图像异常检测与定位方法的鲁棒性研究

随着计算机算力和视觉算法的快速发展，基于计算机视觉的检测和定位算法能广泛应用在工业质检、医疗诊断和遥感监控等多个行业中。无监督异常检测算法是近年其中一个重要的研究内容，它只需要以正常数据作为训练集而不需要任何的标注，就能检测出偏离正常分布的异常数据，大幅降低了算法的应用门槛。

本文以图像异常检测与定位任务为主要研究方向，回顾了现有研究，并针对工业质检任务通过实验指出了现有算法的两个鲁棒性问题。第一个是噪声鲁棒性问题，即训练集中的异常样本会对检测模型造成较大影响。第二个是多类别鲁棒性问题，即在多类别统一异常检测训练时，不同类别的数据互相影响成为噪声。为了解决这两个问题，本文提出了基于核心集去噪的异常检测算法和基于多类分布隐式神经表示的统一异常检测算法。本文的主要研究内容和成果如下：

（1）提出一个基于核心集去噪的无监督异常检测算法。由于现有的无监督算法要求一个只有正常数据的训练集，这对实际应用来说依然需要人工分类出正常数据，这种隐式的监督信号耗费成本且可能出现错误分类。本文提出一个噪声鲁棒的基于内存的无监督异常检测结构SoftPatch，利用离群检测的思想对核心集去噪，并通过添加软权重的方式进一步提升鲁棒性。在带噪声的设置中，SoftPatch显著优于现有异常检测和去噪算法。

（2）提出一个多类别异常检测方法。一般的异常检测方法只针对一个种类图像进行建模，而对多类别图像进行学习时，会由于类别分布差异而相互影响。本文提出一个多类鲁棒的基于重构的异常检测方法MINT-AD，利用类别信息对不同类别和不同分布进行隔离，利用隐式神经表示和先验分布损失将类别信息映射到特征维度，从而与重构过程结合。在MVTec、VisA和BTAD等多个数据集上的实验表明，多类别异常检测要求模型对复杂分布具有鲁棒性，MINT-AD在检测和定位性能上都优于现有算法。

With the rapid development of computer power and visual algorithms, computer vision-based detection and localization algorithms have been applied in various industries, such as industrial quality inspection, medical diagnosis, and remote sensing monitoring. In recent years, unsupervised visual anomaly detection algorithms have been an important research area. They can detect anomalous data that deviates from normal distribution without annotation, significantly reducing the threshold for algorithm applications.

We review existing anomaly detection and localization research and point out two robustness problems of existing algorithms through experiments in the context of industrial quality inspection. One is the noise robustness problem, the problem that the appearance of anomaly samples in the training set will greatly affect the performance of the detection model. The other is the multi-class robustness problem, which refers to the problem of mutual interference among multi-class data during unified anomaly detection training. So we propose a coreset-based denoising anomaly detection algorithm and a multi-class distribution implicit-neural-representation-based unified anomaly detection algorithm. Our main research content and achievements are as follows:

(1) We propose a noise-robust unsupervised anomaly detection algorithm based on the coreset denoising. As existing unsupervised algorithms require a training set with only normal data, this still requires manual classification, which is implicit supervision and can be costly and may result in misclassification. SoftPatch, a noise-robust memory-based unsupervised anomaly detection structure, is proposed to denoise the coreset using outlier detection and improve robustness by adding soft weights. SoftPatch is significantly better than anomaly detection and denoising algorithms in noisy settings. Meanwhile, we also discovered that some datasets inherently contain noise, proving that improving noise robustness is important.

(2) We propose a multi-class anomaly detection method. The general anomaly detection methods only model a single type of image, and learning on multiple types of images may interfere with each other due to differences in distribution. MINT-AD, a multi-class robust reconstruction-based anomaly detection method, is proposed to isolate different distributions and classes. We use category information and map it to the feature dimension using implicit neural representation and prior distribution loss. Experiments on multiple datasets, such as MVTec, VisA, and BTAD, show that multi-class anomaly detection requires the model to have the robustness to complex distributions, and MINT-AD outperforms existing algorithms in both detection and localization performance.

