{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T05:14:22Z","timestamp":1783574062001,"version":"3.55.0"},"reference-count":36,"publisher":"MDPI AG","issue":"7","license":[{"start":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T00:00:00Z","timestamp":1782777600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["J. Imaging"],"abstract":"<jats:p>Knowledge distillation is a cornerstone of image anomaly detection for amplifying subtle defects via teacher\u2013student discrepancy, yet existing methods rely on feature alignment loss that causes reconstruction error confusion and degrades accuracy. To address this critical limitation, this study proposes an abnormal discrepancy-guided knowledge distillation method (DiffKD) that differentially guides student feature reconstruction through channel-level discrepancy masks, leveraging normal features as supervisory signals and abnormal discrepancy features as constraints to enhance anomaly detection performance. The approach integrates a knowledge distillation network for feature reconstruction with a segmentation network for anomaly localization, while utilizing prior anomaly samples and synthetic anomaly samples to provide real-time training data of anomalous samples. Extensive evaluations on the SUT-Crack and MVTec AD benchmarks validate the effectiveness and generalizability of our approach. On MVTec AD, it achieves 80.7% average precision (AP) and 81.9% instance-level average precision (IAP), showing competitive performance against the representative methods evaluated under the same protocol. These results not only demonstrate significant improvements in IAD accuracy but also highlight its promise for enabling real-time, automated anomaly detection in practical applications.<\/jats:p>","DOI":"10.3390\/jimaging12070291","type":"journal-article","created":{"date-parts":[[2026,6,30]],"date-time":"2026-06-30T16:56:55Z","timestamp":1782838615000},"page":"291","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Abnormal Discrepancy-Guided Knowledge Distillation for Image Anomaly Detection"],"prefix":"10.3390","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-3533-2443","authenticated-orcid":false,"given":"Zhenjun","family":"Yu","sequence":"first","affiliation":[{"name":"College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Lin","family":"Sun","sequence":"additional","affiliation":[{"name":"College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kai","family":"Wang","sequence":"additional","affiliation":[{"name":"Qingdao Star-Rising Technologies Co., Ltd., Qingdao 266000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fengxiang","family":"Jin","sequence":"additional","affiliation":[{"name":"College of Geodesy and Geomatics, Shandong University of Science and Technology, Qingdao 266000, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1968","published-online":{"date-parts":[[2026,6,30]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1049\/2024\/8821891","article-title":"A Comprehensive Investigation of Anomaly Detection Methods in Deep Learning and Machine Learning: 2019\u20132023","volume":"2024","author":"Kumari","year":"2024","journal-title":"IET Inf. Secur."},{"key":"ref_2","doi-asserted-by":"crossref","unstructured":"Liu, X., Wang, J., Leng, B., and Zhang, S. (2024). Dual-Modeling Decouple Distillation for Unsupervised Anomaly Detection. arXiv.","DOI":"10.1145\/3664647.3681669"},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"160","DOI":"10.62762\/CJIF.2024.734267","article-title":"Unsupervised Industrial Anomaly Detection Based on Feature Mask Generation and Reverse Distillation","volume":"1","author":"Qi","year":"2024","journal-title":"Chin. J. Inf. Fusion"},{"key":"ref_4","first-page":"5640","article-title":"Unlocking the Potential of Reverse Distillation for Anomaly Detection","volume":"39","author":"Liu","year":"2025","journal-title":"Proc. AAAI Conf. Artif. Intell."},{"key":"ref_5","doi-asserted-by":"crossref","first-page":"109642","DOI":"10.1016\/j.dib.2023.109642","article-title":"SUT-crack: A comprehensive dataset for pavement crack detection across all methods","volume":"51","author":"Sabouri","year":"2023","journal-title":"Data Brief."},{"key":"ref_6","doi-asserted-by":"crossref","unstructured":"Bergmann, P., Fauser, M., Sattlegger, D., and Steger, C. (2019). MVTec AD: A comprehensive real-world dataset for unsupervised anomaly detection. 