Browse > Article
http://dx.doi.org/10.9708/jksci.2021.26.08.023

Comparative Analysis of Anomaly Detection Models using AE and Suggestion of Criteria for Determining Outliers  

Kang, Gun-Ha (Epozen's research institute)
Sohn, Jung-Mo (Epozen's research institute)
Sim, Gun-Wu (Epozen's research institute)
Abstract
In this study, we present a comparative analysis of major autoencoder(AE)-based anomaly detection methods for quality determination in the manufacturing process and a new anomaly discrimination criterion. Due to the characteristics of manufacturing site, anomalous instances are few and their types greatly vary. These properties degrade the performance of an AI-based anomaly detection model using the dataset for both normal and anomalous cases, and incur a lot of time and costs in obtaining additional data for performance improvement. To solve this problem, the studies on AE-based models such as AE and VAE are underway, which perform anomaly detection using only normal data. In this work, based on Convolutional AE, VAE, and Dilated VAE models, statistics on residual images, MSE, and information entropy were selected as outlier discriminant criteria to compare and analyze the performance of each model. In particular, the range value applied to the Convolutional AE model showed the best performance with AUC PRC 0.9570, F1 Score 0.8812 and AUC ROC 0.9548, accuracy 87.60%. This shows a performance improvement of an accuracy about 20%P(Percentage Point) compared to MSE, which was frequently used as a standard for determining outliers, and confirmed that model performance can be improved according to the criteria for determining outliers.
Keywords
Deep Learning; Manufacture Process; Anomaly Detection; Autoencoder;
Citations & Related Records
연도 인용수 순위
  • Reference
1 Yeong-Tae Baek, Jae-Gyu Sim, Chan-Young Pak, Se-Hoon Lee, PCB Defect Inspection using Deep Learning. KSCI, pp. 325-326, 2018.
2 Chul-jin, Park. A study on the quality information management for preventing human errors at quality of ship.. Ulsan University Graduate School of Automotive Vessel Technology, 2016, UCI:I084:48009-000002213977.
3 E.S. Lee, H.C. Bae, H.J. Kim, H.N. Han, Y.K. Lee, J.Y. Son, Trends in AI Technology for Smart Manufacturing in the Future. Electronics and telecommunications trends, Vol.35, No.1, pp. 60-70, 2020, DOI:10.22648/ETRI.2020.J.350106.   DOI
4 Raghavendra Chalapathy and Sanjay Chawla, Deep Learning for Anomaly Detection: A Survey. CoRR, (abs/1901.03407), 2019.
5 Alberto Tellaeche Iglesias, Miguel Angel Campos Anaya, Gonzalo Pajares Martinsanz and Iker Pastor-Lopez, On Combining Convolutional Autoencoders and Support Vector Machines for Fault Detection in Industrial Textures, Sensors, 21(10), 3339, 2021, DOI:10.3390/s21103339.   DOI
6 Jungsuk Kim,Jungbeom Ko,Hojong Choi and Hyunchul Kim, Printed Circuit Board Defect Detection Using Deep Learning via A Skip-Connected Convolutional Autoencoder, Sensors, 21(15), 4968, 2021, DOI:10.3390/s21154968.   DOI
7 Jonathan Masci, Ueli Meier, Dan Ciresan, and Jurgen Schmidhuber, Stacked convolutional auto-encoders for hierarchical feature extraction. In International Conference on Artificial Neural Networks, ICANN, pp. 52-59, 2011, DOI:10.1007/978-3-642-21735-7_7.   DOI
8 Robert Muller, Fabian Ritz, Steffen lllium, Claudia Linnhoff-Popien, Acoustic Anomaly Detection for Machine Sounds based on ImageTransfer Learning, CVPR, 2021, DOI:10.5220/0010185800490056.   DOI
9 Diederik P Kingma and Max Welling, Auto-encoding variational bayes. ICLR. arXiv preprintarXiv:1312.6114, 2013.
10 Omid E. David, Nathan S.Netanyahu, DeepPainter: Painter Classification Using Deep Convolutional Autoencoders. ICANN, Vol.9887, pp. 20-28, 2016, DOI:10/1007/978-3-319-44781-0_3.
11 Taeu Kim, https://taeu.github.io/paper/deeplearning-paper-vae/.
12 Sunil Park, Taeho Oh, Application of Receiver Operating Characteristic(ROC) Curve for Evaluation of Diagnostic Test Performance. Journal of Veterinary Clinics, pp. 97-101, 2016, DOI:10.17555/jvc.2016.04.33.2.97.   DOI
13 Shengyuan Piao and Jiaming Liu, Accuracy improvement of unet based on dilated convolution. Journal of Physics: Conference Series. Vol. 1345. No. 5. IOP Publishing, 2019, DOI:10.1088/1742-6596/1345/5/052066.   DOI
14 Jinwon An and Sungzoon Cho, Variational Autoencoder based AnomalyDetection using Reconstruction Probability. Technical Report, SNU Data MiningCenter, pp. 1-18, 2015.
15 VESAL.Sulaiman, RAVIKUMAR.Nishant, MAIER.Andreas. A 2D dilated residual U-Net for multi-organ segmentation in thoracic CT. arXiv preprint arXiv:1905.07710, 2019.
16 Ravirajsinh Dabhi, Casting product image data for quality inspection, Kaggle, https://www.kaggle.com/ravirajsinh45/real-life-industrial-dataset-of-casting-product.
17 The Nature of Statistics Bigpicture, Statistics Basics Lecture 1. Mean, Deviation, Variance, Standard Deviations, https://hsm-edu.tistory.com/1182
18 Garlic, https://m.blog.naver.com/yk60park/222100758577
19 Wikipedia, Entropy(Information Theory), https://en.wikipedia.org/wiki/Entropy_(information_theory)
20 Kim, Chungyun, Evaluation Method of Classification Model, https://velog.io/@skyepodium/%EB%B6%84%EB%A5%98-%EB%AA%A8%EB%8D%B8-%ED%8F%89%EA%B0%80-%EB%B0%A9%EB%B2%95.
21 Jakub Czakon, https://neptune.ai/blog/f1-score-accuracy-roc-aucpr-auc
22 Ho Young, Lee. Changes in labour demand and acceptability due to the Fourth Industrial Revoluation. Korea Information Society Development, 2019.
23 CityWizard, https://m.blog.naver.com/PostView.naver?blogId=y4769&logNo=220505513170&proxyReferer=https:%2F%2Fwww.google.com%2F