[1] LIU J, XIE G, WANG J, et al. Deep Visual Anomaly Detection in Industrial Manufacturing: A Survey[A]. 2023.
[2] LI W, ZHAN J, WANG J, et al. Towards continual adaptation in industrial anomaly detection [C]//Proceedings of the 30th ACM International Conference on Multimedia. 2022: 2871-2880.
[3] ZHANG S, ZHANG L, XIE G, et al. What makes a good data augmentation for few-shot unsupervised image anomaly detection?[A]. 2023.
[4] XIE G, WANG J, LIU J, et al. Pushing the Limits of Fewshot Anomaly Detection in Industry Vision: Graphcore[A]. 2023.
[5] CHALAPATHY R, CHAWLA S. Deep learning for anomaly detection: A survey[A]. 2019.
[6] BULUSU S, KAILKHURA B, LI B, et al. Anomalous instance detection in deep learning: A survey[R]. Lawrence Livermore National Lab.(LLNL), Livermore, CA (United States), 2020.
[7] PANG G, SHEN C, CAO L, et al. Deep learning for anomaly detection: A review[J]. ACM Computing Surveys (CSUR), 2021, 54(2): 1-38.
[8] RUFF L, KAUFFMANN J R, VANDERMEULEN R A, et al. A unifying review of deep and shallow anomaly detection[J]. Proceedings of the IEEE, 2021.
[9] SALEHI M, MIRZAEI H, HENDRYCKS D, et al. A Unified Survey on Anomaly, Novelty, Open-Set, and Out-of-Distribution Detection: Solutions and Future Challenges[A]. 2021.
[10] CHO H, SEOL J, LEE S G. Masked Contrastive Learning for Anomaly Detection[A]. 2021.
[11] MAZIARKA L, SMIEJA M, SENDERA M, et al. OneFlow: One-class flow for anomaly detec- tion based on a minimal volume region[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.
[12] YANG J, ZHOU K, LI Y, et al. Generalized out-of-distribution detection: A survey[A]. 2021.
[13] KONG S, SHEN Y, HUANG L. Resolving Training Biases via Influence-based Data Relabeling [C]//International Conference on Learning Representations. 2021.
[14] YOU Z, CUI L, SHEN Y, et al. A Unified Model for Multi-class Anomaly Detection[A]. 2022.
[15] ZHOU Z H. A brief introduction to weakly supervised learning[J]. National science review, 2018, 5(1): 44-53.
[16] YANG J, XU R, QI Z, et al. Visual Anomaly Detection for Images: A Survey[A]. 2021.
[17] LI C L, SOHN K, YOON J, et al. CutPaste: Self-Supervised Learning for Anomaly Detec- tion and Localization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 9664-9674.
[18] SHEYNIN S, BENAIM S, WOLF L. A Hierarchical Transformation-Discriminating Genera- tive Model for Few Shot Anomaly Detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV). 2021: 8495-8504.
[19] ZAVRTANIK V, KRISTAN M, SKOČAJ D. DRAEM-A discriminatively trained reconstruc- tion embedding for surface anomaly detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 8330-8339.
[20] REISS T, COHEN N, BERGMAN L, et al. PANDA: Adapting Pretrained Features for Anomaly Detection and Segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 2806-2814.
[21] SALEHI M, SADJADI N, BASELIZADEH S, et al. Multiresolution knowledge distillation for anomaly detection[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 14902-14912.
[22] BERGMANN P, FAUSER M, SATTLEGGER D, et al. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 4183-4192.
[23] YAN X, ZHANG H, XU X, et al. Learning Semantic Context from Normal Samples for Un- supervised Anomaly Detection[C]//Proceedings of the AAAI Conference on Artificial Intelli- gence: volume 35. 2021: 3110-3118.
[24] ZHOU K, XIAO Y, YANG J, et al. Encoding structure-texture relation with P-Net for anomaly detection in retinal images[C]//Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XX 16. Springer, 2020: 360-377.
[25] DEHAENE D, FRIGO O, COMBREXELLE S, et al. Iterative energy-based projection on a normal data manifold for anomaly localization[C]//International Conference on Learning Rep- resentations. 2019.
[26] HOU J, ZHANG Y, ZHONG Q, et al. Divide-and-assemble: Learning block-wise memory for unsupervised anomaly detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 8791-8800.
