정보처리학회논문지:소프트웨어 및 데이터공학 (KIPS Transactions on Software and Data Engineering)

  • 제10권11호
  • /
  • Pages.439-448
  • /
  • 2021
  • /
  • 2287-5905(pISSN)
  • /
  • 2734-0503(eISSN)
한국정보처리학회 (Korea Information Processing Society)
권호 표지
DOI QR코드

DOI QR Code

CAM 기반의 계층적 및 수평적 분류 모델을 결합한 운전자 부주의 검출 및 특징 영역 지역화

Distracted Driver Detection and Characteristic Area Localization by Combining CAM-Based Hierarchical and Horizontal Classification Models

  • 고수연 (숙명여자대학교 컴퓨터과학과) ;
  • 최영우 (숙명여자대학교 컴퓨터과학과)
  • 투고 : 2021.08.27
  • 심사 : 2021.10.12
  • 발행 : 2021.11.30
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

교통사고 원인 중 가장 큰 비율을 차지하는 것이 운전자의 부주의로서 이를 검출하는 연구가 꾸준히 진행되고 있다. 본 논문은 부주의한 운전자를 정확히 검출하고, 검출된 운전자의 모습에서 가장 특징적인 영역을 선정(Localize)하는 방법을 제안한다. 제안하는 방법은 운전자의 부주의를 검출하기 위해서 CAM(Class Activation Map) 기반의 전체 클래스를 분류하는 CNN 모델과 이 모델에서 혼동하거나 공통된 특징 영역을 갖는 클래스들에 대한 상세 분류가 가능한 네 개의 서브 클래스 CNN 모델을 계층적으로 구성한다. 각 모델에서 출력한 분류 결과는 CNN 특징맵들과의 매칭 정도를 표현하는 새로운 특징으로 간주해서 수평적으로 결합하고 학습하여 분류의 정확성을 높였다. 또한 전체 및 상세 분류 모델의 분류 결과를 반영한 히트맵 결과를 결합하여 이미지의 특징적인 주의 영역을 찾아낸다. 제안한 방법은 State Farm 데이터 셋을 이용한 실험에서 95.14%의 정확도를 얻었으며, 이는 기존에 동일한 데이터 셋을 이용한 결과 중 가장 높은 정확도인 92.2%보다 2.94% 향상된 우수한 결과이다. 또한 전체 모델만을 이용했을 때 찾아진 주의 영역보다 훨씬 의미 있고 정확한 주의 영역이 찾아짐을 실험으로 확인하였다.

Driver negligence accounts for the largest proportion of the causes of traffic accidents, and research to detect them is continuously being conducted. This paper proposes a method to accurately detect a distracted driver and localize the most characteristic parts of the driver. The proposed method hierarchically constructs a CNN basic model that classifies 10 classes based on CAM in order to detect driver distration and 4 subclass models for detailed classification of classes having a confusing or common feature area in this model. The classification result output from each model can be considered as a new feature indicating the degree of matching with the CNN feature maps, and the accuracy of classification is improved by horizontally combining and learning them. In addition, by combining the heat map results reflecting the classification results of the basic and detailed classification models, the characteristic areas of attention in the image are found. The proposed method obtained an accuracy of 95.14% in an experiment using the State Farm data set, which is 2.94% higher than the 92.2%, which is the highest accuracy among the results using this data set. Also, it was confirmed by the experiment that more meaningful and accurate attention areas were found than the results of the attention area found when only the basic model was used.

키워드

과제정보

이 논문은 한국연구재단 기초연구과제에 의하여 연구되었음(No. NRF-2017R1D1A1B04035633).

