Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data
In a nuclear power plant (NPP), the used tools are visually inspected to ensure their integrity before and after their use in the nuclear reactor. The manual inspection is usually performed by qualified technicians and takes a large amount of time (weeks up to months). In this work, we propose an au...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2023-07-01
|
| Series: | Computation |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-3197/11/7/139 |
| _version_ | 1827733396552417280 |
|---|---|
| author | Hossam A. Gabbar Oluwabukola Grace Adegboro Abderrazak Chahid Jing Ren |
| author_facet | Hossam A. Gabbar Oluwabukola Grace Adegboro Abderrazak Chahid Jing Ren |
| author_sort | Hossam A. Gabbar |
| collection | DOAJ |
| description | In a nuclear power plant (NPP), the used tools are visually inspected to ensure their integrity before and after their use in the nuclear reactor. The manual inspection is usually performed by qualified technicians and takes a large amount of time (weeks up to months). In this work, we propose an automated tool inspection that uses a classification model for anomaly detection. The deep learning model classifies the computed tomography (CT) images as defective (with missing components) or defect-free. Moreover, the proposed algorithm enables incremental learning (IL) using a proposed thresholding technique to ensure a high prediction confidence by continuous online training of the deployed online anomaly detection model. The proposed algorithm is tested with existing state-of-the-art IL methods showing that it helps the model quickly learn the anomaly patterns. In addition, it enhances the classification model confidence while preserving a desired minimal performance. |
| first_indexed | 2024-03-11T01:10:48Z |
| format | Article |
| id | doaj.art-829a35fa8bdc46469e86726adcd2dfa6 |
| institution | Directory Open Access Journal |
| issn | 2079-3197 |
| language | English |
| last_indexed | 2024-03-11T01:10:48Z |
| publishDate | 2023-07-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Computation |
| spelling | doaj.art-829a35fa8bdc46469e86726adcd2dfa62023-11-18T18:52:12ZengMDPI AGComputation2079-31972023-07-0111713910.3390/computation11070139Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography DataHossam A. Gabbar0Oluwabukola Grace Adegboro1Abderrazak Chahid2Jing Ren3Faculty of Engineering and Applied Science, Ontario Tech University (UOIT), Oshawa, ON L1G 0C5, CanadaFaculty of Engineering and Applied Science, Ontario Tech University (UOIT), Oshawa, ON L1G 0C5, CanadaFaculty of Engineering and Applied Science, Ontario Tech University (UOIT), Oshawa, ON L1G 0C5, CanadaFaculty of Engineering and Applied Science, Ontario Tech University (UOIT), Oshawa, ON L1G 0C5, CanadaIn a nuclear power plant (NPP), the used tools are visually inspected to ensure their integrity before and after their use in the nuclear reactor. The manual inspection is usually performed by qualified technicians and takes a large amount of time (weeks up to months). In this work, we propose an automated tool inspection that uses a classification model for anomaly detection. The deep learning model classifies the computed tomography (CT) images as defective (with missing components) or defect-free. Moreover, the proposed algorithm enables incremental learning (IL) using a proposed thresholding technique to ensure a high prediction confidence by continuous online training of the deployed online anomaly detection model. The proposed algorithm is tested with existing state-of-the-art IL methods showing that it helps the model quickly learn the anomaly patterns. In addition, it enhances the classification model confidence while preserving a desired minimal performance.https://www.mdpi.com/2079-3197/11/7/139incremental learningcontinual learningcomputed tomographyanomaly detectionclassificationindustrial inspection |
| spellingShingle | Hossam A. Gabbar Oluwabukola Grace Adegboro Abderrazak Chahid Jing Ren Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data Computation incremental learning continual learning computed tomography anomaly detection classification industrial inspection |
| title | Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data |
| title_full | Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data |
| title_fullStr | Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data |
| title_full_unstemmed | Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data |
| title_short | Incremental Learning-Based Algorithm for Anomaly Detection Using Computed Tomography Data |
| title_sort | incremental learning based algorithm for anomaly detection using computed tomography data |
| topic | incremental learning continual learning computed tomography anomaly detection classification industrial inspection |
| url | https://www.mdpi.com/2079-3197/11/7/139 |
| work_keys_str_mv | AT hossamagabbar incrementallearningbasedalgorithmforanomalydetectionusingcomputedtomographydata AT oluwabukolagraceadegboro incrementallearningbasedalgorithmforanomalydetectionusingcomputedtomographydata AT abderrazakchahid incrementallearningbasedalgorithmforanomalydetectionusingcomputedtomographydata AT jingren incrementallearningbasedalgorithmforanomalydetectionusingcomputedtomographydata |