[27] RUDOLPH M, WANDT B, ROSENHAHN B. Same same but differnet: Semi-supervised de- fect detection with normalizing flows[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2021: 1907-1916.
[28] CHU W H, KITANI K M. Neural Batch Sampling with Reinforcement Learning for Semi- Supervised Anomaly Detection[C]//Computer Vision–ECCV 2020: 16th European Conference, Glasgow, UK, August 23–28, 2020, Proceedings, Part XXVI 16. Springer, 2020: 751-766.
[29] VENKATARAMANAN S, PENG K C, SINGH R V, et al. Attention guided anomaly localiza- tion in images[C]//European Conference on Computer Vision. Springer, 2020: 485-503.
[30] LIU J, WANG C, SU H, et al. Multistage GAN for fabric defect detection[J]. IEEE Transactions on Image Processing, 2019, 29: 3388-3400.
[31] SU J, SHEN H, PENG L, et al. Few-shot domain-adaptive anomaly detection for cross-site brain images[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021.
[32] SINDAGI V A, SRIVASTAVA S. Domain adaptation for automatic OLED panel defect detection using adaptive support vector data description[J]. International Journal of Computer Vision, 2017, 122(2): 193-211.
[33] SIDDIQUEE M M R, ZHOU Z, TAJBAKHSH N, et al. Learning fixed points in generative adversarial networks: From image-to-image translation to disease detection and localization [C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019: 191- 200.
[34] BHATTACHARYA G, MANDAL B, PUHAN N B. Interleaved Deep Artifacts-Aware Atten- tion Mechanism for Concrete Structural Defect Classification[J]. IEEE Transactions on Image Processing, 2021, 30: 6957-6969.
[35] YANG L, LI B, YANG G, et al. Deep neural network based visual inspection with 3d metric measurement of concrete defects using wall-climbing robot[C]//2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS). IEEE, 2019: 2849-2854.
[36] ZENG Z, LIU B, FU J, et al. Reference-Based Defect Detection Network[J]. IEEE Transactions on Image Processing, 2021, 30: 6637-6647.
[37] LONG X, FANG B, ZHANG Y, et al. Fabric defect detection using tactile information[C]// 2021 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2021: 11169- 11174.
[38] LIS K, NAKKA K, FUA P, et al. Detecting the unexpected via image resynthesis[C]// Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019: 2152-2161.
[39] OBERDIEK P, ROTTMANN M, FINK G A. Detection and retrieval of out-of-distribution objects in semantic segmentation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops. 2020: 328-329.
[40] DI BIASE G, BLUM H, SIEGWART R, et al. Pixel-wise anomaly detection in complex driving scenes[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recogni- tion. 2021: 16918-16927.
[41] VOJIR T, ŠIPKA T, ALJUNDI R, et al. Road anomaly detection by partial image reconstruc- tion with segmentation coupling[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 15651-15660.
[42] KENDALL A, GAL Y. What uncertainties do we need in bayesian deep learning for computer vision?[J]. Advances in neural information processing systems, 2017, 30.
[43] RAMOS S, GEHRIG S, PINGGERA P, et al. Detecting unexpected obstacles for self-driving cars: Fusing deep learning and geometric modeling[C]//2017 IEEE Intelligent Vehicles Sym- posium (IV). IEEE, 2017: 1025-1032.
[44] GUPTA K, JAVED S A, GANDHI V, et al. Mergenet: A deep net architecture for small obstacle discovery[C]//2018 IEEE International Conference on Robotics and Automation (ICRA). IEEE, 2018: 5856-5862.
[45] SUN L, YANG K, HU X, et al. Real-time fusion network for RGB-D semantic segmentation incorporating unexpected obstacle detection for road-driving images[J]. IEEE Robotics and Automation Letters, 2020, 5(4): 5558-5565.
[46] XUE F, MING A, ZHOU M, et al. A novel multi-layer framework for tiny obstacle discovery [C]//2019 International Conference on Robotics and Automation (ICRA). IEEE, 2019: 2939- 2945.
[47] CHAN R, LIS K, UHLEMEYER S, et al. SegmentMeIfYouCan: A Benchmark for Anomaly Segmentation[A]. 2021.
