All Issue

2017 Vol.19, Issue 1 Preview Page
A preliminary study for development of an automatic incident detection system on CCTV in tunnels based on a machine learning algorithm

기계학습(machine learning) 기반 터널 영상유고 자동 감지 시스템 개발을 위한 사전검토 연구

Hyu-Soung Shin1
,
Dong-Gyou Kim2*
,
Min-Jin Yim3
,
Kyu-Beom Lee4
,
Young-Sup Oh5
신 휴성1
,
김 동규2*
,
임 민진3
,
이 규범4
,
오 영섭5
1Head, Extreme Construction Research Center, Korea Institute of Civil Engineering and Building Technology
2Research Fellow, Geotechnical Engineering Research Institute, Korea Institute of Civil Engineering and Building Technology
3Geotechnical Engineering Research Institute, Korea Institute of Civil Engineering and Building Technology
4PhD Student, Geotechnical Engineering Research Institute, Korea Institute of Civil Engineering and Building Technology
5Head, Research & Development Department, SB Networks Ltd
1한국건설기술연구원 극한건설연구단 단장
2한국건설기술연구원 지반연구소 연구위원
3한국건설기술연구원 지반연구소 연구원
4한국건설기술연구원 지반연구소 학생연구원
5(주)SB네트워크, 연구개발실장
*교신저자. *Corresponding Author. 
31 January 2017. pp. 95-107
PDF XML Citation
Abstract

In this study, a preliminary study was undertaken for development of a tunnel incident automatic detection system based on a machine learning algorithm which is to detect a number of incidents taking place in tunnel in real time and also to be able to identify the type of incident. Two road sites where CCTVs are operating have been selected and a part of CCTV images are treated to produce sets of training data. The data sets are composed of position and time information of moving objects on CCTV screen which are extracted by initially detecting and tracking of incoming objects into CCTV screen by using a conventional image processing technique available in this study. And the data sets are matched with 6 categories of events such as lane change, stoping, etc which are also involved in the training data sets. The training data are learnt by a resilience neural network where two hidden layers are applied and 9 architectural models are set up for parametric studies, from which the architectural model, 300(first hidden layer)-150(second hidden layer) is found to be optimum in highest accuracy with respect to training data as well as testing data not used for training. From this study, it was shown that the highly variable and complex traffic and incident features could be well identified without any definition of feature regulation by using a concept of machine learning. In addition, detection capability and accuracy of the machine learning based system will be automatically enhanced as much as big data of CCTV images in tunnel becomes rich.

Keywords
Tunnel Incident Automatic Detection
Tunnel CCTV
유Machine Learning지관리
Image Processing

본 논문에서는 제도적으로 운영 중인 터널내 CCTV들로부터 실시간으로 들어오는 영상들을 최신 딥러닝 알고리즘을 이용, 학습시켜 다양한 조건의 터널환경에서 돌발 상황을 감지하고 그 돌발 상황의 종류들을 분류해 내는 시스템 개발을 위한 사전검토 연구를 수행하였다. 사전검토 연구를 위해, 2개의 도로현장의 교통류 CCTV영상 일부를 이용하여 가용한 전통적인 영상처리기법으로 영상내부로 집입하는 차량을 감지하고, 이동경로를 추적하여 일정 시간간격의 이동 차량의 좌표와 시간정보를 추출하고 학습자료를 구성하였다. 각 차량의 이동정보는 차선변경, 정차 등 6가지의 이벤트 정보와 연계된다. 차량 이동정보와 이벤트로 구성된 학습자료는 레질리언스(resilience) 기계학습 알고리즘을 이용하여 학습하였다. 2개의 은닉층을 설정하고, 각 은닉층의 노드수에 대한 9개의 은닉구조 모델을 설정하여 매개변수 연구를 수행하였다. 본 사전검토의 경우에는 첫 번째, 두 번째 은닉층 노드수가 각각 300개와 150개로 설정된 모델이 합리적으로 가장 추론정확도가 높은 것으로 평가되었다. 이로부터 일반화되기 매우 힘든 복잡한 교통류 상황을 기계학습을 이용하여 어떠한 사전 규칙설정 없이도 교통류의 특징들을 정확히 자동으로 감지할 수 있는 가능성을 보였다. 본 시스템은 시스템의 운용을 통해 지속적으로 교통류 영상과 이벤트 정보가 늘어난다면, 자동으로 그 시스템의 인지능력과 정확도가 자동으로 향상되는 효과도 기대할 수 있다.

