An Unsupervised-Learning-Based Approach for Automated Defect Inspection on Textured Surfaces
The DOI matches its Crossref record; 50 of the 51 references it cites were found in a registry or an index.
Abstract
Automated defect inspection has long been a challenging task especially in industrial applications, where collecting and labeling large amounts of defective samples are usually harsh and impracticable. In this paper, we propose an approach to detect and localize defects with only defect-free samples for model training. This approach is carried out by reconstructing image patches with convolutional denoising autoencoder networks at different Gaussian pyramid levels, and synthesizing detection results from these different resolution channels. Reconstruction residuals of the training patches are used as the indicator for direct pixelwise defect prediction, and the reconstruction residual map generated in each channel is combined to generate the final inspection result. This novel method has two prominent characteristics, which benefit the implementation of automatic defect inspection in practice. First, it is absolutely unsupervised that no human intervention is needed throughout the inspection process. Second, multimodal strategy is utilized in this method to synthesize results from multiple pyramid levels. This strategy is capable of improving the robustness and accuracy of the method. To evaluate this approach, experiments on convergence, noise immunity, and defect inspection accuracy are conducted. Furthermore, comparative tests with some excellent algorithms on actual and simulated data sets are performed. Experimental results demonstrated the effectiveness and superiority of the proposed method on homogeneous and nonregular textured surfaces.
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
“Moreover, AE-based algorithms also demonstrate strong competitiveness in steel surface defect detection, which are reported to be fairly noise-robust.”
From Automated Visual Defect Detection for Flat Steel Surface: A Survey (2020) · cites this paper as [114]
In the citing paper
“Mei et al. [114] utilized convolutional denoising AE network to reconstruct image patches, combined with the reconstruction residual maps, this scheme can reliably learning final detection results, where no manual intervention is needed throughout all the detection process.”
From Automated Visual Defect Detection for Flat Steel Surface: A Survey (2020) · cites this paper as [114]
Citations
Cites
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“The packages involved in the implementation mainly include Keras [47], Theano [48], Scikit-Learn [49], Numpy [50], and Scipy [51].”
MethodThe NumPy Array: A Structure for Efficient Numerical Computation
van der Walt, S., Colbert, S. C., & Varoquaux, G. (2011) · Computing in Science & Engineering · doi:10.1109/mcse.2011.37 · Cited by 9,381 · cited 1×
“The packages involved in the implementation mainly include Keras [47], Theano [48], Scikit-Learn [49], Numpy [50], and Scipy [51].”
Method(2010) · Journal of Machine Learning Research · doi:10.5555/1756006.1953039 · Cited by 4,051 · cited 1×
“This is done by first corrupting the initial input x into x̃ by means of a stochastic mapping x̃ ∼ q(x̃|x) [29].”
MethodPractical Recommendations for Gradient-Based Training of Deep Architectures
Bengio, Y. (2012) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-35289-8_26 · Cited by 2,259 · cited 1×
“As for the learning rate α, a heuristic principle is applied by choosing a value close to the largest learning rate that does not cause divergence [42].”
MethodStatic Analysis of Shape in TensorFlow Programs
The Theano Development Team, Al‐Rfou, R., Alain, G., Almahairi, A., Angermueller, C., Bahdanau, D., Ballas, N., Bastien, F., Bayer, J., Belikov, A., et al. (2016) · arXiv (Cornell University) · doi:10.4230/lipics.ecoop.2020.15 · Cited by 1,589 · cited 1×
“The packages involved in the implementation mainly include Keras [47], Theano [48], Scikit-Learn [49], Numpy [50], and Scipy [51].”
MethodRefaeilzadeh, P., Tang, L., & Liu, H. (2009) · Encyclopedia of Database Systems · doi:10.1007/978-0-387-39940-9_565 · Cited by 1,549
Adaptive deconvolutional networks for mid and high level feature learning
Zeiler, M. D., Taylor, G. W., & Fergus, R. (2011) · 2011 International Conference on Computer Vision · doi:10.1109/iccv.2011.6126474 · Cited by 1,217
Salt-and-pepper noise removal by median-type noise detectors and detail-preserving regularization
Chan, R. H., Chung-Wa, & Nikolova, M. (2005) · IEEE Transactions on Image Processing · doi:10.1109/tip.2005.852196 · Cited by 1,152 · cited 1×
“c) Noise corruption: As stated, the salt and pepper noise [34] is utilized for data corruption in the MSCDAE”
MethodWLD: A Robust Local Image Descriptor
Chen, J., Shan, S., He, C., Zhao, G., Pietikäinen, M., Chen, X., & Gao, W. (2010) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2009.155 · Cited by 1,024
Weimer, D., Scholz-Reiter, B., & Shpitalni, M. (2016) · CIRP Annals · doi:10.1016/j.cirp.2016.04.072 · Cited by 577 · cited 1×
“Weimer et al. [23] and Park et al. [24] also proposed novel deep CNN architectures, which obviously improve the accuracy of automated defect detection.”
