A Pixel-Wise Foreign Object Debris Detection Method Based on Multi-Scale Feature Inpainting
Abstract
:1. Introduction
2. Related Work
3. Method
3.1. Multi-Scale Feature Extraction
3.2. Multi-Scale Grid Masks
3.3. Deep Feature Inpainting
3.4. Anomaly Detection and Localization
4. Experimental Setup
4.1. Datasets
4.2. Evaluation Metrics
4.3. Implementation Details
5. Results
5.1. Anomaly Detection
5.2. Anomaly Localization
6. Ablation Studies
6.1. Effectiveness of Multi-Scale Features
6.2. Loss Function
6.3. Grid Size
7. Discussion
8. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Conflicts of Interest
References
- Federal Aviation Administration (FAA). Foreign Object Debris (Fod) Management; Document Advisory Circular(ac) 150/5220-24; FAA: Washington, DC, USA, 2010. [Google Scholar]
- Munyer, T.; Huang, P.-C.; Huang, C.; Zhong, X. Fod-a: A dataset for foreign object debris in airports. arXiv 2021, arXiv:2110.03072. [Google Scholar]
- Cao, X.; Wang, P.; Meng, C.; Bai, X.; Gong, G.; Liu, M.; Qi, J. Region based CNN for foreign object debris detection on airfield pavement. Sensors 2018, 18, 737. [Google Scholar] [CrossRef] [PubMed]
- Hu, K.; Cui, D.; Zhang, Y.; Cao, C.; Xiao, F.; Huang, G. Classication of foreign object debris using integrated visual features and extreme learning machine. In CCF Chinese Conference on Computer Vision; Springer: Tianjin, China, 2017; pp. 3–13. [Google Scholar]
- Jing, Y.; Zheng, H.; Lin, C.; Zheng, W.; Dong, K.; Li, X. Foreign object debris detection for optical imaging sensors based on random forest. Sensors 2022, 22, 2463. [Google Scholar] [CrossRef] [PubMed]
- Doğru, A.; Bouarfa, S.; Arizar, R.; Aydoğan, R. Using Convolutional Neural Networks to Automate Aircraft Maintenance Visual Inspection. Aerospace 2020, 7, 171. [Google Scholar] [CrossRef]
- He, K.; Gkioxari, G.; Dollár, P.; Girshick, R. Mask r-cnn. In Proceedings of the 2017 IEEE International Conference on Computer Vision (ICCV), Venice, Italy, 22–29 October 2017; IEEE: Venice, Italy, 2017; pp. 2980–2988. [Google Scholar]
- Redmon, J.; Divvala, S.; Girshick, R.; Farhadi, A. You only look once: Unied, real-time object detection. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 27–30 June 2016; IEEE: LasVegas, NV, USA, 2016; pp. 77–788. [Google Scholar]
- Yang, Y.; Gong, H.; Wang, X.; Sun, P. Aerial target tracking algorithm based on faster r-cnn combined with frame differencing. Aerospace 2017, 4, 32. [Google Scholar] [CrossRef]
- Chalapathy, R.; Chawla, S. Deep learning for anomaly detection: A survey. arXiv 2019, arXiv:1901.03407. [Google Scholar]
- Pang, G.; Shen, C.; Cao, L.; Hengel, A.V.D. Deep learning for anomaly detection: A review. ACM Comput. 2021, 54, 38. [Google Scholar] [CrossRef]
- Luo, Z.; He, K.; Yu, Z. A robust unsupervised anomaly detection framework. Appl. Intell. 2022, 52, 6022–6036. [Google Scholar] [CrossRef]
