Development of a Forecasting Model for Traffic Accident Probability on Icy Roads Using Deep Learning

Younggon Kim; Minwoo Lee; Yeojeong Yun; Yongjoo Jun; KwangSik Kim

doi:10.7470/jkst.2022.40.1.111

All Issue

2022 Vol.40, Issue 1 Preview Page Next Page

Article

Development of a Forecasting Model for Traffic Accident Probability on Icy Roads Using Deep Learning 딥러닝을 이용한 결빙 노면 상의 교통사고발생 예측 모형 개발

28 February 2022. pp. 111-127

PDF XML

Abstract

The number of traffic accidents caused by wintertime icy roads is increasing every year. Vehicle collisions on such frozen roads can cause fatalities, serious injuries, and property damage. So, comprehensive countermeasures need to be adopted. This requires accurate prediction and identification of hazardous areas. In this study, a forecasting model was developed to predict the chances of traffic accidents that can occur on icy roads. Deep learning algorithms were employed. Orthophotos and traffic accident data were used for training purposes. Historical data on accidents that occurred in the areas of Jongno-gu, Jung-gu, Mapo-gu, Yongsan-gu, Seodaemun-gu, and Seongbuk-gu in Seoul were collected for the period from year 2013 to year 2019. The orthophotos images were pre-processed before using them as input data for forecasting models. The procedure involved image segmentation techniques as well as Z-Curve index. Results show that there are significantly high probabilities of traffic accidents to occur on icy roads in the following geographical areas: urban parts of Gwanghwamun and Hoehyeon Station-Bank of Korea with particularly large volume of vehicular traffic, bridge areas adjacent to waterfronts like Olympic-daero-Yeongdong Bridge and Nonhyeon-dong-Hangang Bridge.

Keywords

deep learning

forecasting model

icy road

orthophotos

traffic accident probability

겨울철 도로결빙으로 인해 교통사고 발생건수가 매년 증가하는 추세를 보이고 있다. 결빙된 도로에서의 차량 충돌은 심각한 인적, 물적 사고를 야기할 수 있기 때문에 이에 대한 종합적 대책 마련이 필요하다. 본 연구에서는 겨울철 교통사고 발생 피해를 줄이기 위해 결빙 노면에서 발생하는 교통사고를 예측하는 딥러닝 기반의 예측모형을 개발하였다. 모형개발에는 정사영상 이미지와 교통사고 자료가 사용되었으며 2013년에서 2019년까지 서울의 종로구, 중구, 마포구, 용산구, 서대문구, 성북구 등의 지역에서 수집된 자료를 기반으로 하였다. 정사영상 이미지는 Z-Curve Index를 이용한 정사영상 이미지 분할 기법을 적용한 전처리 과정을 거쳐 예측 모형의 자료로 활용되었다. 본 연구 결과, 유동인구와 교통량이 많은 광화문 일대, 회현역-한국은행 구간 등 도심지역과 한강 등 수변지역에 인접한 올림픽대로-영동대교 구간, 논현동-한강대교 등 교량 지역이 교통사고 발생 가능성이 높은 지역으로 분석되었다.

키워드

딥러닝

예측모형

결빙 노면

정사영상

교통사고 발생 가능성

References

Chea J. Y., Kim G. M., Han B. J., Park D. S., Koo K. W. (2014), Road Ice Protection System Using Carbon Thread, The Korean Institute of Electrical Engineers, 1522-1523.

Choe Y. J., Baek C. S., Yom J. H. (2017), Machine Learning Based Accuracy Improvement of Land Surface Temperature from Satellite Imagery: With Application to Cloud Cover Enhancement, Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography, 2017(4), 280-286.

Daegu District Prosecutors’ Office Uiseong District Office (2020), Result of the Investigation into the Sangju Yeongcheon Expressway Multiple Collisions Case.

Hippi M., Juga I., Nurmi P. (2009, September), A Statistical Model for Road Surface Friction Forecasting Applying Optical Road Weather Measurements, In 9th EMS Annual Meeting, EMS2009-378.

Hong J. H. (1994), Qualitative Distance and Direction Reasoning in Geographic Space, Doctoral dissertation, The University of Maine.

Jung H. C., Kim J. E. (2020), Analysis on Factors Affecting Traffic Accident Severity, Journal of the Korea Institute of Information and Communication Engineering, 1162-1164.

Kim D. Y., Jung S. W. (2020), A Study on the Application of Spatial Clustering Technique for Crime Prediction, Proceedings of Korean Society for Geospatial Information Science, 111-114.

Kim J. U., Kim H. G. (2010), Introduction of Prevention and Prediction Techniques of Road Surface Freezing, 10(4), 35-39.

Kim J., Lee H., Paik J. (2020), Survey of Distinction of Black Ice Using Sensors, In Proceedings of the Korean Society of Computer Information Conference, Korean Society of Computer Information, 79-81.

Kim Y., Park J., Baik N. (2012), Research for Generating Road Surface Condition Information with Road Eye Laser Sensor, Korean Society of Civil Engineers Conference, 1522-1525.