Background
Research works printed but not linked: 4
- mario fritz, 2017, The KTH-TIPS and KTH-TIPS2 Image Databases
- jones, 2014, SciPy Open source scientific tools for Python
- al-rfou, 2016, Theano A Python framework for fast computation of mathematical expressions
- chollet, 2017, Keras
Cited by
Items 101–200 of 323
Textile defect detection using multilevel and attentional deep learning network (MLMA-Net)
Wei, B., Xu, B., Hao, K., & Gao, L. (2022) · Textile Research Journal · doi:10.1177/00405175211073773 · Cited by 26
Anomaly Composition and Decomposition Network for Accurate Visual Inspection of Texture Defects
Song, K., Yang, H. S., & Yin, Z. (2022) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2022.3196133 · Cited by 26
Self-citationChang-lin, C., Wu, Q., Zhang, J., Xia, H., Lin, P., Wang, Y., Tian, M., & Song, R. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3386210 · Cited by 26
“Mei et al. [17] trained the model on defect-free samples using a multiscale convolutional denoising autoencoder and Gaussian pyramid.”
MethodZheng, X., Liu, W., & Huang, Y. (2025) · Measurement · doi:10.1016/j.measurement.2025.117393 · Cited by 26
Online Yarn Breakage Detection: A Reflection-Based Anomaly Detection Method
Yan, N., Zhu, L., Yang, H., Li, N., & Zhang, X. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3071227 · Cited by 25
Self-citationUnsupervised anomaly detection and localization with one model for all category
Tan, P., & Wong, W. K. (2024) · Knowledge-Based Systems · doi:10.1016/j.knosys.2024.111533 · Cited by 25
Zhang, H., Liu, S., Wang, C., Lu, S., & Xiong, W. (2023) · The Journal of Supercomputing · doi:10.1007/s11227-023-05639-9 · Cited by 24
“Mei et al. [8] proposed a multi-scale denoising convolutional auto-encoder (MSDCAE) model to detect and locate defect areas through residual analysis, but the performance of this method is not good when the fabric textures become complex.”
Cloud-GAN: Cloud Generation Adversarial Networks for anomaly detection
Zeng, X., Zhuo, Y., Liao, T., & Guo, J. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110866 · Cited by 24
Autolabeling-Enhanced Active Learning for Cost-Efficient Surface Defect Visual Classification
Yang, H. S., Song, K., Mao, F., & Yin, Z. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2020.3032190 · Cited by 24
“Next, visual detection [9], [10] is adopted to locate defect regions in the image.”
MethodSelf-citationAutomatic Rail Component Detection Based on AttnConv-Net
Wang, T., Zhang, Z., Yang, F., & Tsui, K. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3132460 · Cited by 23
The survey of industrial anomaly detection for industry 5.0
Wen, L., Zhang, Y., Hu, W., & Li, X. (2024) · International Journal of Computer Integrated Manufacturing · doi:10.1080/0951192x.2024.2397821 · Cited by 23
“Mei, Yang, and Yin (2018) 2021) maximized the reconstruction error by varying the granularity information of the features and fusing multi-scale information, which in turn improved the discrimination of subtle anomalies.”
On the Generalization of Handwritten Text Recognition Models
Garrido-Munoz, C., & Calvo-Zaragoza, J. (2025) · IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings - IEEE Computer Society Conference on Computer Vision and Pattern Recognition/Proceedings · doi:10.1109/cvpr52734.2025.01423 · Cited by 23
“Following the reconstruction-based literature [14, 15, 53], we measure whether a sample of a target T is out-of-distribution based on the reconstruction error of an Autoencoder (AE) trained on a source S that we denote as � ✓ S .”