- Park, H.; Noh, J.; Ham, B. Learning memory-guided normality for anomaly detection. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 13–19 June 2020; IEEE: Seattle, WA, USA, 2020; pp. 14360–14369. [Google Scholar]
- An, J.; Cho, S. Variational autoencoder based anomaly detection using reconstruction probability. Spec. Lect. IE 2015, 2, 1–18. [Google Scholar]
- Cohen, N.; Hoshen, Y. Sub-image anomaly detection with deep pyramid correspondences. arXiv 2020, arXiv:2005.02357. [Google Scholar]
- Rippel, O.; Mertens, P.; König, E.; Merhof, D. Gaussian anomaly detection by modeling the distribution of normal data in pretrained deep features. IEEE Trans. Instrum. Meas. 2021, 70, 1–13. [Google Scholar] [CrossRef]
- Wan, Q.; Gao, L.; Li, X.; Wen, L. Industrial image anomaly localization based on gaussian clustering of pretrained feature. IEEE Trans. Ind. Electron. 2022, 69, 6182–6192. [Google Scholar] [CrossRef]
- Bergmann, P.; Fauser, M.; Sattlegger, D.; Steger, C. Mvtec ad a comprehensive real-world dataset for unsupervised anomaly detection. In Proceedings of the 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Long Beach, CA, USA, 15–20 June 2019; IEEE: Long Beach, CA, USA, 2019; pp. 9584–9592. [Google Scholar]
- Bergmann, P.; Löwe, S.; Fauser, M.; Sattlegger, D.; Steger, C. Improving unsupervised defect segmentation by applying structural similarity to autoencoders. arXiv 2018, arXiv:1807.02011. [Google Scholar]
- Qin, K.; Wang, Q.; Lu, B.; Sun, H.; Shu, P. Flight Anomaly Detection via a Deep Hybrid Model. Aerospace 2022, 9, 329. [Google Scholar] [CrossRef]
- Kingma, D.P.; Welling, M. Auto-encoding variational bayes. In Proceedings of the 2nd International Conference on Learning Representations (ICLR), Banff, AB, Canada, 14–16 April 2014; pp. 9584–9592. [Google Scholar]
- Memarzadeh, M.; Matthews, B.; Avrekh, I. Unsupervised Anomaly Detection in Flight Data Using Convolutional Variational Auto-Encoder. Aerospace 2020, 7, 115. [Google Scholar] [CrossRef]
- Hou, J.; Zhang, Y.; Zhong, Q.; Xie, D.; Pu, S.; Zhou, H. Divideand-assemble: Learning block-wise memory for unsupervised anomaly detection. In Proceedings of the 2021 IEEE/CVF International Conference on Computer Vision (ICCV), Montreal, BC, Canada, 11–17 October 2021; IEEE: Montreal, QC, Canada, 2021; pp. 8771–8780. [Google Scholar]
- Gong, D.; Liu, L.; Le, V.; Saha, B.; Mansour, M.R.; Venkatesh, S.; Van Den Hengel, A. Memorizing normality to detect anomaly: Memoryaugmented deep autoencoder for unsupervised anomaly detection. In Proceedings of the 2019 IEEE/CVF International Conference on Computer Vision (ICCV), Seoul, Korea, 27 October–2 November 2019; IEEE: Seoul, Korea, 2019; pp. 1705–1714. [Google Scholar]
- Goodfellow, I.J.; Pouget-Abadie, J.; Mirza, M.; Xu, B.; Warde-Farley, D.; Ozair, S.; Courville, A.; Bengio, Y. Generative adversarial nets. In Proceedings of the 27th International Conference on Neural Information Processing Systems, Montreal, BC, Canada, 8–13 December 2014; MIT Press: Cambridge, MA, USA, 2014; pp. 2672–2680. [Google Scholar]