[48] BARZ B, RODNER E, GARCIA Y G, et al. Detecting Regions of Maximal Divergence for Spatio-Temporal Anomaly Detection[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2019, 41: 1088-1101.
[49] CHANG Y, TU Z, XIE W, et al. Clustering Driven Deep Autoencoder for Video Anomaly Detection[C]//ECCV. 2020.
[50] GEORGESCU M I, IONESCU R T, KHAN F S, et al. A Background-Agnostic Framework with Adversarial Training for Abnormal Event Detection in Video.[J]. IEEE transactions on pattern analysis and machine intelligence, 2021, PP.
[51] GONG D, LIU L, LE V, et al. Memorizing Normality to Detect Anomaly: Memory-Augmented Deep Autoencoder for Unsupervised Anomaly Detection[J]. 2019 IEEE/CVF International Con- ference on Computer Vision (ICCV), 2019: 1705-1714.
[52] IONESCU R T, KHAN F S, GEORGESCU M I, et al. Object-Centric Auto-Encoders and Dummy Anomalies for Abnormal Event Detection in Video[J]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2019: 7834-7843.
[53] JARDIM E, THOMAZ L A, DA SILVA E A B, et al. Domain-Transformable Sparse Repre- sentation for Anomaly Detection in Moving-Camera Videos[J]. IEEE Transactions on Image Processing, 2020, 29: 1329-1343.
[54] LI X, CHEN M, WANG Q. Quantifying and Detecting Collective Motion in Crowd Scenes[J]. IEEE Transactions on Image Processing, 2020, 29: 5571-5583.
[55] LUO W, LIU W, LIAN D, et al. Future Frame Prediction Network for Video Anomaly Detection. [J]. IEEE transactions on pattern analysis and machine intelligence, 2021, PP.
[56] MARKOVITZ A, SHARIR G, FRIEDMAN I, et al. Graph Embedded Pose Clustering for Anomaly Detection[J]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2020: 10536-10544.
[57] MORAIS R, LE V, TRAN T, et al. Learning Regularity in Skeleton Trajectories for Anomaly Detection in Videos[J]. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recog- nition (CVPR), 2019: 11988-11996.
[58] PANG G, YAN C, SHEN C, et al. Self-Trained Deep Ordinal Regression for End-to-End Video Anomaly Detection[J]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recogni- tion (CVPR), 2020: 12170-12179.
[59] PARK H, NOH J, HAM B. Learning Memory-Guided Normality for Anomaly Detection [J]. 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2020: 14360-14369.
[60] SABOKROU M, FAYYAZ M, FATHY M, et al. Deep-Cascade: Cascading 3D Deep Neural Networks for Fast Anomaly Detection and Localization in Crowded Scenes[J]. IEEE Transac- tions on Image Processing, 2017, 26: 1992-2004.
[61] BAPPY J H, PAUL S, TUNCEL E, et al. Exploiting Typicality for Selecting Informative and Anomalous Samples in Videos[J]. IEEE Transactions on Image Processing, 2019, 28: 5214- 5226.
[62] YAO Y, XU M, WANG Y, et al. Unsupervised Traffic Accident Detection in First-Person Videos [J]. 2019 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 2019: 273-280.
[63] LIU Z, NIE Y, LONG C, et al. A Hybrid Video Anomaly Detection Framework via Memory- Augmented Flow Reconstruction and Flow-Guided Frame Prediction: abs/2108.06852[A]. 2021.
[64] LUO W, LIU W, LIAN D, et al. Video Anomaly Detection with Sparse Coding Inspired Deep Neural Networks[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2021, 43: 1070-1084.
[65] LIU W, LUO W, LI Z, et al. Margin Learning Embedded Prediction for Video Anomaly De- tection with A Few Anomalies[C]//IJCAI. 2019.
[66] LV H, ZHOU C, CUI Z, et al. Localizing Anomalies From Weakly-Labeled Videos[J]. IEEE Transactions on Image Processing, 2021, 30: 4505-4515.
[67] PURWANTO D, CHEN Y T, FANG W H. Dance With Self-Attention: A New Look of Con- ditional Random Fields on Anomaly Detection in Videos[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 173-183.
