Journal of Korea Society of Industrial Information Systems (한국산업정보학회논문지)

Korea Society of Industrial Information Systems (한국산업정보학회)
권호 표지
DOI QR코드

DOI QR Code

A study on the detection of pedestrians in crosswalks using multi-spectrum

다중스펙트럼을 이용한 횡단보도 보행자 검지에 관한 연구

  • Received : 2021.11.07
  • Accepted : 2021.12.20
  • Published : 2022.02.28
Download PDF
⟨ Previous Next ⟩

Abstract

The use of multi-spectral cameras is essential for day and night pedestrian detection. In this paper, a color camera and a thermal imaging infrared camera were used to detect pedestrians near a crosswalk for 24 hours at an intersection with a high risk of traffic accidents. For pedestrian detection, the YOLOv5 object detector was used, and the detection performance was improved by using color images and thermal images at the same time. The proposed system showed a high performance of 0.940 mAP in the day/night multi-spectral (color and thermal image) pedestrian dataset obtained from the actual crosswalk site.

주간 및 야간의 보행자 감지를 위해서는 다중 스펙트럼 활용이 필수적이다. 본 논문에서는 교통사고의 위험성이 높은 교차로에서 횡단보도 근처의 보행자를 24시간 검출하기 위해 컬러 카메라 및 열화상 적외선 카메라를 사용하였다. 보행자 탐지를 위해서 YOLO v5 객체 검출기를 사용하였으며 컬러 이미지와 열화상 이미지를 동시에 사용하여 감지 성능을 향상 시켰다. 제안된 시스템은 실제 횡단보도 현장에서 확보한 주·야간 다중 스펙트럼(색상 및 열화상) 보행자 데이터 셋에서 Iou 0.5 기준 0.94 mAP의 높은 성능을 보였다.

Keywords

Acknowledgement

이 논문은 2017학년도 대구대학교 학술연구비지원에 의한 논문임.

References

  1. Cho, N. H., Ann, J. D. and Gwak, Y. S. (2014). A Study on the Pedestrians Monitoring System on the Crossroads, Proceedings of KIIT Conference, pp.353-357.
  2. Girshick, R., Donahue, J. and Darrell T. (2016). Region based convolutional networks for accurate object detection and segmentation, IEEE transactions on pattern analysis and machine intelligence, 38(1), 142-158. https://doi.org/10.1109/TPAMI.2015.2437384
  3. Jung, G. Y., Jung, J. H. and Lee H. C. (2013). Efficient Implementation of Candidate Region Extractor for Pedestrian Detection System with Stereo Camera based on GP-GPU, IEMEK J. Embed. Sys. Appl., 8(2), 121-128. https://doi.org/10.14372/IEMEK.2013.8.2.121
  4. Jeon, N. Y., Kim, S. J. and Chu, S. H. (2018). A study on the Application of Living Lab in Transportation : Focused on the Auto-Image Sensing Signal System for Pedestrian, The Korea Institute of Intelligent Transport Systems, 17(2), 1-17. https://doi.org/10.12815/kits.2018.17.2.01
  5. Kim, J. H., Jung Y. M. (2021). Implementation of pedestrian safety system for crosswalks in the late-night time, J-KICS, pp.604-605.
  6. Ling, Bo., Zeifman, M.I., Gibson, D. R. P. (2007). Multiple pedestrian detection using IR LED stereo camera, Intelligent Robots and Computer Vision XXV: Algorithms.
  7. Lee, T. H., Kim, K. J., Yun, K. S. and Choi D. H. (2018). A method of Counting Vehicle and Pedestrian using Deep Learning based on CCTV, Journal of Korean Institute of Intelligent Systems, 28(3), 219-224. https://doi.org/10.5391/JKIIS.2018.28.3.219
  8. Lee, M. H., Nam, K. W. and Lee, C. W. (2019). Crack Detection on the Road in Aerial Image using Mask R-CNN, Journal of the Korea Industrial Information System Research, 24(3), 23-29, http://dx.doi.org./10.9723/jksiis.2019.24.3.023.
  9. Malisiewicz, T., Gupta, A. and Efros A. A. (2011). Ensemble of exemplarsvms for object detection and beyond, in Computer Vision (ICCV) 2011 IEEE International Conference on. IEEE, pp.89-96.
  10. Papageorgiou, C., Poggio T. (2000). A trainable system for object detection, International journal of computer vision, 38(1), 15-33. https://doi.org/10.1023/A:1008162616689
  11. Pradhan, R. N. (2020). Towards Modular Multispectral Object Detection using Adaptive Weighted Fusion, Order No. 28149629, University of Toronto (Canada).
  12. Redmon, J., Divvala S. and Girshick, R. (2016). You Only Look Once: Unified, Real-Time Object Detection, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pp.779-788.
  13. Viola, P., Jones, M. (2001). Rapid object detection using a boosted cascade of simple features, CVPR Proceedings of the 2001 IEEE Computer Society Conference.
  14. Wang, C., Liao, H. M., and Wu, Y. (2020). CSPNet: A New Backbone that can Enhance Learning Capability of CNN, 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), pp.1571-1580.
  15. Youdong L., Yongtao W., and Ting-Ting L. (2019). CBNet: A Novel Composite Backbone Network Architecture for Object Detection, Proceedings of the AAAI Conference on Artificial Intelligence, 34(07), 11653-11660.
  16. Zhang, H., Fromont, E. and Lefevre, E. (2020). Multispectral Fusion for Object Detection with Cyclic Fuse-and-Refine Blocks, 2020 IEEE International Conference on Image Processing (ICIP), pp. 276-280.