- Schlegl, T.; Seeböck, P.; Waldstein, S.M.; Schmidt-Erfurth, U.; Langs, G. Unsupervised anomaly detection with generative adversarial networks to guide marker discovery. In Information Processing in Medical Imaging; Springer: Cham, Switzerland, 2017; pp. 146–157. [Google Scholar]
- Li, Z.; Li, N.; Jiang, K.; Ma, Z.; Wei, X.; Hong, X.; Gong, Y. Superpixel masking and inpainting for self-supervised anomaly detection. In Proceedings of the 31st British Machine Vision Conference 2020, (BMVC), Cardiff, UK, 7–10 September 2020; BMVA Press: Cardiff, UK, 2020; pp. 7–10. [Google Scholar]
- Yan, X.; Zhang, H.; Xu, X.; Hu, X.; Heng, P.-A. Learning semantic context from normal samples for unsupervised anomaly detection. In Proceedings of the AAAI Conference on Articial Intelligence, Virtual, 2–9 February 2021; AAAI Press: Palo Alto, CA, USA, 2021; pp. 3110–3118. [Google Scholar]
- Ruff, L.; Vandermeulen, R.A.; Franks, B.J.; Müller, K.-R.; Kloft, M. Rethinking assumptions in deep anomaly detection. arXiv 2020, arXiv:2006.00339. [Google Scholar]
- Liznerski, P.; Ruff, L.; Vandermeulen, R.A.; Franks, B.J.; Kloft, M.; Müller, K.-R. Explainable deep one-class classication. arXiv 2020, arXiv:2007.01760. [Google Scholar]
- Salehi, M.; Sadjadi, N.; Baselizadeh, S.; Rohban, M.H.; Rabiee, H.R. Multiresolution knowledge distillation for anomaly detection. In Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 20–25 June 2021; IEEE: Nashville, TN, USA, 2021; pp. 14897–14907. [Google Scholar]
- Wang, S.; Wu, L.; Cui, L.; Shen, Y. Glancing at the patch: Anomaly localization with global and local feature comparison. In Proceedings of the 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Nashville, TN, USA, 20–25 June 2021; IEEE: Nashville, TN, USA, 2021; pp. 254–263. [Google Scholar]
- Ruff, L.; Vandermeulen, R.; Goernitz, N.; Deecke, L.; Siddiqui, S.A.; Binder, A.; Müller, E.; Kloft, M. Deep one-class classication. In Proceedings of the International Conference on Machine Learning, Stockholm, Sweden, 10–15 July 2018; pp. 4393–4402. [Google Scholar]
- Yi, J.; Yoon, S. Patch svdd: Patch-level svdd for anomaly detection and segmentation. In Computer Vision—ACCV 2020; Springer: Cham, Switzerland, 2021; pp. 375–390. [Google Scholar]
- Bergmann, P.; Fauser, M.; Sattlegger, D.; Steger, C. Uninformed students: Student-teacher anomaly detection with discriminative latent embeddings. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Seattle, WA, USA, 13–19 June 2020; IEEE: Seattle, WA, USA, 2020; pp. 4182–4191. [Google Scholar]
- He, K.; Zhang, X.; Ren, S.; Sun, J. Deep residual learning for image recognition. In Proceedings of the 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), Las Vegas, NV, USA, 27–30 June 2016; IEEE: Las Vegas, NV, USA, 2016; pp. 770–778. [Google Scholar]
- Ronneberger, O.; Fischer, P.; Brox, T. U-net: Convolutional networks for biomedical image segmentation. In International Conference on Medical Image Computing and Computer-Assisted Intervention; Springer: Cham, Switzerland, 2015; pp. 234–241. [Google Scholar]
