ABSTRACT
Reconstruction-based approaches attract increasing attention in unsupervised visual anomaly detection. Such methods rely on the assumption that reconstruction models trained using only normal images are not well reconstructed for anomalous data, so that the anomalous regions can be identified by computing the differences in the reconstructions. However, it's usually challenging to manage the model's generalization boundaries in practice. The model with a poor generalization capability will produce additional image differences in the normal regions, while the model with an overly strong generalization capability can even well reconstruct the abnormal regions, making the abnormal regions less distinguishable from the normal regions. In this paper, we propose a two-stage reconstruction network to solve the above problems, which can generate high-accuracy and anomaly-free reconstruction. Specifically, our two-stage anomaly detection network consists of the edge extraction sub-network and the image restoration sub-network. The first sub-network generates anomaly-free grayscale edges of the image as intermediate results, and then the second one recovers normal color and low-frequency information for the intermediate results. The proposed network can generate high-accuracy and anomaly-free reconstructions to maximize the difference between the reconstruction errors of normal and abnormal samples, which can ultimately improve the accuracy of anomaly detection. Experiments on the dataset MVTec AD show that our proposed method achieves competitive results in anomaly detection and anomaly localization (97.8% for detection and 97.7% for localization, AUROC).
- Nina Shvetsova, Bart Bakker, Irina Fedulova, Heinrich Schulz, Dmitry V. Dylov. 2021. Anomaly detection in medical imaging with deep perceptual autoencoders. IEEE Access, 9, 118571-118583. Doi: 10.1109/access.2021.310716.Google ScholarCross Ref
- Clifton Phua, Vincent Lee, Kate Smith, Ross Gayler. 2010. A comprehensive survey of data mining-based fraud detection research. arXiv preprint, arXiv:1009.6119. https://doi.org/10.48550/arXiv.1009.6119.Google ScholarCross Ref
- Yajie Cui, Zhaoxiang Liu, and Shiguo Lian. 2022. A survey on unsupervised industrial anomaly detection algorithms. arXiv preprint, arXiv:2204.11161. https://doi.org/10.48550/arXiv.2204.11161.Google ScholarCross Ref
- Paul Bergmann, Sindy Löwe, Michael Fauser, David Sattlegger, Carsten Steger. 2018. Improving unsupervised defect segmentation by applying structural similarity to autoencoders. arXiv preprint, arXiv:1807.02011. https://doi.org/10.48550/arXiv.1807.02011.Google ScholarCross Ref
- Dong Gong, Lingqiao Liu, Vuong Le, Budhaditya Saha, Moussa Reda Mansour, Svetha Venkatesh, Anton van den Hengel. 2019. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision(ICCV 2019), 1705-1714. Doi: 10.1109/ICCV.2019.00179.Google Scholar
- Thomas Schlegl, Philipp Seeböck, Sebastian M. Waldstein, Ursula Schmidt-Erfurth and Georg Langs. 2017. Unsupervised anomaly detection with generative adversarial networks to guide marker discovery. In Information Processing in Medical Imaging: 25th International Conference, IPMI 2017, Boone, NC, USA, June 25-30, 2017, Proceedings. Cham, Springer International Publishing, 146-157. https://doi.org/10.1007/978-3-319-59050-9_12.Google ScholarCross Ref
- Samet Akcay, Amir Atapour-Abarghouei and Toby P. Breckon. 2019. Ganomaly: Semi-supervised anomaly detection via adversarial training. In Computer Vision-ACCV 2018: 14th Asian Conference on Computer Vision, Perth, Australia, December 2-6, 2018, Revised Selected Papers, Part III 14(ACCV 2019), Springer International Publishing, 622-637. https://doi.org/10.1007/978-3-030-20893-6_39.Google ScholarCross Ref
- Samet Akçay, Amir Atapour-Abarghouei and Toby P. Breckon. 2019. Skip-ganomaly: Skip connected and adversarially trained encoder-decoder anomaly detection. In 2019 International Joint Conference on Neural Networks (IJCNN 2019), IEEE, 1-8. Doi: 10.1109/IJCNN.2019.8851808.Google ScholarCross Ref
- Chaoqin Huang, Fei Ye, Jinkun Cao, Maosen Li, Ya Zhang, Cewu Lu. 2020. Attribute restoration framework for anomaly detection. IEEE Transactions on Multimedia, 24, 116-127. Doi: 10.1109/TMM.2020.3046884.Google Scholar
- Vitjan Zavrtanik, Matej Kristan, Danijel Skočaj. 2021. Draem-a discriminatively trained reconstruction embedding for surface anomaly detection. In Proceedings of the IEEE/CVF International Conference on Computer Vision(ICCV 2021), 8330-8339. Doi: 10.1109/ICCV48922.2021.00822.Google ScholarCross Ref
- Chunliang Li, Kihyuk Sohn, Jinsung Yoon, Tomas Pfister. 2021. Cutpaste: Self-supervised learning for anomaly detection and localization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR 2021), 9664-9674. Doi: 10.1109/CVPR46437.2021.00954.Google ScholarCross Ref
- Paul Bergmann, Michael Fauser, David Sattlegger, Carsten Steger. 2019. Mvtec ad-a comprehensive real-world dataset for unsupervised anomaly detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR 2019), 9592-9600. Doi: 10.1109/cvpr.2019.00982.Google ScholarCross Ref
- Guansong Pang, Chunhua Shen, Longbing Cao, Anton Van Den Hengel. 2021.Deep learning for anomaly detection: A review, ACM computing surveys (CSUR), 54(2), 1-38. https://doi.org/10.1145/3439950.Google ScholarDigital Library
