상하수도학회지 (Journal of Korean Society of Water and Wastewater)

  • 제38권2호
  • /
  • Pages.83-93
  • /
  • 2024
  • /
  • 1225-7672(pISSN)
  • /
  • 2287-822X(eISSN)
대한상하수도학회 (The Korean Society of Water and Wastewater)
권호 표지
DOI QR코드

DOI QR Code

관로 조사를 위한 오토 인코더 기반 이상 탐지기법에 관한 연구

A study on the auto encoder-based anomaly detection technique for pipeline inspection

  • 김관태 (주식회사 키네틱스 기술연구소) ;
  • 이준원 (주식회사 키네틱스 기술연구소)
  • 투고 : 2023.12.27
  • 심사 : 2024.03.19
  • 발행 : 2024.04.15
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

In this study, we present a sewer pipe inspection technique through a combination of active sonar technology and deep learning algorithms. It is difficult to inspect pipes containing water using conventional CCTV inspection methods, and there are various limitations, so a new approach is needed. In this paper, we introduce a inspection method using active sonar, and apply an auto encoder deep learning model to process sonar data to distinguish between normal and abnormal pipelines. This model underwent training on sonar data from a controlled environment under the assumption of normal pipeline conditions and utilized anomaly detection techniques to identify deviations from established standards. This approach presents a new perspective in pipeline inspection, promising to reduce the time and resources required for sewer system management and to enhance the reliability of pipeline inspections.

키워드

과제정보

본 연구는 국토교통부/국토교통과학기술진흥원의 지원으로 수행되었음(과제번호 RS-2022-00144031).

참고문헌

  1. Chae, M.B., Bae, Y.H. and Kim, H.S. (2018). Cause Analysis for Reduced Effect of SewerPipe Improvement Project Based On Investigation of Interceptor Sewers, J. Wetl. Res., 20, 219.
  2. Joo, G.H., Park, C.H. and Im, H.S. (2020). Performance Evaluation of Machine Learning Optimizers, J. Inst. Korean Electr. Eectron., 24, 775.
  3. Ji, H.W., Yoo, S.S. and Kang, J.H. (2021). Extraction of sewer pipe relative elevation from CCTV investigation video using deep learning, J. Korea Acad. Coop. Soc., 1085.
  4. Kang, K.H. (2020). Network Anomaly Detection Technologies Using Unsupervised Learning Auto Encoders, Korea Inst. Inf. Secur. Cryptol., 30, 620.
  5. Kang, H.G., Seo, D.S., Lee, B.S. and Kang, M.S. (2017). Applying CEE(Cross Entropy Error) to improve performance of Q-Learning algorithm, Korea Artif. Intell. Assoc., 5, 5.
  6. Kim, H.J. (2020). Unsupervised method of rising keyword and important sentence extraction within a specific period time, master degree, Yonsei University.
  7. Kim, I.S., Lee, M.G. and Jeon, Y.H. (2021). Comparative Analysis of Defect Detection Using YOLO of Deep Learning, J. Korean Soc. Manuf. Technol. Eng., 513, 514.
  8. Kwon, S.H. Wikidocs, https://wikidocs.net/28147 (March 07, 2024)
  9. Lee, H.J. and Jung, Y.S. (2021). Comparison of deep learning-based autoencoders for recommender systems, Korean J. Appl. Stat., 34, 331.
  10. Lee, J.H. and Sohn, J.M. (2021). Escalator Anomaly Detection Using LSTM autoencoder, J. Korea Soc. Comput. Inform., 29, 7.
  11. Lee, J.W. and Kim, K.S. (2023). Intrusion Detection Method Using Unsupervised Learning-Based Embedding and Autoencoder, KIPS Trans. Softw. Data Eng., 12, 356.
  12. Lin, S.B. (2022). Determination of coagulant dose using deep learning-based forecasting model with 9 years of field operation data, master degree, KAIST.
  13. NHN Cloud Meetup, https://meetup.nhncloud.com/posts/362 (November 11, 2023).
  14. Saitoh, Koki (2017). Deep Learning from the Basics: Python and Deep Learning: Theory and Implementation. Packt Publishing, Birmingham.
  15. Sin, D.H. (2020). A Comparative study on Deep Neural network Active function, master degree, Hoseo University.
  16. Srivastava Nitish, Hinton Geoffrey, Krizhevsky Alex, SutsKever Ilya and Salakhutdinov Ruslan. (2014). Dropout: A Simply Way to Prevent Neural Networks from Overfitting, J. Mach. Learn. Res., 15, 1930.