- Simonyan, K.; Zisserman, A. Very deep convolutional networks for largescale image recognition. arXiv 2014, arXiv:1409.1556. [Google Scholar]
- Deng, J.; Dong, W.; Socher, R.; Li, L.-J.; Li, K.; Li, F.-F. Imagenet: A large-scale hierarchical image database. In Proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, Miami, FL, USA, 20–25 June 2009; IEEE: Miami, FL, USA, 2009; pp. 248–255. [Google Scholar]
- Zavrtanik, V.; Kristan, M.; Skočaj, D. Reconstruction by inpainting for visual anomaly detection. Pattern Recognit. 2021, 112, 107706. [Google Scholar] [CrossRef]
- Shi, Y.; Yang, J.; Qi, Z. Unsupervised anomaly segmentation via deep feature reconstruction. Neurocomputing 2021, 424, 9–22. [Google Scholar] [CrossRef]
Category | AE- | RIAD | MRKD | DFR | MSFI | |||||
---|---|---|---|---|---|---|---|---|---|---|
AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | |
Carpet | 0.539 | 0.863 | 0.842 | 0.859 | 0.792 | 0.859 | 0.961 | 0.938 | 0.976 | 0.960 |
Grid | 0.779 | 0.855 | 0.996 | 0.957 | 0.780 | 0.862 | 0.968 | 0.927 | 0.921 | 0.922 |
Leather | 0.841 | 0.865 | 1.000 | 0.956 | 0.950 | 0.847 | 0.984 | 0.963 | 1.000 | 0.995 |
Tile | 0.795 | 0.847 | 0.987 | 0.850 | 0.915 | 0.830 | 0.896 | 0.883 | 0.870 | 0.894 |
Wood | 0.892 | 0.902 | 0.930 | 0.884 | 0.942 | 0.923 | 0.981 | 0.977 | 0.996 | 0.983 |
Bottle | 0.877 | 0.889 | 0.999 | 0.968 | 0.993 | 0.855 | 0.993 | 0.980 | 1.000 | 0.986 |
Cable | 0.477 | 0.755 | 0.819 | 0.755 | 0.891 | 0.755 | 0.831 | 0.809 | 0.967 | 0.923 |
Capsule | 0.660 | 0.904 | 0.884 | 0.906 | 0.804 | 0.926 | 0.975 | 0.983 | 0.888 | 0.943 |
Hazelnut | 0.951 | 0.924 | 0.833 | 0.865 | 0.983 | 0.843 | 0.989 | 0.982 | 0.995 | 0.978 |
Metal Nut | 0.415 | 0.889 | 0.885 | 0.893 | 0.735 | 0.889 | 0.929 | 0.921 | 0.955 | 0.963 |
Pill | 0.625 | 0.912 | 0.838 | 0.919 | 0.827 | 0.912 | 0.931 | 0.928 | 0.942 | 0.954 |
Screw | 0.746 | 0.878 | 0.845 | 0.865 | 0.833 | 0.863 | 0.958 | 0.931 | 0.866 | 0.878 |
Toothbrush | 0.589 | 0.829 | 1.000 | 0.967 | 0.921 | 0.817 | 0.981 | 0.964 | 0.969 | 0.933 |
Transistor | 0.703 | 0.619 | 0.909 | 0.619 | 0.855 | 0.569 | 0.801 | 0.787 | 0.964 | 0.916 |
Zipper | 0.765 | 0.881 | 0.981 | 0.963 | 0.932 | 0.877 | 0.903 | 0.915 | 0.925 | 0.955 |
Mean | 0.710 | 0.854 | 0.917 | 0.882 | 0.877 | 0.842 | 0.939 | 0.926 | 0.949 | 0.945 |
Category | AE- | RIAD | MRKD | DFR | MSFI | |||||
---|---|---|---|---|---|---|---|---|---|---|
AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | |
Screws | 0.715 | 0.712 | 0.788 | 0.754 | 0.933 | 0.908 | 0.777 | 0.775 | 0.997 | 0.947 |
Nuts | 0.670 | 0.629 | 0.759 | 0.758 | 0.944 | 0.897 | 0.975 | 0.970 | 0.995 | 0.954 |
Steel balls | 0.713 | 0.710 | 0.741 | 0.763 | 0.924 | 0.873 | 0.745 | 0.714 | 0.993 | 0.984 |
Gaskets | 0.799 | 0.726 | 0.843 | 0.765 | 0.959 | 0.951 | 0.991 | 0.983 | 0.998 | 0.990 |