[68] SULTANI W, CHEN C, SHAH M. Real-World Anomaly Detection in Surveillance Videos[J]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 6479-6488.
[69] TIAN Y, PANG G, CHEN Y, et al. Weakly-supervised video anomaly detection with robust temporal feature magnitude learning[C]//Proceedings of the IEEE/CVF international conference on computer vision. 2021: 4975-4986.
[70] WU J, ZHANG W, LI G, et al. Weakly-Supervised Spatio-Temporal Anomaly Detection in Surveillance Video[C]//IJCAI. 2021.
[71] WU P, LIU J. Learning Causal Temporal Relation and Feature Discrimination for Anomaly Detection[J]. IEEE Transactions on Image Processing, 2021, 30: 3513-3527.
[72] ZAHEER M, MAHMOOD A, ASTRID M, et al. CLAWS: Clustering Assisted Weakly Super- vised Learning with Normalcy Suppression for Anomalous Event Detection[C]//ECCV. 2020.
[73] ZHONG J X, LI N, KONG W, et al. Graph Convolutional Label Noise Cleaner: Train a Plug- And-Play Action Classifier for Anomaly Detection[J]. 2019 IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR), 2019: 1237-1246.
[74] LEYVA R, SANCHEZ V, LI C T. Video Anomaly Detection With Compact Feature Sets for Online Performance[J]. IEEE Transactions on Image Processing, 2017, 26: 3463-3478.
[75] LI J, HUANG Q, DU Y, et al. Variational Abnormal Behavior Detection With Motion Consis- tency[J]. IEEE Transactions on Image Processing, 2022, 31: 275-286.
[76] LIU W, LUO W, LIAN D, et al. Future Frame Prediction for Anomaly Detection - A New Baseline[J]. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2018: 6536-6545.
[77] LUO W, LIU W, GAO S. A Revisit of Sparse Coding Based Anomaly Detection in Stacked RNN Framework[J]. 2017 IEEE International Conference on Computer Vision (ICCV), 2017: 341-349.
[78] LU C, SHI J, WANG W, et al. Fast Abnormal Event Detection[J]. International Journal of Computer Vision, 2018: 1-19.
[79] LIZNERSKI P, RUFF L, VANDERMEULEN R A, et al. Explainable deep one-class classifi- cation[A]. 2020.
[80] BERGMANN P, FAUSER M, SATTLEGGER D, et al. MVTec AD–A comprehensive real- world dataset for unsupervised anomaly detection[C]//Proceedings of the IEEE/CVF Confer- ence on Computer Vision and Pattern Recognition. 2019: 9592-9600.
[81] BERGMANN P, FAUSER M, SATTLEGGER D, et al. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 4183-4192.
[82] YAMADA S, HOTTA K. Reconstruction Student with Attention for Student-Teacher Pyramid Matching[A]. 2021.
[83] SALEHI M, SADJADI N, BASELIZADEH S, et al. Multiresolution knowledge distillation for anomaly detection[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2021: 14902-14912.
[84] WANG G, HAN S, DING E, et al. Student-Teacher Feature Pyramid Matching for Anomaly Detection[C]//BMVC. 2021.
[85] DENG H, LI X. Anomaly Detection via Reverse Distillation from One-Class Embedding[A]. 2022.
[86] RUDOLPH M, WANDT B, ROSENHAHN B. Same same but differnet: Semi-supervised de- fect detection with normalizing flows[C]//Proceedings of the IEEE/CVF winter conference on applications of computer vision. 2021: 1907-1916.
[87] YU J, ZHENG Y, WANG X, et al. Fastflow: Unsupervised anomaly detection and localization via 2d normalizing flows[A]. 2021.
[88] RUDOLPH M, WEHRBEIN T, ROSENHAHN B, et al. Fully convolutional cross-scale-flows for image-based defect detection[C]//Proceedings of the IEEE/CVF Winter Conference on Ap- plications of Computer Vision. 2022: 1088-1097.
[89] GUDOVSKIY D, ISHIZAKA S, KOZUKA K. Cflow-ad: Real-time unsupervised anomaly detection with localization via conditional normalizing flows[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2022: 98-107.
