Deep Learning-based Object Detection of Panels Door Open in Underground Utility Tunnel

doi:10.15683/kosdi.2023.9.30.665

All Issue

2023 Vol.19, Issue 3 Preview Page Next Page

Original Article

Deep Learning-based Object Detection of Panels Door Open in Underground Utility Tunnel 딥러닝 기반 지하공동구 제어반 문열림 인식: 김경환¹ㆍ김지은²*ㆍ정우석³
Gyunghwan Kim¹, Jieun Kim²*, Woosug Jung³; ¹Student Researcher, Electronics and Telecommunications Research Institute, Daejeon, Republic of Korea
²Principal Researcher, Electronics and Telecommunications Research Institute, Daejeon, Republic of Korea
³Director of Disaster & Safety AI Convergence Center, Electronics and Telecommunications Research Institute, Daejeon, Republic of Korea

30 September 2023. pp. 665-672

PDF

Abstract

Purpose: Underground utility tunnel is facility that is jointly house infrastructure such as electricity, water and gas in city, causing condensation problems due to lack of airflow. This paper aims to prevent electricity leakage fires caused by condensation by detecting whether the control panel door in the underground utility tunnel is open using a deep learning model. Method: YOLO, a deep learning object recognition model, is trained to recognize the opening and closing of the control panel door using video data taken by a robot patrolling the underground utility tunnel. To improve the recognition rate, image augmentation is used. Result: Among the image enhancement techniques, we compared the performance of the YOLO model trained using mosaic with that of the YOLO model without mosaic, and found that the mosaic technique performed better. The mAP for all classes were 0.994, which is high evaluation result. Conclusion: It was able to detect the control panel even when there were lights off or other objects in the underground cavity. This allows you to effectively manage the underground utility tunnel and prevent disasters.

Keywords

Underground Utility Tunnel

Deep Learning

Panel

Object Detection

YOLO

연구목적: 지하공동구는 도시 지하에 전기, 수도, 가스 등의 인프라를 공동 수용하는 시설로 공기 흐름이 부족하여 계절에 상관없이 결로가 자주 발생한다. 결로는 전기 설비의 누전 화재를 일으키는 원인이 되므로 지하공동구 내의 조명 등 각종 시설물 관리를 위해 필요한 제어반은 결로에 노출되지 않도록 문이 닫힌 상태로 관리되어야 한다. 본 논문에서는 딥러닝 객체인식 기술을 활용하여 수km 거리에 반복 배치된 공동구 제어반의 문 열림 여부를 이동 카메라 조건과 조명이 꺼진 조건에서도 인식하고자 한다. 연구방법: 지하공동구를 순찰하는 로봇이 촬영한 영상데이터를 이용하여 딥러닝 객체인식 모델인 YOLO를 모자이크 이미지 증강기법으로 학습시켜 제어반 문 열림과 문 닫힘을 인식한다. 연구결과: 모자이크 이미지 증강기법으로 학습시킨 모델과 사용하지 않은 모델의 성능을 비교한 결과, 모자이크 학습 모델이 더 우수한 성능(모든 클래스에 대한 mAP가 0.994 이상임)을 보이는 것을 확인하였다. 결론: 지하공동구의 조명이 꺼진 상태에서도, 공동구 내부 시설물이 복잡한 환경에서도 제어반의 문열림 여부를 우수한 성능으로 인식하여 지하공동구 재난안전관리에 도움이 될 것으로 기대된다.