Ko H. G., Lee J. S., Kim J. H. (2012), A Study on Provision of Real-Time Safety Information Considering Real-Time Vehicular Data and Road Traffic Condition, Journal of Civil and Environmental Engineering Research, 32(4D), Korean Society of Civil Engeneers, 291-303. 10.12652/Ksce.2012.32.4D.291

Koperski K., Han J., Stefanovic N. (1998), An Efficient Two-Step Method for Classification of Spatial Data, In Proc. Int’l. Symp. on Spatial Data Handling, SDH, 45-54.

Koroad, http://taas.koroad.or.kr/sta/acs/exs/typical.do?menuId=WEB_KMP_OVT_UAS_ASA#, 2020.11.11.

Kwak G. H., Park N. W., Lee K. D., Choi K. Y. (2017), Crop Classification for Inaccessible Areas using Semi-Supervised Learning and Spatial Similarity: A Case Study in the Daehongdan Region, North Korea, Korean Journal of Remote Sensing, 33(5_2), 689-698.

Lee J. Y., Jung J. H., Son B. S. (2008), Analysis of Traffic Accident Severity for Korean Highway Using Structural Equations Model, J. Korean Soc. Transp., 26(2), Korean Society of Transportation, 17-24.

Lee K. B., Lee S. W. (2007), Effects of Deicer on Concrete Pavement Distress, Journal of Civil and Environmental Engineering Research, 27(1D), Korean Society of Civil Engeneers, 73-80.

Lee M., Ryu Y., Jun Y. (2020), The Spatial Estimation of Road Surface Condition using Spatiotemporal Features, In 2020 IEEE 91st Vehicular Technology Conference VTC2020-Spring, 1-5. 10.1109/VTC2020-Spring48590.2020.9128799

Lee S. J. (2017), A Study on Factors that Influence Traffic Accident Severity in Road Surface Freezing, Journal of the Korean Society of Safety, 32(6), 150-156.

Liu S., Tian G., Xu Y. (2019), A Novel Scene Classification Model Combining ResNet Based Transfer Learning and Data Augmentation with a Filter, Neurocomputing, 338, 191-206. 10.1016/j.neucom.2019.01.090

McLaughlin P. W., Evans C., Feng M., Narayana V. (2010), Radiographic and Anatomic Basis for Prostate Contouring Errors and Methods to Improve Prostate Contouring Accuracy, International Journal of Radiation Oncology* Biology* Physics, 76(2), 369-378. 10.1016/j.ijrobp.2009.02.01919515511

Nair V., Hinton G. E. (2010), Rectified Linear Units Improve Restricted Boltzmann Machines, In Icml.

Park J. H., Han S. H. Shin S. (2020), Black Ice Detection and Accident Prevention System Using Thermal Imaging Camera and Drone, Proceedings of Symposium of the Korean Institute of Communications and Information Sciences, 893-894.

Perez L., Wang J. (2017), The Effectiveness of Data Augmentation in Image Classification Using Deep Learning, arXiv preprint arXiv:1712.04621.

Qiu L., Nixon W. A. (2008), Effects of Adverse Weather on Traffic Crashes: Systematic Review and Meta-analysis, Transportation Research Record, 2055(1), 139-146. 10.3141/2055-16

Sander J., Ester M., Kriegel H. P., Xu X. (1998), Density-based Clustering in Spatial Databases: The Algorithm Gdbscan and Its Applications, Data Mining and Knowledge Discovery, 2(2), 169-194. 10.1023/A:1009745219419

Shorten C., Khoshgoftaar T. M. (2019), A Survey on Image Data Augmentation for Deep Learning, Journal of Big Data, 6(1), 1-48. 10.1186/s40537-019-0197-0

Srivastava N., Hinton G., Krizhevsky A., Sutskever I., Salakhutdinov R. (2014), Dropout: A Simple Way to Prevent Neural Networks from Overfitting, The Journal of Machine Learning Research, 15(1), 1929-1958.

Tensor Flow Core v2.1.0, https://www.tensorflow.org/versions/r2.1/api_docs/python/tf?authuser=3&hl=es-AR, 2020.10.01.

Wåhlin J., Baad H., Roads D. P., Vaa T., Roads N. P., Jonsson P. (2016), Laboratory Test of Five Different Optical Road Condition Sensors, Proceedings of the SIRWEC.

Yan H., Li J. (2015), Spatial Similarity Relations in Multi-scale Map Spaces, Heidelberg: Springer. 10.1007/978-3-319-09743-5

Information

Publisher :Korean Society of Transportation
Publisher(Ko) :대한교통학회
Journal Title :Journal of Korean Society of Transportation
Journal Title(Ko) :대한교통학회지
Volume : 40
No :1
Pages :111-127
Received Date : 2021-11-03
Revised Date : 2021-11-19
Accepted Date : 2021-11-24
DOI :https://doi.org/10.7470/jkst.2022.40.1.111

Journal of Korean Society of Transportation ISSN:1229-1366(Print) 2234-4217(Online) 대한교통학회지

All Issue