Mei, S., Cheng, J., He, X., Hu, H., & Wen, G. (2021) · IEEE Sensors Journal · doi:10.1109/jsen.2021.3131908 · Cited by 22
Self-citationAccumulated and aggregated shifting of intensity for defect detection on micro 3D textured surfaces
Yan, Y., Kaneko, S., & Asano, H. (2019) · Pattern Recognition · doi:10.1016/j.patcog.2019.107057 · Cited by 22
Adibhatla, V. A., Huang, Y., Chang, M.-C., Kuo, H.-C., Utekar, A., Chih, H.-C., Abbod, M., & Shieh, J.-S. (2021) · Electronics · doi:10.3390/electronics10243177 · Cited by 22
A simulation-based few samples learning method for surface defect segmentation
Wei, T., Cao, D., Zheng, C., & Yang, Q. (2020) · Neurocomputing · doi:10.1016/j.neucom.2020.06.090 · Cited by 21
Zhao, Z., Hu, B.-T., Feng, Y.-X., Zhao, B., Yang, C., Hong, Z.-X., & Tan, J.-R. (2022) · International Journal of Production Research · doi:10.1080/00207543.2022.2138613 · Cited by 22
“Unsuper-vised learning methods can realise effective surface defect detection with unlabelled samples, but they are susceptible to image noise (Mei, Yang, and Yin 2018).”
BackgroundBearing Defect Detection with Unsupervised Neural Networks
Xu, J., Zuo, Z., Danchao, W., Li, B., Li, X., & Kong, D. (2021) · Shock and Vibration · doi:10.1155/2021/9544809 · Cited by 20
“Mei et al. [11, 12] adopt the idea of image pyramid hierarchy and convolutional denoising autoencoder network to realize defect detection of cloth texture images. results show that full use of unsupervised learning and multimodal result fusion strategy can improve the robustness and accuracy of…”
BackgroundZhang, H., Yuan, X., Li, D., You, J., Liu, B., Zhao, X., Cai, W.-M., & Ju, S. (2022) · Applied Intelligence · doi:10.1007/s10489-021-02981-4 · Cited by 20
“Image-based defect diagnosis methods are now widely used in textured fabrics [20], metallic surfaces [6], magnetic tiles [16], electronic components [5, 29], nanofiber materials [21] and others.”
BackgroundMethodGraph-Based Active Learning With Uncertainty and Representativeness for Industrial Anomaly Detection
Xiao, K., Cao, J., Zeng, Z., & Ling, B. W. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3279422 · Cited by 20
“Unsupervised learning methods [7], [8], [9] learn models without labeling information and heavily rely on a priori assumptions about data distribution.”
MethodEnhancing surface detection: A comprehensive analysis of various YOLO models
Raj, G. D., & Prabadevi, B. (2025) · Heliyon · doi:10.1016/j.heliyon.2025.e42433 · Cited by 20
“Mei et al. [22] used reconstructed residual maps and a convolutional AE network to reduce noise, thus achieving successful recognition without human involvement.”
Sun, T., Guo, B., Ruan, T., Zhou, X., bai, D., Yu, H., & Wang, Y. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127005 · Cited by 20
Fan, Y., Li, Y., Shi, Y., & Wang, S. (2022) · KSII Transactions on Internet and Information Systems · doi:10.3837/tiis.2022.01.014 · Cited by 20
“With the development of deep learning technology in the field of computer vision, fruitful research results regarding defect detection have been achieved in recent years [12-17].”
BackgroundRDAD: A reconstructive and discriminative anomaly detection model based on transformer
Xie, X., Huang, Y., Ning, W., Wu, D., Li, Z., & Yang, H. (2022) · International Journal of Intelligent Systems · doi:10.1002/int.22974 · Cited by 20
“Finally, the anomalous image Ma, the anomaly source image D, and the input image X are mixed to generate the anomalous image Xa which is just‐out‐of the normal distribution, and the definition of Xa is shown in Equation (7).”
Self-citationA Fast Surface Defect Detection Method Based on Background Reconstruction
Lv, C., Zhang, Z., Shen, F., Zhang, F., & Su, H. (2019) · International Journal of Precision Engineering and Manufacturing · doi:10.1007/s12541-019-00262-2 · Cited by 20
Small Defect Detection Based on Local Structure Similarity for Magnetic Tile Surface
Zhong, Z., Wang, H., & Xiang, D. (2022) · Electronics · doi:10.3390/electronics12010185 · Cited by 19
Adaptive vision inspection for multi-type electronic products based on prior knowledge
Zhao, D., Xue, D., Wang, X., & Du, F. (2021) · Journal of Industrial Information Integration · doi:10.1016/j.jii.2021.100283 · Cited by 19
Zheng, X., Liu, W., & Huang, Y. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112179 · Cited by 18
Cheng, J., Wen, G., He, X., Liu, X., Hu, Y., & Mei, S. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3334368 · Cited by 18
“To solve this issue, many researchers turned their attention to the background reconstruction-based methods, which mainly adopt the autoencoder (AE) [10], [11], [12].”
BackgroundSelf-citationXuezhen, L. I., Jing, J., Bao, J., Lu, P., Xie, Y., & An, Y. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3280508 · Cited by 18
“According to the rich experience in the past [16], [17], the application of anomaly detection algorithms for high-dimensional data is rising in industrial field [18].”
