Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection
In industrial contexts, anomaly detection is crucial for ensuring quality control and maintaining operational efficiency in manufacturing processes. Leveraging high-level features extracted from ImageNet-trained networks and the robust capabilities of the Deep Support Vector Data Description (SVDD)...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2024-12-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/25/1/67 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841548970857332736 |
|---|---|
| author | Wei Huang Yongjie Li Zhaonan Xu Xinwei Yao Rongchun Wan |
| author_facet | Wei Huang Yongjie Li Zhaonan Xu Xinwei Yao Rongchun Wan |
| author_sort | Wei Huang |
| collection | DOAJ |
| description | In industrial contexts, anomaly detection is crucial for ensuring quality control and maintaining operational efficiency in manufacturing processes. Leveraging high-level features extracted from ImageNet-trained networks and the robust capabilities of the Deep Support Vector Data Description (SVDD) model for anomaly detection, this paper proposes an improved Deep SVDD model, termed Feature-Patching SVDD (FPSVDD), designed for unsupervised anomaly detection in industrial applications. This model integrates a feature-patching technique with the Deep SVDD framework. Features are extracted from a pre-trained backbone network on ImageNet, and each extracted feature is split into multiple small patches of appropriate size. This approach effectively captures both macro-structural information and fine-grained local information from the extracted features, enhancing the model’s sensitivity to anomalies. The feature patches are then aggregated and concatenated for further training with the Deep SVDD model. Experimental results on both the MvTec AD and CIFAR-10 datasets demonstrate that our model outperforms current mainstream approaches and provides significant improvements in anomaly detection performance, which is vital for industrial quality assurance and defect detection in real-time manufacturing scenarios. |
| format | Article |
| id | doaj-art-18f6b3a8bf8b4353b2047d05b53bf8c5 |
| institution | Kabale University |
| issn | 1424-8220 |
| language | English |
| publishDate | 2024-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj-art-18f6b3a8bf8b4353b2047d05b53bf8c52025-01-10T13:20:46ZengMDPI AGSensors1424-82202024-12-012516710.3390/s25010067Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly DetectionWei Huang0Yongjie Li1Zhaonan Xu2Xinwei Yao3Rongchun Wan4College of Computer Science, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Computer Science, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Computer Science, Zhejiang University of Technology, Hangzhou 310023, ChinaCollege of Computer Science, Zhejiang University of Technology, Hangzhou 310023, ChinaZhejiang HOUDAR Intelligent Technology Co., Ltd., Hangzhou 310023, ChinaIn industrial contexts, anomaly detection is crucial for ensuring quality control and maintaining operational efficiency in manufacturing processes. Leveraging high-level features extracted from ImageNet-trained networks and the robust capabilities of the Deep Support Vector Data Description (SVDD) model for anomaly detection, this paper proposes an improved Deep SVDD model, termed Feature-Patching SVDD (FPSVDD), designed for unsupervised anomaly detection in industrial applications. This model integrates a feature-patching technique with the Deep SVDD framework. Features are extracted from a pre-trained backbone network on ImageNet, and each extracted feature is split into multiple small patches of appropriate size. This approach effectively captures both macro-structural information and fine-grained local information from the extracted features, enhancing the model’s sensitivity to anomalies. The feature patches are then aggregated and concatenated for further training with the Deep SVDD model. Experimental results on both the MvTec AD and CIFAR-10 datasets demonstrate that our model outperforms current mainstream approaches and provides significant improvements in anomaly detection performance, which is vital for industrial quality assurance and defect detection in real-time manufacturing scenarios.https://www.mdpi.com/1424-8220/25/1/67anomaly detectionpre-trained networkdeep SVDDfeature patchingunsupervised learning |
| spellingShingle | Wei Huang Yongjie Li Zhaonan Xu Xinwei Yao Rongchun Wan Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection Sensors anomaly detection pre-trained network deep SVDD feature patching unsupervised learning |
| title | Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection |
| title_full | Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection |
| title_fullStr | Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection |
| title_full_unstemmed | Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection |
| title_short | Improved Deep Support Vector Data Description Model Using Feature Patching for Industrial Anomaly Detection |
| title_sort | improved deep support vector data description model using feature patching for industrial anomaly detection |
| topic | anomaly detection pre-trained network deep SVDD feature patching unsupervised learning |
| url | https://www.mdpi.com/1424-8220/25/1/67 |
| work_keys_str_mv | AT weihuang improveddeepsupportvectordatadescriptionmodelusingfeaturepatchingforindustrialanomalydetection AT yongjieli improveddeepsupportvectordatadescriptionmodelusingfeaturepatchingforindustrialanomalydetection AT zhaonanxu improveddeepsupportvectordatadescriptionmodelusingfeaturepatchingforindustrialanomalydetection AT xinweiyao improveddeepsupportvectordatadescriptionmodelusingfeaturepatchingforindustrialanomalydetection AT rongchunwan improveddeepsupportvectordatadescriptionmodelusingfeaturepatchingforindustrialanomalydetection |