2019 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR.2019.00982"},{"key":"ref_7","doi-asserted-by":"crossref","unstructured":"Rudolph, M., Wehrbein, T., Rosenhahn, B., and Wandt, B. (2022). Fully convolutional cross-scale-flows for image-based defect detection. Proceedings of the IEEE\/CVF Winter Conference on Applications of Computer Vision, IEEE.","DOI":"10.1109\/WACV51458.2022.00189"},{"key":"ref_8","doi-asserted-by":"crossref","unstructured":"Rudolph, M., Wandt, B., and Rosenhahn, B. (2021). Same same but DifferNet: Semi-supervised defect detection with normalizing flows. 2021 IEEE Winter Conference on Applications of Computer Vision (WACV), IEEE.","DOI":"10.1109\/WACV48630.2021.00195"},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Roth, K., Pemula, L., Zepeda, J., Sch\u00f6lkopf, B., Brox, T., and Gehler, P. (2022). Towards total recall in industrial anomaly detection. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR52688.2022.01392"},{"key":"ref_10","doi-asserted-by":"crossref","unstructured":"Defard, T., Setkov, A., Loesch, A., and Audigier, R. (2021). Padim: A patch distribution modeling framework for anomaly detection and localization. International Conference on Pattern Recognition, Springer.","DOI":"10.1007\/978-3-030-68799-1_35"},{"key":"ref_11","doi-asserted-by":"crossref","unstructured":"Bergmann, P., Fauser, M., Sattlegger, D., and Steger, C. (2020). Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, IEEE.","DOI":"10.1109\/CVPR42600.2020.00424"},{"key":"ref_12","doi-asserted-by":"crossref","unstructured":"Deng, H., and Li, X. (2022). Anomaly Detection via Reverse Distillation from One-Class Embedding. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR52688.2022.00951"},{"key":"ref_13","doi-asserted-by":"crossref","unstructured":"Salehi, M., Sadjadi, N., Baselizadeh, S., Rohban, M.H., and Rabiee, H.R. (2021). Multiresolution knowledge distillation for anomaly detection. 2021 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR46437.2021.01466"},{"key":"ref_14","doi-asserted-by":"crossref","unstructured":"Wang, G., Han, S., Ding, E., and Huang, D. (2021). Student-teacher feature pyramid matching for anomaly detection. arXiv.","DOI":"10.5244\/C.35.349"},{"key":"ref_15","unstructured":"Yamada, S., and Hotta, K. (2022). Reconstruction student with attention for student-teacher pyramid matching. arXiv."},{"key":"ref_16","doi-asserted-by":"crossref","unstructured":"Zhang, X., Li, S., Li, X., Huang, P., Shan, J., and Chen, T. (2023). DeSTSeg: Segmentation guided denoising student-teacher for anomaly detection. 2023 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR52729.2023.00381"},{"key":"ref_17","doi-asserted-by":"crossref","first-page":"4023","DOI":"10.1007\/s00371-024-03412-4","article-title":"Revisiting segmentation-guided denoising student-teacher in anomaly detection","volume":"40","author":"Zang","year":"2024","journal-title":"Vis. Comput."},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Liu, X., Wang, J., Leng, B., and Zhang, S. (2024). Dual-Modeling Decouple Distillation for Unsupervised Anomaly Detection. Proceedings of the 32nd ACM International Conference on Multimedia (MM \u201924), Association for Computing Machinery.","DOI":"10.1145\/3664647.3681669"},{"key":"ref_19","doi-asserted-by":"crossref","first-page":"131423","DOI":"10.1016\/j.neucom.2025.131423","article-title":"KD-KI: Knowledge distillation with knowledge infusion for anomaly detection and localization","volume":"656","author":"Huang","year":"2025","journal-title":"Neurocomputing"},{"key":"ref_20","doi-asserted-by":"crossref","first-page":"112761","DOI":"10.1016\/j.patcog.2025.112761","article-title":"Efficient industrial anomaly detection via cross-scale distillation with enhanced feature compression","volume":"173","author":"Hu","year":"2026","journal-title":"Pattern Recognit."},{"key":"ref_21","doi-asserted-by":"crossref","first-page":"167","DOI":"10.5614\/itbj.ict.res.appl.2023.17.2.3","article-title":"Improving robustness using MixUp and CutMix augmentation for corn leaf diseases classification based on ConvMixer architecture","volume":"17","author":"Li","year":"2023","journal-title":"J. ICT Res. Appl."