BackgroundAnomalySeg: Deep Learning-Based Fast Anomaly Segmentation Approach for Surface Defect Detection
Song, Y., Xia, W., Li, Y., Li, H., Yuan, M., & Zhang, Q. (2024) · Electronics · doi:10.3390/electronics13020284 · Cited by 17
“Mei et al. [39] proposed an unsupervised learning model that only needs defect-free samples to implement defect detection.”
DAS3D: Dual-Modality Anomaly Synthesis for 3D Anomaly Detection
Li, K., Dai, B., Fu, J., & Hou, X. (2025) · Lecture notes in computer science · doi:10.1007/978-3-031-92805-5_10 · Cited by 17
One-class anomaly detection through color-to-thermal AI for building envelope inspection
Kurtser, P., Feng, K., Olofsson, T., & Andrés, A. D. (2024) · Energy and Buildings · doi:10.1016/j.enbuild.2024.115052 · Cited by 16
Dual-Branch Knowledge Distillation via Residual Features Aggregation Module for Anomaly Segmentation
Zhou, Y., Huang, Z., Zeng, D., Qu, Y., & Wu, Z. (2024) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2024.3507036 · Cited by 16
“Among them, the reconstruction methods based on autoencoders (AEs) [8], [9], [10] and generative adversarial networks (GANs) [11], [12] have been well studied, but this type of generative model has the problem of disappearing reconstruction errors and is prone to unstable convergence of training,…”
Zhu, L., Baolin, D., Xiaomeng, Z., Fang, S., Zhen, C., Zhou, J., & Shumin, C. (2022) · Scientific Programming · doi:10.1155/2022/4481495 · Cited by 16
Xiao, S., Liu, Z., Yan, Z., & Wang, M. (2023) · Sensors · doi:10.3390/s23094563 · Cited by 16
“Mei et al. [14] reconstructed image blocks using convolutional denoising autoencoder networks at different levels of Gaussian pyramid and integrated the detection results from these different resolution channels.”
BackgroundMethodKnowledge Graph-guided Convolutional Neural Network for Surface Defect Recognition
Wang, Y., Gao, L., Gao, Y., Li, X., & Gao, L. (2020) · IEEE International Conference on Automation Science and Engineering (CASE) · doi:10.1109/case48305.2020.9216752 · Cited by 16
He, K., Liu, X., Liu, J., & Wu, P. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3112784 · Cited by 16
Siamese Transition Masked Autoencoders as Uniform Unsupervised Visual Anomaly DetectorPreprint
Yao, H., Wang, X., & Yu, W. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2211.00349 · Cited by 16
“For instance, an MSCDAE [24] model is proposed to localize textural defects using a three-layer pyramid structure.”
MethodActive Jamming Mitigation for Short-Range Detection System
Chen, K., Zhang, J., Chen, S., Zhang, S., & Zhao, H. (2023) · IEEE Transactions on Vehicular Technology · doi:10.1109/tvt.2023.3266380 · Cited by 15
Ma, Q., Zhang, E., Chen, Y., Duan, J., & Shao, L. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2023.107131 · Cited by 15
Generalizable Industrial Visual Anomaly Detection with Self-Induction Vision TransformerPreprint
Yao, H., & Yu, W. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2211.12311 · Cited by 15
“For example, in [6], the MSCDAE is proposed to apply a three-layer pyramid denoising AE for textural surface defect detection.”
MethodChip Appearance Defect Recognition Based on Convolutional Neural Network
Wang, J., Zhou, X., & Wu, J. (2021) · Sensors · doi:10.3390/s21217076 · Cited by 14
“[15] proposed a novel approach known as multiscale convolutional denoising autoencoder, which had different Gaussian pyramid levels for detecting and localizing defects with defect-free samples for model training; the approach is effective for homogeneous and nonregular textured surfaces.”
BackgroundColour‐patterned fabric‐defect detection using unsupervised and memorial defect‐free features
Zhang, H., Zhang, W., Wang, Y., Lu, S., Yao, L., & Chen, X. (2022) · Coloration Technology · doi:10.1111/cote.12624 · Cited by 14
Tao, X., Yan, S., Gong, X., & Adak, C. (2022) · IEEE Transactions on Artificial Intelligence · doi:10.1109/tai.2022.3227142 · Cited by 14
“Compared to the existing multiscale architectures, the proposed method can avoid the imprecision of image reconstruction [11], [12], [13], [14], [15], [16], [17], [18], [19], [20], and reduce the complex operation of feature alignment [25] and patch setting [26].”