키워드

지하공동구

딥러닝

제어반

객체인식

YOLO

References

Jo, Y.-H., Lee, M.-H., Kim, D.-H., Lee, J.-K. (2020). "Recognizing transportation vulnerable using deep neural network based object detection." Proceedings of Symposium of the Korean Institute of communications and Information Sciences, Republic of Korea, pp.1402-1403.
Kim, H.-Y., Park, T.-Y.(2021). "Detection of wearing a protective gear and entering hazardous areas of worker in construction site using YOLO." Conference on Information and Control Systems 2021, Republic of Korea, pp. 458-459.
Kim, J.-S., Lee, C.-W., Park, S.-H., Lee, J.-H., Hong, C.-H. (2020). "Development of fire detection model for underground utility facilities using deep learning: Training data supplement and bias optimization." Journal of the Korea Academia-Industrial cooperation Society, Vol. 21, No. 12, pp. 320-330.
Lee, M.-S., Jung, W.-S., Kim, E.-S.(2021). "A study on the disaster safety management method of underground lifelines based on digital twin technology." Communications of the Korean Institute of Information Scientists and Engineers, Vol. 39, No. 2, pp. 16-24.
Lee, Y.-J. (2020). A Study on the Disaster Management System Considering the Risk Characteristics of Fire in Utility Tunnel. The Korean Society of Disaster Information, Republic of Korea, pp.71-72.
Nam, C.-H., Nam, E.-J., Jang, K.-S. (2021). "Mask wearing detection system using deep learning." Journal of the Korea Institute of Information and Communication Engineering, Vol. 25, No. 1, pp. 44-49.
Park, S.-M., Hong, C.-H., Park, S.-H., Lee, J.-W., Kim, J.-S. (2022). "Development of a deep learning-based fire extinguisher object detection model in underground utility tunnels." Journal of the Society of Disaster Information, Vol. 18, No. 4, pp. 922-929.
YOLOv5, https://github.com/ultralytics/yolov5
Yoon, D.-W., Seong, N.-C., Shin, D.-C.(2014). "A study on determining dew condensation at the underground utility tunnel as measurement thermal condition." Journal of The Korean Society of Living Environmental System, Vol. 21, No. 6, pp. 1014-1022. 10.21086/ksles.2014.12.21.6.1014

Information

Publisher :The Korean Society of Disaster Information
Publisher(Ko) :한국재난정보학회
Journal Title :Journal of the Society of Disaster Information
Journal Title(Ko) :한국재난정보학회논문집
Volume : 19
No :3
Pages :665-672
DOI :https://doi.org/10.15683/kosdi.2023.9.30.665

[1] Jo, Y.-H., Lee, M.-H., Kim, D.-H., Lee, J.-K. (2020). "Recognizing transportation vulnerable using deep neural network based object detection." Proceedings of Symposium of the Korean Institute of communications and Information Sciences, Republic of Korea, pp.1402-1403.

[2] Kim, H.-Y., Park, T.-Y.(2021). "Detection of wearing a protective gear and entering hazardous areas of worker in construction site using YOLO." Conference on Information and Control Systems 2021, Republic of Korea, pp. 458-459.

[3] Kim, J.-S., Lee, C.-W., Park, S.-H., Lee, J.-H., Hong, C.-H. (2020). "Development of fire detection model for underground utility facilities using deep learning: Training data supplement and bias optimization." Journal of the Korea Academia-Industrial cooperation Society, Vol. 21, No. 12, pp. 320-330.

[4] Lee, M.-S., Jung, W.-S., Kim, E.-S.(2021). "A study on the disaster safety management method of underground lifelines based on digital twin technology." Communications of the Korean Institute of Information Scientists and Engineers, Vol. 39, No. 2, pp. 16-24.

[5] Lee, Y.-J. (2020). A Study on the Disaster Management System Considering the Risk Characteristics of Fire in Utility Tunnel. The Korean Society of Disaster Information, Republic of Korea, pp.71-72.

[6] Nam, C.-H., Nam, E.-J., Jang, K.-S. (2021). "Mask wearing detection system using deep learning." Journal of the Korea Institute of Information and Communication Engineering, Vol. 25, No. 1, pp. 44-49.

[7] Park, S.-M., Hong, C.-H., Park, S.-H., Lee, J.-W., Kim, J.-S. (2022). "Development of a deep learning-based fire extinguisher object detection model in underground utility tunnels." Journal of the Society of Disaster Information, Vol. 18, No. 4, pp. 922-929.

[8] YOLOv5, https://github.com/ultralytics/yolov5

[9] Yoon, D.-W., Seong, N.-C., Shin, D.-C.(2014). "A study on determining dew condensation at the underground utility tunnel as measurement thermal condition." Journal of The Korean Society of Living Environmental System, Vol. 21, No. 6, pp. 1014-1022. 10.21086/ksles.2014.12.21.6.1014

Journal of the Society of Disaster Information ISSN:1976-2208(Print) 2671-5287(Online) 한국재난정보학회논문집

All Issue