao, Q., Li, S., Wang, C., Chan, S., Hu, J., & Bai, C. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19565-2 · Cited by 6
“He et al. [26] achieved the mean average precision (mAP) of 82.3%, but when the detection speed was raised to 20 FPS, the mAP of the detection model dropped to 70%.”
InfluentialAn Improved YOLOv8 Algorithm for Steel Surface Defect Detection
Wang, W., Xiao, H., Zhang, Q., et al. (2024) · Asian Conference on Artificial Intelligence Technology (ACAIT) · doi:10.1109/acait63902.2024.11022247 · Cited by 4
“To validate the effectiveness of our model using the NEU-DET [34] and GC10-DET [35] datasets.”
Liu, J., Liu, Y., Dong, K., et al. (2019) · International Journal of Distributed Sensor Networks · doi:10.1177/1550147719895952 · Cited by 4
Self-citationEvaluation of Data Augmentation on Surface Defect Detection
Farady, I., Lin, C., Akhyar, F., et al. (2021) · IEEE International Conference on Consumer Electronics - Taiwan (ICCE-Taiwan) · doi:10.1109/icce-tw52618.2021.9603212 · Cited by 4
“We compared various augmentation methods on three benchmarked public surface defect detection datasets including the Severstal steel surface dataset from Kaggle 1 , DAGM [5], and NEU [4].”
MethodTowards an Explanation Space to Align Humans and Explainable-AI TeamworkPreprint
Cabour, G., Morales, A., Ledoux, É., & Bassetto, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2106.01503 · Cited by 4
“The sample image on the left side is an original image taken from the NEU surface defect database [73, 74, 75].”
MethodA Review and Benchmark on State-of-The-Art Steel Defects DetectionPreprint
Chazhoor, A. A. P., Ho, E. S. L., Woo, W. L., & Gao, B. (2022) · SSRN Electronic Journal · doi:10.2139/ssrn.4121951 · Cited by 5
Global Attention Module and Cascade Fusion Network for Steel Surface Defect DetectionPreprint
Liu, G., Chu, M., Gong, R., et al. (2024) · SSRN Electronic Journal · doi:10.2139/ssrn.4818932 · Cited by 4
Analysis of Surface Interlaminar Fault Detection Method for Radial Magnetic Bearing Rotor Cores
He, J., Su, Z., Zhang, Q., Hu, J., Jiang, H., Jiang, Q., & Wang, D. (2024) · IEEE Transactions on Energy Conversion · doi:10.1109/tec.2024.3486903 · Cited by 4
“A large amount of literature discussing methods for fault identification on electrical steel sheets’ surfaces [18], [19].”
Wang, Z., & Song, Y. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/add611 · Cited by 3
“(5) Our approach demonstrate superior performance, outper-forming the current mainstream detection models on the NEU-DET [12], GC-10 [13], and TCAL datasets [14].”
Liu, C., Yang, Q., Ma, S., Guo, Q., Gao, P., & Wang, R. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3570366 · Cited by 4
“For instance, an end-to-end steel SDD approach was proposed by He et al. [3], which fuses multiple hierarchical features.”
Yu, Y., Jiang, S., Wang, Y., & Zhu, S. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.111278 · Cited by 5
Mobile phone screen defect detection based on improved YOLOv8n network
Shi, P., Li, X., Feng, Z., & Shang, X. (2025) · Pattern Recognition Letters · doi:10.1016/j.patrec.2025.05.011 · Cited by 4
Zhang, K., Liu, C., & Tang, L. (2025) · IEEE Transactions on Image Processing · doi:10.1109/tip.2025.3580320 · Cited by 3
“3) Comparisons With State-of-the-Art Methods: To evaluate the e ff ectiveness of MOEA-FM-Trans, it is compared with several state-of-the-art methods, including MHFN [38], MIFusion [39], NCFNet [40], CSFF [41], MSMA [6], Swin-T [42], CNN-T [43], HybridDC [10], as listed in Table III.”