MethodInfluentialZhang, H., Qiao, G., Liu, S., Lyu, Y., Yao, L., & Ge, Z. (2022) · Coloration Technology · doi:10.1111/cote.12644 · Cited by 14
Visual Defect Inspection for Deep-Aperture Components With Coarse-to-Fine Contour Extraction
Gong, X., Su, H., Xu, D., Zhang, J., Zhang, L., & Zhang, Z. (2020) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2019.2928347 · Cited by 14
“According to previous publications, shape recognition methods for VBM can be roughly divided into two categories: those that depend upon recognition of feature points or landmarks [4], [7] and those that depend on the contour and skeleton of the object [1]–[3], [5], [8], [9].”
MethodP2 random walk: self-supervised anomaly detection with pixel-point random walk
Hua, L., Qi, Q., & Long, J. (2023) · Complex & Intelligent Systems · doi:10.1007/s40747-023-01285-z · Cited by 14
“The autoencoding methods construct symmetric networks using encoders and decoders in the reconstruction model [29].”
Unsupervised surface defect detection using dictionary-based sparse representation
Meng, F., Gong, T., Wu, D., & Xiang, X. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.110020 · Cited by 13
Li, Z., Ge, Y., & Meng, L. (2025) · Information Fusion · doi:10.1016/j.inffus.2025.103356 · Cited by 13
Wang, J., Yi, G., Zhang, S., & Wang, Y. (2020) · Applied Sciences · doi:10.3390/app11010283 · Cited by 13
“In order to evaluate the effectiveness of our method, we carried out comparative experiments between the proposed method and three other unsupervised methods: PHOT [37], AnoGAN [31] and MSCDAE [30].”
MethodDetection and segmentation of image anomalies based on unsupervised defect reparation
Dai, W., Erdt, M., & Sourin, A. (2021) · The Visual Computer · doi:10.1007/s00371-021-02257-5 · Cited by 13
Image Restoration for Blast Furnace Burden Surface Based on Dust Multiscattering Model
Li, Y., Jiang, Z., Yi, Z., Pan, D., Yang, B., & Gui, W. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3284022 · Cited by 13
“recognition [2], defect detection [3], [4], and fault diagno-”
BackgroundZhang, R., Wang, H., Feng, M., Liu, Y., & Yang, G. (2023) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2023.3295468 · Cited by 13
“Current deep learning-based approaches for anomaly detection were usually based on reconstruction-based neural network architectures, such as AEs [6], [21], [22], variational AEs (VAEs) [23], and GANs [7], [24].”
BackgroundHu, P., He, C., Zhu, Y., & Li, T. (2023) · Journal of Cloud Computing Advances Systems and Applications · doi:10.1186/s13677-023-00487-7 · Cited by 13
“The machine vision methods have been widely adopted in quality inspection scenarios [13].”
MethodYu, Z., Xu, Y., Wang, Y., Wang, Y., & Sheng, X. (2023) · Coloration Technology · doi:10.1111/cote.12744 · Cited by 13
A Self-Supervised CNN for Particle Inspection on Optical Element
Hou, W. S., Tao, X., & Xu, D. (2021) · IEEE Transactions on Instrumentation and Measurement · doi:10.1109/tim.2021.3077994 · Cited by 13
“Mei et al. [22] proposed an unsupervised multiscale convolutional denoising autoencoder (MSCDAE) model that used multilevel scale defect-free images to build autoencoders and detect defects by comparing the reconstructed with original images.”
MethodA high-effective multitask surface defect detection method based on CBAM and atrous convolution
Xie, X., Lei, X. U., Xinlei, L. I., Wang, B., & Wan, T. (2022) · Journal of Advanced Mechanical Design Systems and Manufacturing · doi:10.1299/jamdsm.2022jamdsm0063 · Cited by 12
“It combines the idea of an image pyramid hierarchy with convolutional denoising autoencoder (CDAE) networks to perform defect detection on texture images (Mei et al., 2018).”
Printed Surface Defect Detection Model Based on Positive Samples
Xin, Z., Hongyuan, W., Pengyu, Q., Weidong, D., Zhang, J., & Fuhua, C. (2022) · Computers, materials & continua/Computers, materials & continua (Print) · doi:10.32604/cmc.2022.026943 · Cited by 12
An Aircraft Object Detection Algorithm Based on Small Samples in Optical Remote Sensing Image
Wang, T., Cao, C. C., Zeng, X., Feng, Z., Shen, J., Li, W., Wang, B., Zhou, Y., & Xu, Y. H. (2020) · Applied Sciences · doi:10.3390/app10175778 · Cited by 11
Zhang, Y., Han, T., Wei, B., Hao, K., & Gao, L. (2023) · Complex & Intelligent Systems · doi:10.1007/s40747-023-01070-y · Cited by 11
“Subsequently, Mei et al. [19] proposed a multiscale convolutional denoising autoencoder to detect and localize defects only using defect-free samples for model training.”