[90] COHEN N, HOSHEN Y. Sub-image anomaly detection with deep pyramid correspondences [A]. 2020.
[91] LI N, JIANG K, MA Z, et al. Anomaly Detection Via Self-Organizing Map[C]//2021 IEEE International Conference on Image Processing (ICIP). IEEE, 2021: 974-978.
[92] KIM J H, KIM D H, YI S, et al. Semi-orthogonal embedding for efficient unsupervised anomaly segmentation[A]. 2021.
[93] DEFARD T, SETKOV A, LOESCH A, et al. Padim: a patch distribution modeling framework for anomaly detection and localization[C]//International Conference on Pattern Recognition. Springer, 2021: 475-489.
[94] ROTH K, PEMULA L, ZEPEDA J, et al. Towards total recall in industrial anomaly detection [A]. 2021.
[95] LEE S, LEE S, SONG B C. CFA: Coupled-hypersphere-based Feature Adaptation for Target- Oriented Anomaly Localization[A]. 2022.
[96] SOHN K, LI C L, YOON J, et al. Learning and Evaluating Representations for Deep One-Class Classification[C]//International Conference on Learning Representations. 2020.
[97] MASSOLI F V, FALCHI F, KANTARCI A, et al. MOCCA: Multilayer One-Class Classification for Anomaly Detection[J]. IEEE Transactions on Neural Networks and Learning Systems, 2021.
[98] YI J, YOON S. Patch svdd: Patch-level svdd for anomaly detection and segmentation[C]// Proceedings of the Asian Conference on Computer Vision. 2020.
[99] HU C, CHEN K, SHAO H. A Semantic-Enhanced Method Based On Deep SVDD for Pixel- Wise Anomaly Detection[C]//2021 IEEE International Conference on Multimedia and Expo (ICME). IEEE, 2021: 1-6.
[100] SAUTER D, SCHMITZ A, DIKICI F, et al. Defect Detection of Metal Nuts Applying Convo- lutional Neural Networks[C]//2021 IEEE 45th Annual Computers, Software, and Applications Conference (COMPSAC). IEEE, 2021: 248-257.
[101] REISST,COHENN,BERGMANL,etal.PANDA:AdaptingPretrainedFeaturesforAnomaly Detection and Segmentation[J]. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 2021: 2805-2813.
[102] LI C L, SOHN K, YOON J, et al. Cutpaste: Self-supervised learning for anomaly detection and localization[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2021: 9664-9674.
[103] YANGJ,SHIY,QIZ.Dfr:Deepfeaturereconstructionforunsupervisedanomalysegmentation [A]. 2020.
[104] ZAVRTANIK V, KRISTAN M, SKOČAJ D. Draem-a discriminatively trained reconstruction embedding for surface anomaly detection[C]//Proceedings of the IEEE/CVF International Con- ference on Computer Vision. 2021: 8330-8339.
[105] LIANG Y, ZHANG J, ZHAO S, et al. Omni-frequency Channel-selection Representations for Unsupervised Anomaly Detection[A]. 2022.
[106] BERGMANNP,FAUSERM,SATTLEGGERD,etal.MVTecAD—AComprehensiveReal- World Dataset for Unsupervised Anomaly Detection[J]. 2019 IEEE/CVF Conference on Com- puter Vision and Pattern Recognition (CVPR), 2019: 9584-9592.
[107] BERGMANN P, BATZNER K, FAUSER M, et al. The MVTec anomaly detection dataset: a comprehensive real-world dataset for unsupervised anomaly detection[J]. International Journal of Computer Vision, 2021, 129(4): 1038-1059.
[108] BERGMANN P, JIN X, SATTLEGGER D, et al. The mvtec 3d-ad dataset for unsupervised 3d anomaly detection and localization[A]. 2021.
[109] BERGMANN P, BATZNER K, FAUSER M, et al. Beyond Dents and Scratches: Logical Con- straints in Unsupervised Anomaly Detection and Localization[J]. International Journal of Com- puter Vision, 2022, 130(4): 947-969.
[110] JEZEK S, JONAK M, BURGET R, et al. Deep learning-based defect detection of metal parts:evaluating current methods in complex conditions[C]//2021 13th International Congress on Ultra Modern Telecommunications and Control Systems and Workshops (ICUMT). IEEE, 2021:66-71.