참고문헌

  1. World Health Organization, Global Status Report on Road Safety 2018: Summary, World Health Organization, Geneva, Switzerland, 2018.
  2. National Highway Traffic Safety Administration, "2015 motor vehicle crashes: Overview," Traffic safety facts: research note, U.S. Department of Transportation, August, 2016.
  3. C. H. Zhao, B. L. Zhang, J. He, and J. Lian, "Recognition of driving postures by contourlet transform and random forests," Intelligent Transport Systems, Vol.6, pp.161-168, 2012. https://doi.org/10.1049/iet-its.2011.0116
  4. I. Sultan. Academic purposes, [Internet], https://www.kaggle.com/c/ state-farm-distracted-driver-detection/discussion/20043#117982. 2016.
  5. H. M. Eraqi, Y. Abouelnaga, M. H. Saad, and M. N. Moustafa, "Driver distraction identification with an ensemble of convolutional neural networks," Journal of Advanced Transportation, Vol.2019, 2019.
  6. B. Zhou, A. Khosla, A. Lapedriza, A. Oliva, and A. Torralba, "Learning deep features for discriminative localization," 2016 IEEE Conference on Computer Vision and Pattern Recognition, pp.2921-2929, 2016.
  7. M. Alotaibi and B. Alotaibi, "Distracted driver classification using deep learning," Signal Image Video Process, pp.617-624, 2019.
  8. M. Q. Lu, Y. C. Hu, and X. B. Lu, "Driveraction recognition using deformable and dilated fasterR-CNN with optimized region proposals," Applied Intelligence, Vol.50, pp.1100-1111, 2020. https://doi.org/10.1007/s10489-019-01603-4
  9. L. C. Valeriano, P. Napoletano, and R. Schettini, "Recognition of driver distractions using deep learning," In Proceedings of the 2018 IEEE 8th International Conferenceon Consumer Electronics, pp.1-6, 2018.
  10. N. Moslemi, R. Azmi, and M. Soryani, "Driver distraction recognition using 3D convolutional neural networks," In Proceedings of the 2019 4th International Conferenceon Pattern Recognitionand Image Analysis, pp.145-151, 2019.
  11. K. He, X. Zhang, S. Ren, and J. Sun, "Deep residual learning for image recognition," In Proceedings of the IEEE conference on Computer Vision and Pattern Recognition, Vol.1, pp.770-778, 2016.
  12. J. Chung, S. Ahn, and Y. Bengio, "Hierarchical multiscale recurrent neural networks," IInternational Conference on Learning Representations, 2017.
  13. C. Szegedy, W. Liu, Y. Jia, P. Sermanet, S. Reed, D. Anguelov, and A. Rabinovich, "Going deeper with convolutions," In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp.1-9, 2015.
  14. S. Masood, A. Rai, A. Aggarwal, and M. N. Doja, "Detecting distraction of drivers using Convolutional Neural Network," Pattern Recognition Letters, Vol.139, 2018.
  15. K. Simonyan and A. Zisserman. "Very deep convolutional networks for large-scale image recognition," International Conference on Learning Representations, 2015.
  16. S. Ren, K. He, R. Girshick, and J. Sun. "Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks," IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol.39, pp.1137-1149, 2015. https://doi.org/10.1109/TPAMI.2016.2577031
  17. R. R. Selvaraju, M. Cogswell, A. Das, R. Vedantam, D. Parikh, and D. Batra. "Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization," IEEE International Conference on Computer Vision, pp.618-626, 2017.
  18. K. K. Singh and Y. J. Lee, "Hide-and-seek: Forcing a network to be meticulous for weakly- supervised object and action localization," IEEE International Conference on Computer Vision, pp.3544-3553, 2017.
  19. S. Yun, D. Han, S. J. Oh, S. Chun, J. Choe, and Y. Yoo, "Cutmix: Regularization strategy to train strong classifiers with localizable features," IEEE/CVF International Conference on Computer Vision, pp.6022-6031, 2019.
  20. X. Zhang, Y. Wei, J. Feng, Y. Yang, and T. Huang, "Adversarial complementary learning for weakly supervised object localization," in IEEE Computer Vision and Pattern Recognition, pp.1325-1334, 2018.
  21. J. Deng, W. Dong, R. Socher, L.-J. Li, K. Li, and L. Fei-Fei, "Imagenet: A large-scale hierarchical image database," Conference on Computer Vision and Pattern Recognition, pp. 248-255, 2009.