BackgroundEnhanced Non-Maximum Suppression for the Detection of Steel Surface Defects
Kang, S., Palakonda, V., Kim, I., Kang, J., & Yun, S. (2023) · Mathematics · doi:10.3390/math11183898 · Cited by 11
Multi-defect detection and classification for aluminum alloys with enhanced YOLOv8
Han, Y., Li, X., Cui, G., Song, J., Zhou, F., & Wang, Y. (2025) · PLoS ONE · doi:10.1371/journal.pone.0316817 · Cited by 11
“The automatic defect detection algorithm proposed by Mei et al. [48] performs better on defect-free samples, yet is not applicable to the defect dataset in this study compared to the YOLOv8-ALGP algorithm.”
Han, X., & Huang, Z. R. (2021) · Materials Letters · doi:10.1016/j.matlet.2021.130065 · Cited by 11
Unsupervised defect inspection algorithm based on cascaded GAN with edge repair feature fusion
He, L., Shi, N., Malik, K., & Li, F. (2021) · Applied Intelligence · doi:10.1007/s10489-021-02556-3 · Cited by 11
A robust defect detection method for syringe scale without positive samples
Wang, X., Xu, X., Wang, Y., Wu, P., Yan, F., & Zeng, Z. (2022) · The Visual Computer · doi:10.1007/s00371-022-02671-3 · Cited by 11
“To relieve the difficulty of collecting and labeling defective samples, some unsupervised defect detection methods based on image reconstruction [23,24] or image reparation [25,26] are proposed.”
MethodShancheng, T., Jin, Z., Zhang, Y., Lu, J., Li, H., & Jiqing, Y. (2023) · Processes · doi:10.3390/pr11092615 · Cited by 11
Li, Y., He, L., Zhang, M., Cheng, Z., Liu, W., & Wu, Z. (2023) · Electronics · doi:10.3390/electronics12112440 · Cited by 10
Learning the Color Space for Effective Fabric Defect Detection
Zhang, C., Qi, Y., & Wang, Y. (2023) · IEEE Transactions on Emerging Topics in Computational Intelligence · doi:10.1109/tetci.2023.3320565 · Cited by 10
“Convolutional Auto-Encoders (CAEs) are commonly used to obtain better image reconstruction quality [16], [27], [28], while methods such as [29], [30] use Generative Adversarial Networks (GANs) [31] to generate more realistic textures using the same reconstruction heuristics.”
MethodJeong, K., Shin, D. Y., Park, J., Yi, D., Hong, H., Kim, H., & Chung, K. (2024) · Advanced Science · doi:10.1002/advs.202407923 · Cited by 10
Weakly Supervised Surface Defect Detection Based on Attention Mechanism
Sun, M., Lyu, C., Han, Y., Li, S., & Wang, Z. (2021) · Journal of Computer-Aided Design & Computer Graphics · doi:10.3724/sp.j.1089.2021.18594 · Cited by 10
Deep learning object detection applied to defect recognition of memory modules
Huang, J.-T., & Ting, C.-H. (2022) · The International Journal of Advanced Manufacturing Technology · doi:10.1007/s00170-022-09716-w · Cited by 10
“In terms of the matching matrix [24], the binary matching matrix contains four elements: true positives (TPs), true negatives (TNs), false positives (FPs), and false negatives (FNs).”
BackgroundLuo, W., Niu, T., Yao, H., Tang, L., Yu, W., & Li, B. (2023) · IEEE Sensors Journal · doi:10.1109/jsen.2023.3276762 · Cited by 10
“Mei et al. [21] proposed MSCDAE, which uses a Gaussian pyramid structure to obtain different receptive fields to generate more realistic background images.”
BackgroundA Hierarchical Feature Fusion-based Method for Defect Recognition with a Small Sample
Gao, Y., Gao, L., & Li, X. (2019) · IEEE International Conference on Industrial Engineering and Engineering Management (IEEM) · doi:10.1109/ieem44572.2019.8978912 · Cited by 10
“[12] proposed an unsupervised method for a textured surface.”