Locks | 0.864 | 0.808 | 0.941 | 0.846 | 0.924 | 0.907 | 0.985 | 0.968 | 0.979 | 0.634 |
Rubber blocks | 0.744 | 0.769 | 0.891 | 0.853 | 0.931 | 0.918 | 0.990 | 0.977 | 0.993 | 0.977 |
Stones | 0.771 | 0.833 | 0.787 | 0.782 | 0.921 | 0.900 | 0.932 | 0.905 | 0.968 | 0.918 |
Silver metal cylinders | 0.777 | 0.699 | 0.835 | 0.770 | 0.928 | 0.906 | 0.989 | 0.984 | 0.986 | 0.976 |
White plastic cylinders | 0.733 | 0.734 | 0.844 | 0.822 | 0.924 | 0.915 | 0.993 | 0.984 | 0.980 | 0.971 |
Golden plastic cylinders | 0.744 | 0.681 | 0.861 | 0.829 | 0.925 | 0.913 | 0.982 | 0.970 | 0.973 | 0.961 |
Silver metal spheres | 0.742 | 0.837 | 0.823 | 0.781 | 0.937 | 0.929 | 0.992 | 0.984 | 0.998 | 0.990 |
Golden metal spheres | 0.898 | 0.855 | 0.981 | 0.851 | 0.944 | 0.921 | 0.987 | 0.964 | 0.997 | 0.974 |
Glass spheres | 0.802 | 0.807 | 0.876 | 0.780 | 0.932 | 0.922 | 0.985 | 0.975 | 0.978 | 0.958 |
Golden marble spheres | 0.771 | 0.702 | 0.841 | 0.762 | 0.932 | 0.922 | 0.987 | 0.977 | 0.989 | 0.979 |
White marble spheres | 0.721 | 0.711 | 0.943 | 0.860 | 0.926 | 0.922 | 0.986 | 0.982 | 0.976 | 0.972 |
Mean | 0.764 | 0.748 | 0.850 | 0.798 | 0.932 | 0.914 | 0.953 | 0.941 | 0.987 | 0.946 |
Category | AE- | RIAD | MRKD | DFR | MSFI | |||||
---|---|---|---|---|---|---|---|---|---|---|
AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | |
Carpet | 0.566 | 0.153 | 0.963 | 0.386 | 0.956 | 0.458 | 0.970 | 0.554 | 0.980 | 0.680 |
Grid | 0.605 | 0.132 | 0.988 | 0.392 | 0.917 | 0.432 | 0.980 | 0.406 | 0.992 | 0.535 |
Leather | 0.735 | 0.275 | 0.994 | 0.558 | 0.981 | 0.236 | 0.980 | 0.380 | 0.996 | 0.518 |
Tile | 0.593 | 0.258 | 0.891 | 0.425 | 0.827 | 0.655 | 0.870 | 0.535 | 0.928 | 0.546 |
Wood | 0.734 | 0.346 | 0.858 | 0.317 | 0.848 | 0.426 | 0.930 | 0.449 | 0.914 | 0.522 |
Bottle | 0.704 | 0.294 | 0.984 | 0.650 | 0.963 | 0.340 | 0.970 | 0.719 | 0.968 | 0.760 |
Cable | 0.750 | 0.254 | 0.842 | 0.311 | 0.824 | 0.411 | 0.920 | 0.635 | 0.971 | 0.465 |
Capsule | 0.788 | 0.205 | 0.928 | 0.383 | 0.958 | 0.248 | 0.990 | 0.499 | 0.974 | 0.591 |
Hazelnut | 0.788 | 0.544 | 0.961 | 0.468 | 0.946 | 0.238 | 0.990 | 0.634 | 0.981 | 0.729 |
Metal Nut | 0.704 | 0.424 | 0.925 | 0.523 | 0.863 | 0.530 | 0.930 | 0.862 | 0.972 | 0.769 |
Pill | 0.855 | 0.376 | 0.957 | 0.514 | 0.896 | 0.171 | 0.970 | 0.738 | 0.977 | 0.727 |
Screw | 0.898 | 0.156 | 0.988 | 0.390 | 0.959 | 0.390 | 0.990 | 0.281 | 0.963 | 0.591 |
Toothbrush | 0.864 | 0.217 | 0.989 | 0.552 | 0.961 | 0.547 | 0.990 | 0.656 | 0.985 | 0.647 |
Transistor | 0.548 | 0.212 | 0.877 | 0.395 | 0.764 | 0.381 | 0.800 | 0.642 | 0.924 | 0.489 |
Zipper | 0.682 | 0.220 | 0.978 | 0.627 | 0.939 | 0.320 | 0.960 | 0.441 | 0.952 | 0.660 |
Mean | 0.720 | 0.271 | 0.942 | 0.459 | 0.907 | 0.385 | 0.949 | 0.562 | 0.965 | 0.615 |
Category | AE- | RIAD | MRKD | DFR | MSFI | |||||
---|---|---|---|---|---|---|---|---|---|---|
AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | AUROC | F1 | |
Screws | 0.714 | 0.241 | 0.826 | 0.301 | 0.964 | 0.479 | 0.936 | 0.651 | 0.995 | 0.704 |