키워드
터널 영상유고 자동 감지
터널 CCTV
기계학습
영상처리
References
1
1.Choi, J.M., Kwon, J.O. (2010), “Converged Security Market Trend Report”, Samsung SDS, Samsung SDS Journal of IT Services, Vol. 7, No. 2, pp. 13-29.
2
2.Karpathy, A., Toderici, G., Shetty, S., Leung, T., Sukthankar, R., Fei-Fei, L. (2014). “Large-scale video classification with convolutional neural networks”, Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. pp. 1725-1732.
3
3.Kim, D.G., Shin, Y.W., Shin, Y.S. (2012) “Section enlargement by reinforcement of shotcrete lining on the side wall of operating road tunnel”, Journal of Korean Tunneling and Underground Space Association, Vol. 14, No. 6, pp. 637-652.
4
4.Kim, T.B. (2016), “The National Highway, Expressway Tunnel Video Incident Detection System performance analysis and reflect attributes for double deck tunnel in great depth underground space”, J. Korea Inst. Inf. Commun. Eng., Vol. 20, No. 7, pp. 1325-1334.
5
5.Ministry of Land, Infrastructure and Transport (MOLIT) (2015), “Study on efficient ways of tunnel maintenance”, Technical Report by Korea Institute of Civil Engineering and Building Technology.
6
6.Ministry of Land, Infrastructure and Transport (MOLIT) (2016). “Attemp for faultless safety system of road tunnels”, Press Release.
7
7.Ministry of Land, Infrastructure and Transport (MOLIT) (2016), “Guideline of installation of disaster prevention facilities on road tunnels”,
8
8.National Committee for Land and Transport (2016), “Tunnel accidents increase, but tunnel insident automatic detection system often fails in operation”, Press Release from parliamentary inspection of MOLIT. Congressman Yoon Hoo-Dyuk.
9
9.Park, J.K., Park, Y.K., On, H.I., Kang, D.J. (2015), “Object Perception Methods in Image Using Deep Learning”, Institute of Control, Robotics and Systems, Journal of Institute of Control, Robotics and Systems, Vol. 21, No. 4, pp. 21-26, 2015.
10
10.Prechelt, L. (1998), “Automatic early stopping using cross validation: quantifying the criteria”, Neural Networks, Vol. 11, pp. 761-767.
11
11.Le, Q., Ranzato, M., Monga, R., Devin, M., Chen, K., Corrado, G., Dean, J., Ng, A. (2012), “Building high-level features using large scale unsupervised learning”, International Conference on Machine Learning.
12
12.Roh, C.G., Park, B.J., Kim, J.S. (2016), “A Study on the Contents for Operation of Tunnel Management Systems Using a View Synthesis Technology”, The Korea Contents Society, Journal of the Korea Contents Association, Vol. 16, No. 6, pp. 507-515.
13
13.Shin, H.S. (2001), “Neural network based constitutive models for finite element analysis”, Ph.D. thesis, University of Wales Swansea, C/Ph/250/01.
14
14.Shin, H.S., Kwon, Y.C. (2009a), “Development of a window-shifting ANN training method for a quantitative rock classification in un-sampled rock zone”, Journal of Korean Tunneling and Underground Space Association, Vol. 11, No. 2, pp. 151-162.
15
15.Shin, H.S., Kwon, Y.C., Kim, D.G., Bae, G.J., Lee, H.G., Shin, Y.W. (2009b), “ Quantitative preliminary hazard level simulation for tunnel design based on the KICT tunnel collapse hazard index (KTH-index)”, Journal of Korean Tunneling and Underground Space Association, Vol. 11, No. 4, pp. 373-385.
16
16.Skouras, K., Goutis, C., Bramson, M.J. (1994), “Estimation in linear-models using gradient descent with early stopping”, Statistics and Computing, Vol. 4, No. 4, pp. 271-278.
Information
  • Publisher :Korean Tunneling and Underground Space Association
  • Publisher(Ko) :한국터널지하공간학회
  • Journal Title :Journal of Korean Tunnelling and Underground Space Association
  • Journal Title(Ko) :한국터널지하공간학회 논문집
  • Volume : 19
  • No :1
  • Pages :95-107
  • Received Date : 2017-01-19
  • Revised Date : 2017-01-26
  • Accepted Date : 2017-01-31
  • DOI :https://doi.org/10.9711/KTAJ.2017.19.1.095
Journal Informaiton Journal of Korean Tunnelling and Underground Space Association Journal of Korean Tunnelling and Underground Space Association
Online Submission & Review System
  • NRF
  • KOFST
  • KISTI Current Status
  • KISTI Cited-by
  • crosscheck
  • orcid
  • open access
  • ccl
Journal Informaiton Journal Informaiton - close