LScAD: A Large–Small Model Collaboration Framework for Unsupervised Industrial Anomaly Detection
Qu, S., Tao, X., Gong, X., Qu, Z., Prasad, M., Shen, F., Zhang, Z., & Ding, G. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3588927 · Cited by 8
“I NDUSTRIAL defect detection based on supervised deep learning is a critical technology widely used in fields, such as mechanical manufacturing [1], consumer electronics [2], and bridge inspection [3].”
Research on a lightweight cone yarn varieties recognition model
Du, Z., Dai, N., Hu, X., et al. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adf24e · Cited by 3
“To evaluate the generalization capability of the YOLO-FSL model, cross-domain tests were conducted on two publicly available datasets: DUT Anti-UAV [37] and NEU-DET [38].”
Ding, H., Chen, J., Ye, H., & Chen, Y. (2025) · Applied Sciences · doi:10.3390/app15147887 · Cited by 4
A real-time end-to-end detector for detecting surface defects on oversized rings
Liu, J., Zhang, L., Chen, S., Wang, Y., Wu, B., & Dong, J. (2025) · PLoS ONE · doi:10.1371/journal.pone.0330031 · Cited by 3
“The dataset used in this study is the NEU-DET [35] dataset from Northeastern University.”
Zhou, X., Chen, X., Chen, J., & Chen, B. M. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.103891 · Cited by 3
A multi-stage dynamic self-distillation network for industrial defect detection
Zhao, H., Liu, J., Wang, Q., Wen, Z., & Shen, X. (2025) · Advanced Engineering Informatics · doi:10.1016/j.aei.2025.103921 · Cited by 3
S 2 -AMNet: A lightweight Spatial–Spectral Adaptive Modulation Network for surface defect detection
Bao, J., Zhang, C., Bao, L., & Zhang, J. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112778 · Cited by 6
Tang, J., Zhang, Q., Jiang, J., Pan, Y., & Ding, X. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/ae1854 · Cited by 5
“Secondly, the classification process is complicated by the large number of defect types and the fact that these defect types often occur simultaneously [3].”
Kong, L., Mei, J., Jia, F., & Du, W. (2025) · Applied Soft Computing · doi:10.1016/j.asoc.2025.114346 · Cited by 5
Wen, Z., Liu, J., Tang, J., Zhao, H. M., Jiang, L., & Wang, Q. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3643046 · Cited by 14
“A. Dataset 1) NEU-DET: NEU-DET is a surface defect database for hot-rolled steel strips, collected by Northeastern University (NEU) [33].”
Low-latency network-based fast melt pool state recognition method
Lu, Y., Dong, X., Fan, Y., Jia, H., Wang, F., yu, S., Chen, X., & Zhang, T. (2026) · Optics Express · doi:10.1364/oe.583810 · Cited by 3
Huang, W., Zhao, J., Wei, R., et al. (2026) · Information Sciences · doi:10.1016/j.ins.2026.123092 · Cited by 3
Xiang, H., Gu, J., Xiang, J., et al. (2026) · Knowledge-Based Systems · doi:10.1016/j.knosys.2026.115769 · Cited by 3
Su, S., Shen, J., Lin, P., Tang, W., Zhang, W., & Chen, Z. (2026) · Sensors · doi:10.3390/s26082311 · Cited by 3
Supervision dropout: guidance learning in deep neural network
Zeng, L., Zhang, H., Li, Y., Li, M., & Wang, S. (2022) · Multimedia Tools and Applications · doi:10.1007/s11042-022-14274-0 · Cited by 3
“With the development of information technology, deep neural networks(DNNs) [2, 7, 32] have made great progresses in multiple fields, such as image classification [9, 17, 19], target detection [13, 45] and semantic segmentation [42].”
MethodWang, L., Huang, X., Zheng, Z., & Ruan, H. (2023) · PLoS ONE · doi:10.1371/journal.pone.0279035 · Cited by 3
Deep Reinforcement Learning Based Multi-Task Automated Channel Pruning for DNNs
Ma, X., & Fang, W. W. (2023) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn54540.2023.10191092 · Cited by 3
“, VGG or MobileNet, can be used for detect defects for different types of products, such as printed fabric [11], wafer structure [12] and steel surface [13].”