},{"key":"ref_22","unstructured":"DeVries, T., and Taylor, G.W. (2017). Improved regularization of convolutional neural networks with cutout. arXiv."},{"key":"ref_23","doi-asserted-by":"crossref","unstructured":"Yun, S., Han, D., Chun, S., Oh, S.J., Yoo, Y., and Choe, J. (2019). CutMix: Regularization strategy to train strong classifiers with localizable features. 2019 IEEE\/CVF International Conference on Computer Vision (ICCV), IEEE.","DOI":"10.1109\/ICCV.2019.00612"},{"key":"ref_24","doi-asserted-by":"crossref","first-page":"313","DOI":"10.1007\/s00530-025-01941-z","article-title":"Multi-modal semi-supervised semantic segmentation for indoor scenes via adaptive CutMix and contrastive learning","volume":"31","author":"Lyu","year":"2025","journal-title":"Multimed. Syst."},{"key":"ref_25","doi-asserted-by":"crossref","first-page":"5643","DOI":"10.1007\/s00371-024-03743-2","article-title":"Industrial surface defect detection by multi-scale Inpainting-GAN","volume":"41","author":"Liu","year":"2025","journal-title":"Vis. Comput."},{"key":"ref_26","unstructured":"Odena, A., Olah, C., and Shlens, J. (2016). Conditional image synthesis with auxiliary classifier GANs. arXiv."},{"key":"ref_27","unstructured":"Chen, X., Duan, Y., Houthooft, R., Schulman, J., Sutskever, I., and Abbeel, P. (2016). InfoGAN: Interpretable representation learning by information maximizing generative adversarial nets. arXiv."},{"key":"ref_28","unstructured":"Lopez, R., Regier, J., Jordan, M.I., and Yosef, N. (2018). Information constraints on auto-encoding variational bayes. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Yan, H., Yeh, H.M., and Sergin, N. (2019). Image-based process monitoring via adversarial autoencoder with applications to rolling defect detection. 2019 IEEE 15th International Conference on Automation Science and Engineering (CASE), IEEE.","DOI":"10.1109\/COASE.2019.8843313"},{"key":"ref_30","doi-asserted-by":"crossref","unstructured":"Hu, T., Zhang, J., Yi, R., Du, Y., Chen, X., Liu, L., Wang, Y., and Wang, C. (2023). AnomalyDiffusion: Few-Shot Anomaly Image Generation with Diffusion Model. arXiv.","DOI":"10.1609\/aaai.v38i8.28696"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Jin, Y., Peng, J., He, Q., Hu, T., Wu, J., Chen, H., Wang, H., Zhu, W., Chi, M., and Liu, J. (2024). DualAnoDiff: Dual-Interrelated Diffusion Model for Few-Shot Anomaly Image Generation. arXiv.","DOI":"10.1109\/CVPR52734.2025.02832"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Gui, S., Gao, B.-B., Wang, C., and Wu, Y. (2025). Few-Shot Anomaly-Driven Generation for Anomaly Classification and Segmentation. arXiv.","DOI":"10.1007\/978-3-031-73010-8_13"},{"key":"ref_33","unstructured":"He, H., Zhang, J., Chen, H., Chen, X., Li, Z., Chen, X., Wang, Y., Wang, C., and Xie, L. (2023). DiAD: A Diffusion-based Framework for Multi-class Anomaly Detection. arXiv."},{"key":"ref_34","doi-asserted-by":"crossref","first-page":"309","DOI":"10.1145\/1015706.1015720","article-title":"GrabCut: Interactive foreground extraction using iterated graph cuts","volume":"23","author":"Rother","year":"2004","journal-title":"ACM Trans. Graph."},{"key":"ref_35","doi-asserted-by":"crossref","first-page":"1219","DOI":"10.1007\/s10618-022-00829-0","article-title":"The area under the ROC curve as a measure of clustering quality","volume":"36","author":"Jaskowiak","year":"2022","journal-title":"Data Min. Knowl. Discov."},{"key":"ref_36","doi-asserted-by":"crossref","unstructured":"He, K., Lu, Y., and Sclaroff, S. (2018). Local descriptors optimized for average precision. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR), IEEE.","DOI":"10.1109\/CVPR.2018.00069"}],"container-title":["Journal of Imaging"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/291\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,9]],"date-time":"2026-07-09T04:17:22Z","timestamp":1783570642000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2313-433X\/12\/7\/291"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,6,30]]},"references-count":36,"journal-issue":{"issue":"7","published-online":{"date-parts":[[2026,7]]}},"alternative-id":["jimaging12070291"],"URL":"https:\/\/doi.org\/10.3390\/jimaging12070291","relation":{},"ISSN":["2313-433X"],"issn-type":[{"value":"2313-433X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,6,30]]}}}