InfluentialA Real-time Unsupervised Two-stage Framework for Fabric Defect Detection
Xia, D., Yu, Z., & Deng, X. (2021) · International Academic Exchange Conference on Science and Technology Innovation (IAECST) · doi:10.1109/iaecst54258.2021.9695639 · Cited by 10
PNG: an adaptive local-global hybrid framework for unsupervised material surface defect detection
Sun, Q., Xu, K., Zhao, D., Li, H., Jin, L., Liu, C., & Xu, P. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.128711 · Cited by 9
Component-Aware Unsupervised Logical Anomaly Generation for Industrial Anomaly Detection
Tong, X., Chang, Y., Zhao, Q., Yu, J., Wang, B., Lin, J., Lin, Y., Mai, X., Wang, H., Tao, Z., et al. (2025) · Proceedings - IEEE International Conference on Robotics and Automation/Proceedings · doi:10.1109/icra55743.2025.11127296 · Cited by 9
“Visual anomaly detection is fundamental to ensuring quality control in production and serves as a key enabler of industrial automation and evolution of intelligent manufacturing systems [1]–[3].”
Li, L., Wang, Y., Qi, J., Xiao, S., & Gao, H. (2022) · Polymers · doi:10.3390/polym14091855 · Cited by 9
“Comparison of different recognition methods (MSCDAE [40], SDAE [41], LSTS [30], BB [23], ER [25], RB [24], WGIS [26], NN [42]).”
MethodWaveguide quality inspection in quantum cascade lasers: A capsule neural network approach
Ebayyeh, A. A. R. M. A., Mousavi, A., Danishvar, S., Blaser, S., Gresch, T., Landry, O., & Müller, A. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2022.118421 · Cited by 9
Automated Surface Defects Detection on Mirrorlike Materials by using Faster R-CNN
Karangwa, J., Kong, L., You, T., & Zheng, J. (2020) · International Conference on Information Science and Control Engineering · doi:10.1109/icisce50968.2020.00341 · Cited by 9
A Semi-Supervised Deep Learning Approach for the Classification of Steel Surface Defects
Mayuravaani, M., & Manivannan, S. (2021) · International Conference on Information and Automation for Sustainability (ICIAfS) · doi:10.1109/iciafs52090.2021.9606143 · Cited by 9
A novel approach for flip chip inspection based on improved SDELM and vibration signals
Su, L., Zhang, S., Ji, Y., Wang, G., Ming, X., Gu, J., Li, K., & PECHT, M. G. (2022) · Science China Technological Sciences · doi:10.1007/s11431-021-1964-0 · Cited by 9
“Mei et al. [20] proposed a CDAE-based fabric defect detection method, which exhibited good detection performance and stability compared to other methods.”
MethodWang, L., Huang, X., & Zheng, Z. (2023) · PLoS ONE · doi:10.1371/journal.pone.0280363 · Cited by 9
“[26] proposed a method of detecting and locating defects only after training with non-defective samples to detect fabric defects, which is mainly reconstructed by a network with a convolutional noise reduction autoencoder.”
MethodClear Memory-Augmented Auto-Encoder for Surface Defect DetectionPreprint
Luo, W., Niu, T., Tang, L., Yu, W., & Li, B. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2208.03879 · Cited by 9
“[22] proposed MSCDAE, which uses a Gaussian pyramid structure to obtain different receptive fields to generate more realistic background images.”
MethodA diverse embedding-based composite reconstruction encoder–decoder for color fabric defect detection
Zhang, H., Meng, L., Lu, S., Song, Z., & Wu, L. Z. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.127261 · Cited by 8
Cycle-CFM: An unsupervised framework for robust multimodal anomaly detection in industrial settings
yikang, S., Zhan, X., Li, Y., Wu, Z., Zhang, W., & Li, H. (2025) · Expert Systems with Applications · doi:10.1016/j.eswa.2025.129745 · Cited by 8
A New Method for Rapid Detection of Surface Defects on Complex Textured Tiles
Dong, G., Wang, Y., Liu, S., Wu, N., Kong, X., Chen, X., & Wang, Z. (2024) · Journal of Nondestructive Evaluation · doi:10.1007/s10921-024-01145-x · Cited by 8
“In addition, feature extraction operators usually rely on manually set parameters, and the types of defects and textures detected are often limited[30].”
Global-Regularized Neighborhood Regression for Efficient Zero-Shot Texture Anomaly Detection
Yao, H., Luo, W., Cao, Y., Zhang, Y., Yu, W., & Shen, W. (2025) · IEEE Transactions on Systems Man and Cybernetics Systems · doi:10.1109/tsmc.2025.3595539 · Cited by 8
“In order to reduce the labor involved in collecting defective samples and annotating data, unsupervised-learning-based texture anomaly detection methods [1], [6] shown in Fig.”
