Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning
Automated inspection using deep-learning has been attracting attention for visual inspection at the manufacturing site. However, the inability to obtain sufficient abnormal product data for training deep- learning models is a problem in practical application. This study proposes an anomaly detection...
Main Authors: | , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
Taylor & Francis Group
2022-12-01
|
Series: | Applied Artificial Intelligence |
Online Access: | http://dx.doi.org/10.1080/08839514.2022.2094885 |
_version_ | 1797641104781213696 |
---|---|
author | Hironori Takimoto Junya Seki Sulfayanti F. Situju Akihiro Kanagawa |
author_facet | Hironori Takimoto Junya Seki Sulfayanti F. Situju Akihiro Kanagawa |
author_sort | Hironori Takimoto |
collection | DOAJ |
description | Automated inspection using deep-learning has been attracting attention for visual inspection at the manufacturing site. However, the inability to obtain sufficient abnormal product data for training deep- learning models is a problem in practical application. This study proposes an anomaly detection method based on the Siamese network with an attention mechanism for a small dataset. Moreover, attention branch loss (ABL) is proposed for Siamese network to render more task-specific attention maps from attention mechanism. Experimental results confirm that the proposed method with the attention mechanism and ABL is effective even with limited abnormal data. |
first_indexed | 2024-03-11T13:40:44Z |
format | Article |
id | doaj.art-c8e9bbced0a54511a6a7d676387f2507 |
institution | Directory Open Access Journal |
issn | 0883-9514 1087-6545 |
language | English |
last_indexed | 2024-03-11T13:40:44Z |
publishDate | 2022-12-01 |
publisher | Taylor & Francis Group |
record_format | Article |
series | Applied Artificial Intelligence |
spelling | doaj.art-c8e9bbced0a54511a6a7d676387f25072023-11-02T13:36:38ZengTaylor & Francis GroupApplied Artificial Intelligence0883-95141087-65452022-12-0136110.1080/08839514.2022.20948852094885Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot LearningHironori Takimoto0Junya Seki1Sulfayanti F. Situju2Akihiro Kanagawa3Okayama Prefectural UniversityOkayama Prefectural UniversitySulawesi Barat UniversityOkayama Prefectural UniversityAutomated inspection using deep-learning has been attracting attention for visual inspection at the manufacturing site. However, the inability to obtain sufficient abnormal product data for training deep- learning models is a problem in practical application. This study proposes an anomaly detection method based on the Siamese network with an attention mechanism for a small dataset. Moreover, attention branch loss (ABL) is proposed for Siamese network to render more task-specific attention maps from attention mechanism. Experimental results confirm that the proposed method with the attention mechanism and ABL is effective even with limited abnormal data.http://dx.doi.org/10.1080/08839514.2022.2094885 |
spellingShingle | Hironori Takimoto Junya Seki Sulfayanti F. Situju Akihiro Kanagawa Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning Applied Artificial Intelligence |
title | Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning |
title_full | Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning |
title_fullStr | Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning |
title_full_unstemmed | Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning |
title_short | Anomaly Detection Using Siamese Network with Attention Mechanism for Few-Shot Learning |
title_sort | anomaly detection using siamese network with attention mechanism for few shot learning |
url | http://dx.doi.org/10.1080/08839514.2022.2094885 |
work_keys_str_mv | AT hironoritakimoto anomalydetectionusingsiamesenetworkwithattentionmechanismforfewshotlearning AT junyaseki anomalydetectionusingsiamesenetworkwithattentionmechanismforfewshotlearning AT sulfayantifsituju anomalydetectionusingsiamesenetworkwithattentionmechanismforfewshotlearning AT akihirokanagawa anomalydetectionusingsiamesenetworkwithattentionmechanismforfewshotlearning |