MethodPixel-Level Contrastive Pretrainer for Industrial Image Representation
Zhu, B., Chen, Y., Tang, M., & Wang, J. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3353860 · Cited by 6
“It plays a significant role in facilitating the advancement of surface defect detection technologies, such as Strip Steel Surface Defect [2], [3], [4], [5], glass defect detection [6], [7], pavement defect detection [8], [9], [10], LED defect detection, as well as improve industrial anomaly…”
BackgroundMethodAn Enhanced Instance Segmentation Network for Surface Defect Detection
Ren, X., & He, X. (2023) · IEEE International Conference on Computer and Communications (ICCC) · doi:10.1109/iccc59590.2023.10507476 · Cited by 5
MDC-Net: Multimodal Detection and Captioning Network for Steel Surface Defects
Chazhoor, A. A. P., Hu, S., Gao, B., & Woo, W. L. (2024) · Communications in computer and information science · doi:10.1007/978-3-031-59057-3_20 · Cited by 3
FSPDD: A double-branch attention guided network for few-shot PCB defect detection
Shi, K., Xu, Z., Cao, Y., Zhao, L., & Kang, Y. (2024) · Multimedia Tools and Applications · doi:10.1007/s11042-024-19893-3 · Cited by 3
Özdemir, M. E., & Karakuş, F. (2024) · Electronics · doi:10.3390/electronics13203998 · Cited by 3
Fu, G., Chen, J., Qian, S., Miao, J., Li, J., Jiang, Q., Zhu, Q., & Shen, Y. (2024) · Measurement · doi:10.1016/j.measurement.2024.116220 · Cited by 5
Wang, P., Shi, D., & Aguilar, J. (2024) · Signal Image and Video Processing · doi:10.1007/s11760-024-03666-4 · Cited by 3
“To thoroughly validate the efficacy and generalization capability of the presented approach, this study utilizes three public datasets in the domain of defect detection: NEU-DET [32], GC10-DET [33] and PKU-PCB.”
Progressive Feature Enhancement Network for Surface Defect Segmentation
Yan, F., Jiang, X., Zhang, Y., Lu, Y., Nan, X., He, S., & Xu, M. (2025) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2024.3523770 · Cited by 5
“He et al. [19] proposed a multi-level feature fusion method for end-to-end steel surface defect detection.”
Chomklin, A., Jaiyen, S., & Wattanakitrungroj, N. (2024) · International Conference on Business and Industrial Research (ICBIR) · doi:10.1109/icbir61386.2024.10875971 · Cited by 3
Yu, S., & Wang, Z. (2025) · Journal of Real-Time Image Processing · doi:10.1007/s11554-025-01636-8 · Cited by 3
“He et al. [19] proposed adding a multi-level feature fusion network to a pre-trained model, which effectively improved the detection rate of multi-scale surface defects.”
Wu, H., Li, B., Tian, L., Dong, C., Liao, W., & Liu, Y. (2025) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2025.3545193 · Cited by 4
“Those methods are tailored for NEU-DET dataset [2], KolektorSDD2 dataset [49], and RSSDs dataset [51], respectively.”
InfluentialA Scale-Adaptive and Background-Robust Method for Surface Defect Detection
Dong, J., Zuo, Z., Wu, Z., & Liu, M. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10889710 · Cited by 3
“HRIPCB is a public dataset which consists of 693 images of printed circuit boards (PCBs) [18], and NEU-DET comprises 1800 images of strip steel surface defects [19].”
Surface Defect Detection Algorithm for Strip Alloy Material Based on Improved YOLOv8
Yang, W. P., Yang, J. A., & Xia, Y. (2025) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp49660.2025.10890088 · Cited by 3
Steel surface defect detection based on lightweight YOLOv7
Shi, T., Wu, R., Zhu, W., et al. (2025) · Optoelectronics Letters · doi:10.1007/s11801-025-3104-2 · Cited by 3
“HE et al [12] proposed a strip surface defect detection network based on the improved faster R-CNN, which makes better use of the global semantic information.”