[111] MISHRA P, VERK R, FORNASIER D, et al. VT-ADL: A vision transformer network for imageanomaly detection and localization[C]//2021 IEEE 30th International Symposium on IndustrialElectronics (ISIE). IEEE, 2021: 01-06.
[112] ZOU Y, JEONG J, PEMULA L, et al. SPot-the-Diference Self-supervised Pre-training forAnomaly Detection and Segmentation[C]//Computer Vision–ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23–27, 2022, Proceedings, Part XXX. Springer, 2022: 392-408.
[113] HUANG Y, QIU C, YUAN K. Surface defect saliency of magnetic tile[J]. The Visual Computer,2020, 36(1): 85-96.
[114] DAGM (DEUTSCHE ARBEITSGEMEINSCHAFT FüR MUSTERERKENNUNG E.V.G C O T I I A F P R, THE GNSS (GERMAN CHAPTER OF THE EUROPEAN NEURALNETWORK SOCIETY). DAGM dataset[EB/OL]. 2000. http://www.thisisurl/.
[115] BAO T, CHEN J, LI W, et al. MIAD: A Maintenance Inspection Dataset for UnsupervisedAnomaly Detection[A]. 2022.
[116] NDIOUR I, AHUJA N, GENC U, et al. FRE: A Fast Method For Anomaly Detection AndSegmentation[A]. 2022.
[117] HU Z, YANG Z, HU X, et al. SimPLE: Similar Pseudo Label Exploitation for Semi-SupervisedClassifcation[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and PatternRecognition (CVPR). 2021: 15099-15108.
[118] SOHN K, BERTHELOT D, CARLINI N, et al. Fixmatch: Simplifying semi-supervised learningwith consistency and confdence[J]. Advances in Neural Information Processing Systems, 2020,33: 596-608.
[119] LI J, SOCHER R, HOI S C. Dividemix: Learning with noisy labels as semi-supervised learning[A]. 2020.
[120] XU M, ZHANG Z, HU H, et al. End-to-end semi-supervised object detection with soft teacher[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2021: 3060-3069.
[121] LIU Y C, MA C Y, HE Z, et al. Unbiased teacher for semi-supervised object detection[A]. 2021.
[122] YANG F, WU K, ZHANG S, et al. Class-Aware Contrastive Semi-Supervised Learning[A].2022.
[123] PETERSON L E. K-nearest neighbor[J]. Scholarpedia, 2009, 4(2): 1883.
[124] BREUNIG M M, KRIEGEL H P, NG R T, et al. LOF: identifying density-based local outliers[C]//Proceedings of the 2000 ACM SIGMOD international conference on Management of data.2000: 93-104.
[125] BERGMANN P, LÖWE S, FAUSER M, et al. Improving unsupervised defect segmentation byapplying structural similarity to autoencoders[A]. 2018.
[126] COLLIN A S, DE VLEESCHOUWER C. Improved anomaly detection by training an autoencoder with skip connections on images corrupted with stain-shaped noise[C]//2020 25th International Conference on Pattern Recognition (ICPR). IEEE, 2021: 7915-7922.
[127] KINGMA D P, WELLING M. Auto-encoding variational bayes[A]. 2013.
[128] LIU W, LI R, ZHENG M, et al. Towards visually explaining variational autoencoders[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020:8642-8651.
[129] AKCAY S, ATAPOUR-ABARGHOUEI A, BRECKON T P. Ganomaly: Semi-supervisedanomaly detection via adversarial training[C]//Computer Vision–ACCV 2018: 14th Asian Conference on Computer Vision, Perth, Australia, December 2–6, 2018, Revised Selected Papers,Part III 14. Springer, 2019: 622-637.
[130] PERERA P, NALLAPATI R, XIANG B. Ocgan: One-class novelty detection using gans withconstrained latent representations[C]//Proceedings of the IEEE/CVF Conference on ComputerVision and Pattern Recognition. 2019: 2898-2906.
[131] SABOKROU M, KHALOOEI M, FATHY M, et al. Adversarially learned one-class classiferfor novelty detection[C]//Proceedings of the IEEE conference on computer vision and patternrecognition. 2018: 3379-3388.