- Raghavendra Chalapathy, Aditya Krishna Menon, Sanjay Chawla. 2018. Anomaly detection using one-class neural networks. arXiv preprint, arXiv:1802.06360. https://doi.org/10.1145/3439950.Google ScholarDigital Library
- Lukas Ruff, Robert Vandermeulen, Nico Goernitz, Lucas Deecke, Shoaib Ahmed Siddiqui, Alexander Binder, Emmanuel Müller, Marius Kloft. 2018. Deep one-class classification. In International conference on machine learning(ICML 2018), PMLR, 4393-4402. Available from https://proceedings.mlr.press/v80/ruff18a.html.Google Scholar
- Jihun Yi, Sungroh Yoon. 2020. Patch svdd: Patch-level svdd for anomaly detection and segmentation. In Proceedings of the Asian Conference on Computer Vision(ACCV 2020). https://doi.org/10.48550/arXiv.2006.16067.Google ScholarCross Ref
- Niv Cohen, Yedid Hoshen. 2020. Sub-image anomaly detection with deep pyramid correspondences. arXiv preprint, arXiv:2005.02357..https://doi.org/10.48550/arXiv.2005.02357.Google ScholarCross Ref
- Oliver Rippel, Patrick Mertens, and Dorit Merhof. 2021. Modeling the Distribution of Normal Data in Pre-Trained Deep Features for Anomaly Detection. In 2020 25th International Conference on Pattern Recognition (ICPR 2020), IEEE, 6726-6733. Doi: 10.1109/ICPR48806.2021.9412109.Google Scholar
- Shenzhi Wang, Liwei Wu, Lei Cui, Yujun Shen. 2021. Glancing at the Patch: Anomaly Localization with Global and Local Feature Comparison. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition(CVPR 2021), 254-263. Doi: 10.1109/cvpr46437.2021.00032.Google ScholarCross Ref
- Mayu Sakurada, and Takehisa Yairi. 2014. Anomaly detection using autoencoders with nonlinear dimensionality reduction. In Proceedings of the MLSDA 2014 2nd workshop on machine learning for sensory data analysis, 4-11. Doi: 10.1145/2689746.2689747.Google ScholarDigital Library
- VanLoi Cao, Miguel Nicolau, and James McDermott. 2016. A hybrid autoencoder and density estimation model for anomaly detection. In Parallel Problem Solving from Nature–PPSN XIV: 14th International Conference, Edinburgh, UK, September 17-21, 2016, Proceedings 14, Springer International Publishing, 717-726. https://doi.org/10.1007/978-3-319-45823-6_67.Google ScholarCross Ref
- Hyunjong Park, Jongyoun Noh, and Bumsub Ham. 2020. Learning Memory-Guided Normality for Anomaly Detection. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition(CVPR 2020), 14372-14381. https://doi.org/10.48550/arXiv.2003.13228.Google ScholarCross Ref
- Jinlei Hou, Yingying Zhang, Qiaoyong Zhong, Di Xie, Shiliang Pu, Hong Zhou. 2021. Divide-and-Assemble: Learning Block-wise Memory for Unsupervised Anomaly. In Proceedings of the IEEE/CVF International Conference on Computer Vision(CVPR 2021), 8791-8800. https://doi.org/10.48550/arXiv.2107.13118.Google ScholarCross Ref
- Izhak Golan, and Ran El-Yaniv. 2018. Detection Deep anomaly detection using geometric transformations. Advances in neural information processing systems(NeurIPS 2018), 31. https://arxiv.org/abs/1805.10917.Google Scholar
- Jonathan Pirnay, and Keng Chai. 2022. Inpainting Transformer for Anomaly Detection. In Image Analysis and Processing–ICIAP 2022: 21st International Conference, 2022, Proceedings, Part II, Cham, Springer International Publishing. https://doi.org/10.1007/978-3-031-06430-2_33.Google ScholarDigital Library
- Vitjan Zavrtanik, Matej Kristan, and Danijel Skočaj. 2021. Reconstruction by inpainting for visual anomaly detection. Pattern Recognition, 112, 107706. https://doi.org/10.1016/j.patcog.2020.107706.Google ScholarCross Ref
- Tongkun Liu, Bing Li, Zhuo Zhao, Xiao Du, Bingke Jiang, Leqi Geng. 2022. Reconstruction from edge image combined with color and gradient difference for industrial surface anomaly detection. arXiv preprint, arXiv:2210.14485. https://doi.org/10.48550/arXiv.2210.14485.Google ScholarCross Ref
- Paul Bergmann, Michael Fauser, David Sattlegger, Carsten Steger. 2020. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (CVPR 2020), 4183-4192. Doi: 10.1109/cvpr42600.2020.00424.Google ScholarCross Ref
- Yunfei Liu, Chaoqun Zhuang, Feng Lu. 2021. Unsupervised Two-Stage Anomaly Detection. arXiv preprint, arXiv: 2103.11671. https://doi.org/10.48550/arXiv.2103.11671.Google ScholarCross Ref
Index Terms
- A two-stage network based on edge information for visual anomaly detection
Recommendations
Video anomaly detection and localization based on appearance and motion models
AbstractIn this paper, we present an approach to detect and localize anomalies in the surveillance videos. Precise detection, modeling the normality in a context and dealing with false alarms are the major challenges to cope with while performing ...
Toward Explainable Deep Anomaly Detection
KDD '21: Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data MiningAnomaly explanation, also known as anomaly localization, is as important as, if not more than, anomaly detection in many real-world applications. However, it is challenging to build explainable detection models due to the lack of anomaly-supervisory ...
Two-stage anomaly detection algorithm via dynamic community evolution in temporal graph
AbstractDetecting anomalies from a massive amount of user behavioral data is often liken to finding a needle in a haystack. While tremendous efforts have been devoted to anomaly detection from temporal graphs, existing studies rarely consider community ...
Comments