TSDNet: A New Multiscale Texture Surface Defect Detection Model
Dong, M., Li, D., Li, K., & Xu, J. (2023) · Applied Sciences · doi:10.3390/app13053289 · Cited by 8
Research on Surface Defect Detection Based on Semantic Segmentation
Liu, Y., Yang, Y., Wang, C., Xu, X., & Zhang, T. (2019) · DEStech Transactions on Computer Science and Engineering · doi:10.12783/dtcse/aicae2019/31504 · Cited by 8
“In reference [3], an automatic defect Detection method for texture Surface based on unsupervised Learning was proposed by S.”
MethodLiu, X., Wu, Q., Cheng, Y., & Wen, G. (2025) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2025.112408 · Cited by 7
SI3DP: Source Identification Challenges and Benchmark for Consumer-Level 3D Printer Forensics
Shim, B. S., Shin, Y. S., Park, S. W., & Hou, J. (2021) · ACM International Conference on Multimedia (ACM MM) · doi:10.1145/3474085.3475316 · Cited by 7
“The unsupervised learning [28] will lead to performance improvement in further research even if targeted tasks and labels do not match.”
BackgroundTaştimur, C., Karaduman, G., & Akın, E. (2021) · Innovations in Intelligent Systems and Applications Conference (ASYU) · doi:10.1109/asyu52992.2021.9598985 · Cited by 7
Vision-based LCD/OLED defect detection methods: a critical summary
Siyuan, L., & Yiquan, W. (2024) · Journal of Image and Graphics · doi:10.11834/jig.230518 · Cited by 7
Self-supervised Augmented Patches Segmentation for Anomaly Detection
Long, J., Yang, Y., Hua, L., & Ou, Y. (2023) · Lecture notes in computer science · doi:10.1007/978-3-031-26284-5_6 · Cited by 7
Tsai, C., & Chen, H. (2020) · Sensors · doi:10.3390/s20164356 · Cited by 7
“[17] proposed an unsupervised-learning-based approach to detect and localize defects with only defect-free samples for model training.”
BackgroundIndustrial object detection method based on improved CenterNet
Tang, C., Wu, Z., Wang, S., Deng, C., & Luo, L. (2021) · International Conference on Computer Engineering and Artificial Intelligence (ICCEAI) · doi:10.1109/icceai52939.2021.00023 · Cited by 7
Transferable Learning Architecture for Scalable Visual Quality Inspection
Talha, M., Rashid, S. F., Iftikhar, Z., Afzal, M. T., & Liu, Y. (2022) · doi:10.1109/icai55435.2022.9773637 · Cited by 7
DeFRCN-MAM: DeFRCN and multi-scale attention mechanism-based industrial defect detection method
Zheng, T., Sa, L., Yu, C., & Song, A. (2024) · Applied Artificial Intelligence · doi:10.1080/08839514.2024.2349981 · Cited by 7
“Mei, Yang, and Yin (2018) use the image pyramid structure to scale the image to three different sizes, which can add noise randomly.”
Wang, H., He, X., Zhang, C., Liang, X., Zhu, P., Wang, X., Gui, W., Li, X., & Qian, X. (2025) · Measurement · doi:10.1016/j.measurement.2025.116702 · Cited by 6
Metadata
- Authors
- Mei, S., Yang, H., & Yin, Z.
- Affiliations
- State Key Laboratory of Digital Manufacturing Equipment and Technology (CN)
- Venue
- IEEE Transactions on Instrumentation and Measurement
- Published
- 2018-06
- Volume
- 67
- Issue
- 6
- Pages
- 1266-1277
- ISSN
- 0018-9456 (ISSN-L)
- 1557-9662
- OpenAlex
- W2791709739
- Semantic Scholar
- 38ec86557f805ec27e18e5517a34ca43b2bfa60f
- Cited by
- 308
- References
- 51
- Citations by year
- 2026: 16 · 2025: 41 · 2024: 36 · 2023: 59 · 2022: 63 · 2021: 44 · 2020: 36 · 2019: 11 · 2018: 2
- FWCI
- 19.21
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- IEEE Transactions on Instrumentation and Measurement (publishedVersion)
- Funders
- National Science Foundation of China (51327801, 51475193)
- Major Project Foundation of Hubei Province (2016AAA009)
- Topics
- Physical Sciences › Engineering › Industrial and Manufacturing Engineering › Industrial Vision Systems and Defect Detection
- Physical Sciences › Engineering › Mechanical Engineering › Non-Destructive Testing Techniques
- Physical Sciences › Engineering › Civil and Structural Engineering › Infrastructure Maintenance and Monitoring
- Keywords
- automated defect detection, unsupervised learning, textured surfaces, convolutional denoising autoencoder, Gaussian pyramid, reconstruction residuals, defect localization, multimodal fusion, industrial inspection
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.