Journal of Convergence for Information Technology (융합정보논문지)

Convergence Society for SMB (중소기업융합학회)
권호 표지
DOI QR코드

DOI QR Code

AutoML and CNN-based Soft-voting Ensemble Classification Model For Road Traffic Emerging Risk Detection

도로교통 이머징 리스크 탐지를 위한 AutoML과 CNN 기반 소프트 보팅 앙상블 분류 모델

  • Jeon, Byeong-Uk (Division of AI Computer Science and Engineering, Kyonggi University) ;
  • Kang, Ji-Soo (Department of Computer Science, Kyonggi University) ;
  • Chung, Kyungyong (Division of AI Computer Science and Engineering, Kyonggi University)
  • 전병욱 (경기대학교 AI컴퓨터공학부) ;
  • 강지수 (경기대학교 컴퓨터과학과) ;
  • 정경용 (경기대학교 AI컴퓨터공학부)
  • Received : 2021.05.15
  • Accepted : 2021.07.20
  • Published : 2021.07.28
Download PDF
⟨ Previous Next ⟩

Abstract

Most accidents caused by road icing in winter lead to major accidents. Because it is difficult for the driver to detect the road icing in advance. In this work, we study how to accurately detect road traffic emerging risk using AutoML and CNN's ensemble model that use both structured and unstructured data. We train CNN-based road traffic emerging risk classification model using images that are unstructured data and AutoML-based road traffic emerging risk classification model using weather data that is structured data, respectively. After that the ensemble model is designed to complement the CNN-based classification model by inputting probability values derived from of each models. Through this, improves road traffic emerging risk classification performance and alerts drivers more accurately and quickly to enable safe driving.

겨울철 도로 결빙으로 인한 사고는 대부분 큰 사고로 이어진다. 이는 운전자가 도로의 결빙을 사전에 자각하기 어렵기 때문이다. 본 연구에서는 AutoML과 CNN의 앙상블 모델을 이용하여 도로교통 이머징 리스크를 정확하게 탐지하는 방법을 연구한다. 비정형 데이터인 이미지를 이용한 CNN 이미지 특징 추출 기반 도로교통 이머징 리스크 분류 모델과 정형 데이터인 기상 데이터를 이용한 AutoML 기반 도로교통 이머징 리스크 분류 모델을 각각 학습시킨다. 그 후 모델들에서 도출된 확률값을 입력하여 CNN 기반 분류 모델을 보완하도록 앙상블 모델을 설계한다. 이를 통해 도로교통 이머징 리스크 분류 성능을 향상하고 더 정확하고 빠르게 운전자에게 경고하여 안전한 주행이 가능하도록 한다.

Keywords

Acknowledgement

This work is supported by the Korea Agency for Infrastructure Technology Advancement(KAIA) grant funded by the Ministry of Land, Infrastructure and Transport (Grant 21CTAP-C157011-02).

References

  1. S. J. Lee. (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. DOI : 10.14346/JKOSOS.2017.32.6.150
  2. G. Pan, L. Fu, R. Yu & M. Muresan. (2018). Winter road surface condition recognition using a pretrained deep convolutional network. arXiv preprint arXiv:1812.06858.
  3. M. Kangas, M. Heikinheimo & M. Hippi. (2015). RoadSurf: a modelling system for predicting road weather and road surface conditions. Meteorological applications, 22(3), 544-553. DOI : 10.1002/met.1486
  4. L. Zhang., F. Yang, Y. D. Zhang. & Y. J. Zhu. (2016, September). Road crack detection using deep convolutional neural network. 2016 IEEE international conference on image processing (ICIP), 3708-3712. DOI : 10.1109/ICIP.2016.7533052
  5. B. U. Jeon, J. S. Kang, H. J. Kwon & K. Chung, (2021). AutoML And CNN Image Feature Extraction for Road Icing Detection, 2021 Korean Society For Internet Information Spring Conference, 75-76.
  6. S. S. Park, J. W. Baek, S. M. Jo & K. Chung. (2019). Motion Monitoring using Mask R-CNN for Articulation Disease Management. Journal of the Korea Convergence Society, 10(3), 1-6. DOI : 10.15207/JKCS.2019.10.3.001
  7. F. Mohr, M. Wever. & E. Hullermeier. (2018). ML-Plan: Automated machine learning via hierarchical planning. Machine Learning, 107(8), 1495-1515. DOI : 10.1007/s10994-018-5735-z
  8. L. Zimmer. M. Lindauer & F. Hutter. (2021). Auto-Pytorch: Multi-Fidelity MetaLearning for Efficient and Robust AutoDL. IEEE Transactions on Pattern Analysis and Machine Intelligence. DOI : 10.1109/TPAMI.2021.3067763.
  9. O. Sagi & L. Rokach. (2018). Ensemble learning: A survey. Wiley Interdisciplinary Reviews: Data Mining and Knowledge Discovery, 8(4), e1249. DOI : 10.1002/widm.1249
  10. J. Cao, S. Kwong, R. Wang, X. Li, K. Li & X. Kong. (2015). Class-specific soft voting based multiple extreme learning machines ensemble. Neurocomputing, 149, 275-284. DOI : 10.1016/j.neucom.2014.02.072
  11. Korea Meteorological Administration. (n.d.). Weather Data Service - ASOS. (Online). https://data.kma.go.kr (accessed on 18 Mar. 2021)
  12. Traffic Accident Analysis System. (n.d.). Geographic Information System. (Online). http://taas.koroad.or.kr (accessed on 18 Mar. 2021)
  13. K. Shim., M., Lee., I. Choi., Y. Boo, & W. Sung. (2017, December). Svd-softmax: Fast softmax approximation on large vocabulary neural networks. Proceedings of the 31st International Conference on Neural Information Processing Systems, 5469-5479.
  14. H. J. Kim, J. W. Baek, & K. Chung. (2021). Associative Knowledge Graph using Fuzzy Clustering and Min-Max Normalization in Video Contents. IEEE Access, 9(1), 74802-74816. DOI : 10.1109/ACCESS.2021.3080180
  15. J. W. Baek, & K. Chung. (2021) Multi-Level Health Knowledge Mining Process in P2P Edge Network. IEEE Access, 9(1), 61623-61634. DOI : 10.1109/ACCESS.2021.3073775
  16. H. Yoo, S. Han, & K. Chung. (2021) Diagnosis Support Model of Cardiomegaly based on CNN using Resnet and Explainable Feature Map. IEEE Access, 9(1), 55802-55813. DOI : 10.1109/ACCESS.2021.3068597