한국컴퓨터정보학회논문지 (Journal of the Korea Society of Computer and Information)

  • 제24권3호
  • /
  • Pages.87-93
  • /
  • 2019
  • /
  • 1598-849X(pISSN)
  • /
  • 2383-9945(eISSN)
한국컴퓨터정보학회 (Korean Society of Computer Information)
권호 표지
DOI QR코드

DOI QR Code

Deep Learning based violent protest detection system

  • 투고 : 2019.02.12
  • 심사 : 2019.03.13
  • 발행 : 2019.03.29
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

In this paper, we propose a real-time drone-based violent protest detection system. Our proposed system uses drones to detect scenes of violent protest in real-time. The important problem is that the victims and violent actions have to be manually searched in videos when the evidence has been collected. Firstly, we focused to solve the limitations of existing collecting evidence devices by using drone to collect evidence live and upload in AWS(Amazon Web Service)[1]. Secondly, we built a Deep Learning based violence detection model from the videos using Yolov3 Feature Pyramid Network for human activity recognition, in order to detect three types of violent action. The built model classifies people with possession of gun, swinging pipe, and violent activity with the accuracy of 92, 91 and 80.5% respectively. This system is expected to significantly save time and human resource of the existing collecting evidence.

키워드

CPTSCQ_2019_v24n3_87_f0001.png 이미지

Fig. 1. Related Work: Screenshots

CPTSCQ_2019_v24n3_87_f0002.png 이미지

Fig. 2. Architecture of the proposed system

CPTSCQ_2019_v24n3_87_f0003.png 이미지

Fig. 3. Data augmentation

CPTSCQ_2019_v24n3_87_f0004.png 이미지

Fig. 5. Training

CPTSCQ_2019_v24n3_87_f0005.png 이미지

Fig. 4. Data labeling

CPTSCQ_2019_v24n3_87_f0006.png 이미지

Fig. 6. Early stopping point

CPTSCQ_2019_v24n3_87_f0007.png 이미지

Fig. 8. Detection Result

CPTSCQ_2019_v24n3_87_f0008.png 이미지

Fig. 7. Upload screenshot

Table 1. Related Work

CPTSCQ_2019_v24n3_87_t0001.png 이미지

Table 2. Parameters and results

CPTSCQ_2019_v24n3_87_t0002.png 이미지

Table 3. Data set

CPTSCQ_2019_v24n3_87_t0003.png 이미지

참고문헌

  1. AWS(AmazonWebService) "https://aws.amazon.com/?nc2=h_lg"
  2. Naver about protest "https://terms.naver.com/entry.nhn?docId=1820712&cid=46634&categoryId=46634"
  3. ChosunIlbo, ratio of demonstration "http://news.chosun.com/site/data/html_dir/2018/11/03/2018110300250.html"
  4. Navernews research about drone "http://www.sportsseoul.com/news/read/617479"
  5. CatChaser(2016)ByRobertBond Nvidia "https://devblogs.nvidia.com/ai-cat-chaser-jetson-tx1-caffe"
  6. SINGH, Amarjot; PATIL, Devendra; OMKAR, S. N. Eye in the Sky: Real-time Drone Surveillance System (DSS) for Violent Individuals Identification using ScatterNet Hybrid Deep Learning Network. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition Workshops, p. 1629-1637, June 2018
  7. How To Do Real Time Object Detection OnDroneVideoStreams "https://blog.nanonets.com/real-time-object-detection-for-drones/?utm_source=facebook&utm_medium=social&utm_campaign=drone&utm_content=group_DeepLearnng&fbclid=IwAR34ULqditi4eM3Gvm-Ic84lM6LDA9ElQR6bmFaIsqRRJ1_hgauCgMTa-v8"
  8. REDMON, Joseph; FARHADI, Ali. Yolov3: An incremental improvement. arXiv preprint arXiv:1804.02767, April 2018.
  9. AWS Amazon Simple Storage Service "https://aws.amazon.com/s3/?nc2=type_a"
  10. ZHANG, Jianming, et al. A real-time chinese traffic sign detection algorithm based on modified YOLOv2. Algorithms, Vol. 10 Issue 4, p127. 13p, Dec2017.
  11. GoogleAPI "https://support.google.com/webmasters/answer/1061943?hl=ko"
  12. Google crawling using selenium "https://www.slideshare.net/winglessagl/seleniu-61525457"
  13. Image capture FFmpeg "https://www.ffmpeg.org/"
  14. Thumos Challenge2015 "http://www.thumos.info/download.html"
  15. Large Scale Visual Recognition Challenge 2014"http://www.image-net.org/challenges/LSVRC/2014/"
  16. SZEGEDY, Christian, et al. Going deeper with convolutions. In: Proceedings of the IEEE conference on computer vision and pattern recognition. CVPR , p. 1-9, June 2015.
  17. Labeling Yolo_mark "https://github.com/AlexeyAB/Yolo_mark"
  18. REDMON, Joseph; FARHADI, Ali. YOLO9000: better, faster, stronger. In: Proceedings of the IEEE conference on computer vision and pattern recognition. CVPR, p. 7263-7271, July 2017.
  19. Lin, Tsung-Yi, et al. "Feature pyramid networks for object detection." CVPR. Vol. 1. No. 2. July 2017.
  20. Tistory Blog yeonsu "http://yeonsuuu-uuu.tistory.com/"
  21. Live streaming RTMP Server "http://blog.manycam.com/rtmp-live-stream/#.XATy5hMzbcM"