Fu, G., Le, W., Li, J., et al. (2025) · Measurement Science and Technology · doi:10.1088/1361-6501/adc9d7 · Cited by 3
“In addition, on the publicly available NEU-DET dataset, most of the existing works [9, 13, 14] have very low accuracy on the crazing defect class.”
InfluentialZhang, T., You, T., Liu, Z., Rehman, S. U., Shi, Y., & Munshi, A. (2025) · Applied Intelligence · doi:10.1007/s10489-025-06590-3 · Cited by 6
“NEU-DET is a dataset for steel surface defect detection, containing six types of defects with 300 samples each, totaling 1,800 samples and corresponding labels [56].”
EN-YOLO: a featured-enhanced model for precise hot rolled steel surface defect detection
Hao, Z., Liu, B., & Xu, B. (2025) · Nondestructive Testing And Evaluation · doi:10.1080/10589759.2025.2524604 · Cited by 4
“Despite these advancements, challenges persist in steel surface defect detection using deep learning, including noise interference in small defect detection, incomplete feature extraction for large defects, and inter-class feature confusion caused by dynamic lighting conditions [39,40].”
Real-Time Detection of Hole-Type Defects on Industrial Components Using Raspberry Pi 5
Deniz, M., Böğrekçi, İ., & Demircioğlu, P. (2025) · Applied System Innovation · doi:10.3390/asi8040089 · Cited by 7
“He et al. [17] highlighted the adaptability of ResNet to various datasets, instructing ResNet to generate feature maps and combining them into a multi-level feature fusion network for more accurate defect localization on steel plates.”
Strip Surface Defect Detection Algorithm Based on Improved YOLO11
Cao, J., & Liu, Z. (2025) · Chinese Control and Decision Conference (CCDC) · doi:10.1109/ccdc65474.2025.11090264 · Cited by 3
Wang, H., Zhang, C., Kang, W., Liang, X., Cao, L., Yang, C., & Gui, W. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129290 · Cited by 4
An industrial small surface defect detection technique based on CEYolo model
Li, B., Wang, B., & Hu, X. (2025) · Cluster Computing · doi:10.1007/s10586-025-05324-x · Cited by 3
“Deep learning-based methods are primarily categorized into two broad categories: single-stage [12] and two-stage algorithms.”
EIM-YOLO: A Defect Detection Method for Metal-Painted Surfaces on Electrical Sealing Covers
Wu, Z., & Yang, L. (2025) · Applied Sciences · doi:10.3390/app15179380 · Cited by 4
Region Detection Rate: An Applied Measure for Surface Defect Localization
Huxohl, T., & Kümmert, F. (2021) · IEEE International Conference on Signal and Image Processing Applications (ICSIPA) · doi:10.1109/icsipa52582.2021.9576810 · Cited by 3
“Surface defect detection is important to ensure the quality of industrially manufactured products, such as steel plates [1],”
BackgroundDetecting LED Chip Surface Defects with Modified Faster R-CNN
Zhang, Z., Gong, Q., Cao, Y., et al. (2021) · International SoC Design Conference · doi:10.1109/isocc53507.2021.9613949 · Cited by 3
“FPNs use the inherent pyramid hierarchical structure of the deep convolutional network to perform feature fusion at different levels so that the feature map can retain low-latitude features, and at the same time solve the problem of excessive calculation [6].”
BackgroundPDD-Net: A Precise Defect Detection Network Based on Point Set Representation
Ban, M., Ding, R., Zhang, J., et al. (2022) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp43922.2022.9747492 · Cited by 3
Relational-based transfer learning for automatic optical inspection based on domain discrepancy
Salgado, E. I. V., Yan, H., Hong, Y., Zhu, P., Zhu, S., Liao, C., Wen, Y., Li, X., Qian, X., Wang, X., et al. (2022) · Optoelectronic Imaging and Multimedia Technology · doi:10.1117/12.2644087 · Cited by 3
“Clearly, the previous publications show how authors modify models without considering real-world data quality, distribution, and similarity among samples [7] [8].”
BackgroundAn Efficient End-to-End Multitask Network Architecture for Defect Inspection
Zhang, C., Yang, H., Ma, J., & Chen, H. (2022) · Sensors · doi:10.3390/s22249845 · Cited by 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.