基於多尺度區塊的表徵學習應用於影像異常偵測與分割

簡易檢索 / 詳目顯示

回結果列表

研究生：	蔡勁家 Tsai, Chin-Chia
論文名稱：	基於多尺度區塊的表徵學習應用於影像異常偵測與分割 Multi-Scale Patch-Based Representation Learning for Image Anomaly Detection and Segmentation
指導教授：	賴尚宏 Lai, Shang-Hong
口試委員:	邱瀞德 Chiu, Ching-Te 林嘉文 Lin, Chia-Wen 劉庭祿 Liu, Tyng-Luh
學位類別：	碩士 Master
系所名稱：	電機資訊學院 - 資訊系統與應用研究所 Institute of Information Systems and Applications
論文出版年：	2021
畢業學年度：	109
語文別：	英文
論文頁數：	35
中文關鍵詞：	異常偵測、深度學習、多尺度區塊、表徵學習、非監督式學習、自監督學習
外文關鍵詞：	Anomaly detection, Deep learning, Multi-scale patch-based, Representation learning, Unsupervised learning, Self-supervised learning
相關次數：	點閱：2 下載：0
分享至:	分享至facebook 分享至twitter

查詢本校圖書館目錄查詢臺灣博碩士論文知識加值系統勘誤回報

近年來非監督式表徵學習已被證實對於解決挑戰難度較高的異常偵測與分割任務十分有效。本篇論文中，我們提出了一種基於多尺度區塊的表徵學習方法，萃取正常影像中關鍵且具代表性的訊息。通過考慮區域性距離區塊之間的特徵相似度，我們實現了更好的表徵學習。此外，我們改善了預測區塊之間空間組態的自監督學習策略，將其擴展為更加精細的方式，使我們的模型能夠學習更好的區塊之間的幾何關係。通過在影像上滑動不同尺度的區塊，我們的模型從每個區塊中萃取出具代表性的特徵，並將該特徵與正常影像訓練資料集中的特徵做比對，以此來偵測異常區域。我們在公開資料集MVTec AD與BTAD的實驗結果顯示，我們提出的方法在異常偵測與分割的任務中，皆能達到目前為止最好的準確度。

Unsupervised representation learning has been proven to be effective for the challenging anomaly detection/segmentation tasks. In this paper, we propose a multi-scale patch-based representation learning method to extract critical and representative information from normal images. By taking the feature similarity between patches of local distance into account, we can achieve better representation learning. Moreover, we improve the self-supervised learning strategy, i.e., predicting the spatial configuration between the patches, to a more refined manner, thus allowing our model to learn better geometric relationship between the patches. Through sliding patches of different scales all over an image, our model extracts representative features from each patch and compares it with those in the training set of normal images to detect the anomalous regions. Our experimental results on MVTec AD dataset and BTAD dataset demonstrate the proposed method achieves the state-of-the-art accuracy for both anomaly detection and segmentation.

Introduction 1
1 Motivation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
2 Problem Statement . . . . . . . . . . . . . . . . . . . . . . . . . . 2
3 Contributions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
4 Thesis Organization . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Related Works 4
1 Reconstruction-based methods . . . . . . . . . . . . . . . . . . . . 4
2 Feature-based methods . . . . . . . . . . . . . . . . . . . . . . . . 5
Method 7
1 Overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
2 Network Architecture . . . . . . . . . . . . . . . . . . . . . . . . . 8
3 Objective Function . . . . . . . . . . . . . . . . . . . . . . . . . . 9
3.1 SVDD Loss and Cos Loss . . . . . . . . . . . . . . . . . . 9
3.2 Overall Loss . . . . . . . . . . . . . . . . . . . . . . . . . 11
4 Patch Selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
5 Inference and Score Calculation . . . . . . . . . . . . . . . . . . . 13
Experiments 15
1 Experimental Setting . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.1 Datasets . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
1.2 Evaluation Metric . . . . . . . . . . . . . . . . . . . . . . . 16
1.3 Implementation Details . . . . . . . . . . . . . . . . . . . . 16
2 Experimental Comparison . . . . . . . . . . . . . . . . . . . . . . 17
3 Ablation Study . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21
4 Visualization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
4.1 tSNE Visualization . . . . . . . . . . . . . . . . . . . . . 24
4.2 Qualitative Results . . . . . . . . . . . . . . . . . . . . . . 25
5 Detailed Analysis and Discussion . . . . . . . . . . . . . . . . . . 26
5.1 Time Complexity . . . . . . . . . . . . . . . . . . . . . . . 26
5.2 Analysis in Failure Cases . . . . . . . . . . . . . . . . . . . 27
Conclusions 32
References 33
                                

