Anomaly detection from images in pipes using GAN
Abstract In recent years, the number of pipes that have exceeded their service life has increased. For this reason, earthworm-type robots equipped with cameras have been developed to perform regularly inspections of sewer pipes. However, inspection methods have not yet been established. This paper p...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
SpringerOpen
2023-03-01
|
| Series: | ROBOMECH Journal |
| Subjects: | |
| Online Access: | https://doi.org/10.1186/s40648-023-00246-y |
| _version_ | 1797864341315256320 |
|---|---|
| author | Shigeki Yumoto Takumi Kitsukawa Alessandro Moro Sarthak Pathak Taro Nakamura Kazunori Umeda |
| author_facet | Shigeki Yumoto Takumi Kitsukawa Alessandro Moro Sarthak Pathak Taro Nakamura Kazunori Umeda |
| author_sort | Shigeki Yumoto |
| collection | DOAJ |
| description | Abstract In recent years, the number of pipes that have exceeded their service life has increased. For this reason, earthworm-type robots equipped with cameras have been developed to perform regularly inspections of sewer pipes. However, inspection methods have not yet been established. This paper proposes a method for anomaly detection from images in pipes using Generative Adversarial Network (GAN). A model that combines f-AnoGAN and Lightweight GAN is used to detect anomalies by taking the difference between input images and generated images. Since the GANs are only trained with non-defective images, they are able to convert an image containing defects into one without them. Subtraction images is used to estimate the location of anomalies. Experiments were conducted using actual images of cast iron pipes to confirm the effectiveness of the proposed method. It was also validated using sewer-ml, a public dataset. |
| first_indexed | 2024-04-09T22:51:22Z |
| format | Article |
| id | doaj.art-3937e45133354615a8b8d6a656a5d1cd |
| institution | Directory Open Access Journal |
| issn | 2197-4225 |
| language | English |
| last_indexed | 2024-04-09T22:51:22Z |
| publishDate | 2023-03-01 |
| publisher | SpringerOpen |
| record_format | Article |
| series | ROBOMECH Journal |
| spelling | doaj.art-3937e45133354615a8b8d6a656a5d1cd2023-03-22T11:37:52ZengSpringerOpenROBOMECH Journal2197-42252023-03-0110111210.1186/s40648-023-00246-yAnomaly detection from images in pipes using GANShigeki Yumoto0Takumi Kitsukawa1Alessandro Moro2Sarthak Pathak3Taro Nakamura4Kazunori Umeda5Graduate School of Science and Engineering, Chuo UniversityGraduate School of Science and Engineering, Chuo UniversityGraduate School of Science and Engineering, Chuo UniversityGraduate School of Science and Engineering, Chuo UniversityGraduate School of Science and Engineering, Chuo UniversityGraduate School of Science and Engineering, Chuo UniversityAbstract In recent years, the number of pipes that have exceeded their service life has increased. For this reason, earthworm-type robots equipped with cameras have been developed to perform regularly inspections of sewer pipes. However, inspection methods have not yet been established. This paper proposes a method for anomaly detection from images in pipes using Generative Adversarial Network (GAN). A model that combines f-AnoGAN and Lightweight GAN is used to detect anomalies by taking the difference between input images and generated images. Since the GANs are only trained with non-defective images, they are able to convert an image containing defects into one without them. Subtraction images is used to estimate the location of anomalies. Experiments were conducted using actual images of cast iron pipes to confirm the effectiveness of the proposed method. It was also validated using sewer-ml, a public dataset.https://doi.org/10.1186/s40648-023-00246-yInfrastructure inspectionSewer pipeDeep learningGANAnomaly detection |
| spellingShingle | Shigeki Yumoto Takumi Kitsukawa Alessandro Moro Sarthak Pathak Taro Nakamura Kazunori Umeda Anomaly detection from images in pipes using GAN ROBOMECH Journal Infrastructure inspection Sewer pipe Deep learning GAN Anomaly detection |
| title | Anomaly detection from images in pipes using GAN |
| title_full | Anomaly detection from images in pipes using GAN |
| title_fullStr | Anomaly detection from images in pipes using GAN |
| title_full_unstemmed | Anomaly detection from images in pipes using GAN |
| title_short | Anomaly detection from images in pipes using GAN |
| title_sort | anomaly detection from images in pipes using gan |
| topic | Infrastructure inspection Sewer pipe Deep learning GAN Anomaly detection |
| url | https://doi.org/10.1186/s40648-023-00246-y |
| work_keys_str_mv | AT shigekiyumoto anomalydetectionfromimagesinpipesusinggan AT takumikitsukawa anomalydetectionfromimagesinpipesusinggan AT alessandromoro anomalydetectionfromimagesinpipesusinggan AT sarthakpathak anomalydetectionfromimagesinpipesusinggan AT taronakamura anomalydetectionfromimagesinpipesusinggan AT kazunoriumeda anomalydetectionfromimagesinpipesusinggan |