[132] ZHOU K, XIAO Y, YANG J, et al. Encoding structure-texture relation with p-net for anomalydetection in retinal images[C]//Computer Vision–ECCV 2020: 16th European Conference,Glasgow, UK, August 23–28, 2020, Proceedings, Part XX 16. Springer, 2020: 360-377.
[133] SHI Y, YANG J, QI Z. Unsupervised anomaly segmentation via deep feature reconstruction[J].Neurocomputing, 2021, 424: 9-22.
[134] XIA Y, ZHANG Y, LIU F, et al. Synthesize then compare: Detecting failures and anomalies for semantic segmentation[C]//Computer Vision–ECCV 2020: 16th European Conference,Glasgow, UK, August 23–28, 2020, Proceedings, Part I 16. Springer, 2020: 145-161.
[135] GONG D, LIU L, LE V, et al. Memorizing normality to detect anomaly: Memory-augmenteddeep autoencoder for unsupervised anomaly detection[C]//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2019: 1705-1714.
[136] HOU J, ZHANG Y, ZHONG Q, et al. Divide-and-assemble: Learning block-wise memory forunsupervised anomaly detection[C]//Proceedings of the IEEE/CVF International Conference onComputer Vision. 2021: 8791-8800.
[137] PARK H, NOH J, HAM B. Learning memory-guided normality for anomaly detection[C]//Proceedings of the IEEE/CVF conference on computer vision and pattern recognition. 2020:14372-14381.
[138] DEHAENE D, FRIGO O, COMBREXELLE S, et al. Iterative energy-based projection on anormal data manifold for anomaly localization[A]. 2020.
[139] YAN X, ZHANG H, XU X, et al. Learning semantic context from normal samples for unsupervised anomaly detection[C]//Proceedings of the AAAI Conference on Artifcial Intelligence:volume 35. 2021: 3110-3118.
[140] SCHWARTZ E, ARBELLE A, KARLINSKY L, et al. MAEDAY: MAE for few and zero shotAnomalY-Detection[A]. 2022.
[141] POURREZA M, MOHAMMADI B, KHAKI M, et al. G2d: Generate to detect anomaly[C]//Proceedings of the IEEE/CVF Winter Conference on Applications of Computer Vision. 2021:2003-2012.
[142] PARK J J, FLORENCE P, STRAUB J, et al. Deepsdf: Learning continuous signed distancefunctions for shape representation[C]//Proceedings of the IEEE/CVF conference on computervision and pattern recognition. 2019: 165-174.
[143] LI M, PATIL A G, XU K, et al. Grains: Generative recursive autoencoders for indoor scenes[J]. ACM Transactions on Graphics (TOG), 2019, 38(2): 1-16.
[144] SITZMANN V, MARTEL J N P, BERGMAN A W, et al. Implicit Neural Representations withPeriodic Activation Functions[J]. Neural Information Processing Systems, 2020.
[145] LINDELL D B, VAN VEEN D, PARK J J, et al. Bacon: Band-limited coordinate networksfor multiscale scene representation[C]//Proceedings of the IEEE/CVF Conference on ComputerVision and Pattern Recognition. 2022: 16252-16262.
[146] DENG J, DONG W, SOCHER R, et al. Imagenet: A large-scale hierarchical image database[C]//2009 IEEE conference on computer vision and pattern recognition. Ieee, 2009: 248-255.
[147] TAN M, LE Q. Effcientnet: Rethinking model scaling for convolutional neural networks[C]//International conference on machine learning. PMLR, 2019: 6105-6114.
[148] VASWANI A, SHAZEER N, PARMAR N, et al. Attention is all you need[J]. Advances inneural information processing systems, 2017, 30.
[149] MILDENHALL B, SRINIVASAN P P, TANCIK M, et al. Nerf: Representing scenes as neuralradiance felds for view synthesis[J]. Communications of the ACM, 2021, 65(1): 99-106.
[150] MEHTA I, GHARBI M, BARNES C, et al. Modulated periodic activations for generalizablelocal functional representations[C]//Proceedings of the IEEE/CVF International Conference onComputer Vision. 2021: 14214-14223.