[1] Akcay, S., Atapour-Abarghouei, A., and Breckon, T. P. Ganomaly: Semi-supervised anomaly detection via adversarial training. In Asian Conference on Computer Vision (2018), Springer, pp. 622–637.
[2] Bergman, L., and Hoshen, Y. Classification-based anomaly detection for general data. In International Conference on Learning Representations (ICLR) (2020).
[3] Bergmann, P., Fauser, M., Sattlegger, D., and Steger, C. Mvtec ad —a comprehensive real-world dataset for unsupervised anomaly detection. 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2019), 9584–9592.
[4] Bergmann, P., Fauser, M., Sattlegger, D., and Steger, C. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. In 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2020), pp. 4182–4191.
[5] Bergmann, P., Löwe, S., Fauser, M., Sattlegger, D., and Steger, C. Improving unsupervised defect segmentation by applying structural similarity to autoencoders. Proceedings of the 14th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications (2019).
[6] Cohen, N., and Hoshen, Y. Subimage anomaly detection with deep pyramid correspondences. arXiv:2005.02357 (2020).
[7] Defard, T., Setkov, A., Loesch, A., and Audigier, R. Padim: a patch distribution modeling framework for anomaly detection and localization. ICPR (2020).
[8] Dehaene, D., and Eline, P. Anomaly localization by modeling perceptual features. arXiv:2008.05369 (2020).
[9] Deng, J., Dong, W., Socher, R., Li, L.J., Li, K., and FeiFei, L. Imagenet: A large-scale hierarchical image database. In 2009 IEEE conference on computer vision and pattern recognition (2009), Ieee, pp. 248–255.
[10] Doersch, C., Gupta, A., and Efros, A. A. Unsupervised visual representation learning by context prediction. In 2015 IEEE International Conference on Computer Vision (ICCV) (2015), pp. 1422–1430.
[11] Gidaris, S., Singh, P., and Komodakis, N. Unsupervised representation learning by predicting image rotations. In International Conference on Learning Representations (2018).
[12] Gong, D., Liu, L., Le, V., Saha, B., Mansour, M. R., Venkatesh, S., and Van Den Hengel, A. Memorizing normality to detect anomaly: Memory-augmented deep autoencoder for unsupervised anomaly detection. 2019 IEEE/CVF International Conference on Computer Vision (ICCV) (Oct 2019).
[13] Huang, Y., Qiu, C., Guo, Y., Wang, X., and Yuan, K. Surface defect saliency of magnetic tile. In 2018 IEEE 14th International Conference on Automation Science and Engineering (CASE) (2018), pp. 612–617.
[14] Jin, X., and Han, J. K-Means Clustering. Springer US, Boston, MA, 2010, pp. 563–564.
[15] Li, C.L., Sohn, K., Yoon, J., and Pfister, T. Cutpaste: Self-supervised learning for anomaly detection and localization. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) (2021).
[16] Liu, W., Li, R., Zheng, M., Karanam, S., Wu, Z., Bhanu, B., Radke, R. J., and Camps, O. Towards visually explaining variational autoencoders.
arXiv:1911.07389 (2020).
[17] Mei, S., Yang, H., and Yin, Z. An unsupervised-learning-based approach for automated defect inspection on textured surfaces. IEEE Transactions on Instrumentation and Measurement 67, 6 (2018), 1266–1277.
[18] Mishra, P., Verk, R., Fornasier, D., Piciarelli, C., and Foresti, G. L. VT-ADL: A vision transformer network for image anomaly detection and localization. In 30th IEEE/IES International Symposium on Industrial Electronics (ISIE) (June 2021).
[19] Park, H., Noh, J., and Ham, B. Learning memory-guided normality for anomaly detection. In Proceedings of the IEEE/CVF Conference on Computer
Vision and Pattern Recognition (2020), pp. 14372–14381.
[20] Pirnay, J., and Chai, K. Inpainting transformer for anomaly detection. arXiv:2104.13897 (2021).
[21] Rudolph, M., Wandt, B., and Rosenhahn, B. Same same but differnet: Semi-supervised defect detection with normalizing flows. In Winter Conference on Applications of Computer Vision (WACV) (Jan. 2021).
[22] Ruff, L., Vandermeulen, R., Goernitz, N., Deecke, L., Siddiqui, S. A., Binder, A., Müller, E., and Kloft, M. Deep one-class classification. In Proceedings of the 35th International Conference on Machine Learning (10–15 Jul 2018), J. Dy and A. Krause, Eds., vol. 80 of Proceedings of Machine Learning Research, PMLR, pp. 4393–4402.
[23] Sabokrou, M., Khalooei, M., Fathy, M., and Adeli, E. Adversarially learned one-class classifier for novelty detection. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (2018), pp. 3379–3388.
[24] Simonyan, K., and Zisserman, A. Very deep convolutional networks for large-scale image recognition. In 3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 79, 2015, Conference Track Proceedings (2015).
[25] Tax, D. M., and Duin, R. P. Support vector data description. Machine Learning 54 (2004), 45–66.
[26] Venkataramanan, S., Peng, K.C., Singh, R. V., and Mahalanobis, A. Attention guided anomaly localization in images. arXiv:1911.08616 (2020).
[27] Wang, G., Han, S., Ding, E., and Huang, D. Student-teacher feature pyramid matching for unsupervised anomaly detection. arXiv:2103.04257 (2021).
[28] Wang, L., Zhang, D., Guo, J., and Han, Y. Image anomaly detection using normal data only by latent space resampling. Applied Sciences 10, 23 (2020).
[29] Yang, J., Shi, Y., and Qi, Z. Dfr: Deep feature reconstruction for unsupervised anomaly segmentation. ArXiv abs/2012.07122 (2020).
[30] Yi, J., and Yoon, S. Patch svdd: Patch-level svdd for anomaly detection and segmentation. In Proceedings of the Asian Conference on Computer Vision (ACCV) (November 2020).

簡易檢索 / 詳目顯示

相關論文