Nuts | 0.700 | 0.237 | 0.864 | 0.305 | 0.989 | 0.610 | 0.993 | 0.826 | 0.997 | 0.849 |
Steel balls | 0.740 | 0.215 | 0.876 | 0.245 | 0.980 | 0.260 | 0.953 | 0.544 | 0.997 | 0.697 |
Gaskets | 0.718 | 0.253 | 0.927 | 0.496 | 0.990 | 0.633 | 0.996 | 0.862 | 0.998 | 0.924 |
Locks | 0.703 | 0.445 | 0.868 | 0.627 | 0.861 | 0.538 | 0.979 | 0.887 | 0.985 | 0.875 |
Rubber blocks | 0.678 | 0.433 | 0.845 | 0.522 | 0.869 | 0.461 | 0.952 | 0.772 | 0.953 | 0.920 |
Stones | 0.597 | 0.252 | 0.741 | 0.291 | 0.898 | 0.498 | 0.971 | 0.604 | 0.946 | 0.562 |
Silver metal cylinders | 0.576 | 0.284 | 0.719 | 0.330 | 0.929 | 0.561 | 0.989 | 0.809 | 0.987 | 0.783 |
White plastic cylinders | 0.527 | 0.292 | 0.672 | 0.494 | 0.852 | 0.498 | 0.990 | 0.827 | 0.985 | 0.810 |
Golden plastic cylinders | 0.779 | 0.378 | 0.693 | 0.340 | 0.896 | 0.548 | 0.976 | 0.727 | 0.960 | 0.798 |
Silver metal spheres | 0.637 | 0.289 | 0.800 | 0.379 | 0.962 | 0.452 | 0.992 | 0.832 | 0.992 | 0.833 |
Golden metal spheres | 0.695 | 0.308 | 0.644 | 0.291 | 0.968 | 0.487 | 0.992 | 0.823 | 0.994 | 0.862 |
Glass spheres | 0.589 | 0.282 | 0.714 | 0.309 | 0.949 | 0.553 | 0.979 | 0.703 | 0.984 | 0.737 |
Golden marble spheres | 0.695 | 0.325 | 0.595 | 0.290 | 0.957 | 0.526 | 0.990 | 0.819 | 0.991 | 0.851 |
White marble spheres | 0.569 | 0.298 | 0.751f | 0.455 | 0.896 | 0.539 | 0.980 | 0.735 | 0.985 | 0.783 |
Mean | 0.661 | 0.302 | 0.769 | 0.378 | 0.930 | 0.509 | 0.977 | 0.761 | 0.983 | 0.799 |
Category | Metric | Last1 | Last2 | Last3 | Last4 |
---|---|---|---|---|---|
MVTec AD | Image-level AUROC | 0.828 | 0.905 | 0.944 | 0.949 |
Pixel-level AUROC | 0.822 | 0.910 | 0.950 | 0.965 | |
FOD | Image-level AUROC | 0.878 | 0.919 | 0.984 | 0.988 |
Pixel-level AUROC | 0.826 | 0.908 | 0.976 | 0.985 |
Category | Metric | |||
---|---|---|---|---|
MVTec AD | Image-level AUROC | 0.835 | 0.934 | 0.949 |
Pixel-level AUROC | 0.863 | 0.949 | 0.965 | |
FOD | Image-level AUROC | 0.956 | 0.967 | 0.988 |
Pixel-level AUROC | 0.966 | 0.976 | 0.985 |
Category | Metric | |||
---|---|---|---|---|
MVTec AD | Image-level AUROC | 0.914 | 0.936 | 0.920 |
Pixel-level AUROC | 0.922 | 0.952 | 0.931 | |
FOD | Image-level AUROC | 0.948 | 0.967 | 0.935 |
Pixel-level AUROC | 0.936 | 0.956 | 0.926 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Jing, Y.; Zheng, H.; Zheng, W.; Dong, K. A Pixel-Wise Foreign Object Debris Detection Method Based on Multi-Scale Feature Inpainting. Aerospace 2022, 9, 480. https://doi.org/10.3390/aerospace9090480
Jing Y, Zheng H, Zheng W, Dong K. A Pixel-Wise Foreign Object Debris Detection Method Based on Multi-Scale Feature Inpainting. Aerospace. 2022; 9(9):480. https://doi.org/10.3390/aerospace9090480
Chicago/Turabian StyleJing, Ying, Hong Zheng, Wentao Zheng, and Kaihan Dong. 2022. "A Pixel-Wise Foreign Object Debris Detection Method Based on Multi-Scale Feature Inpainting" Aerospace 9, no. 9: 480. https://doi.org/10.3390/aerospace9090480