한국군사과학기술학회지 (Journal of the Korea Institute of Military Science and Technology)

  • 제27권6호
  • /
  • Pages.684-692
  • /
  • 2024
  • /
  • 1598-9127(pISSN)
  • /
  • 2636-0640(eISSN)
한국군사과학기술학회 (The Korea Institute of Military Science and Technology)
권호 표지
DOI QR코드

DOI QR Code

딥러닝 모델 기반 사물-인체 이종 그래프 추출을 활용한 총기 인식 및 위협 감지 기법

Risk Detection through Firearm Recognition Using Deep Learning-Based Object-Human Heterogeneous Graph Extraction

  • 양정은 (포항공과대학교 IT융합공학과) ;
  • 백종은 (포항공과대학교 컴퓨터공학과) ;
  • 샤킬라 쇼제이 (포항공과대학교 IT융합공학과) ;
  • 배효진 (포항공과대학교 IT융합공학과) ;
  • 박주홍 (포항공과대학교 IT융합공학과)
  • Jeongeun Yang (Department of Convergence IT Engineering, Pohang University of Science and Technology) ;
  • Jongeun Baek (Department of Computer Science and Engineering, Pohang University of Science and Technology) ;
  • Shakila Shojaei (Department of Convergence IT Engineering, Pohang University of Science and Technology) ;
  • Hyojin Bae (Department of Convergence IT Engineering, Pohang University of Science and Technology) ;
  • Juhong Park (Department of Convergence IT Engineering, Pohang University of Science and Technology)
  • 투고 : 2024.06.27
  • 심사 : 2024.09.29
  • 발행 : 2024.12.05
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Effective border security is crucial in managing and mitigating firearm-related threats. While prior research has focused on firearm detection, it lacks contextual analysis. This paper advances firearm-related incident assessment by integrating pose estimation to improve gun violence detection. Our novel approach extracts body and firearm pose graphs and employs Graph Attention Networks(GAT) for graph analysis to accurately identify gun violence incidents. By recognizing associated actions, our system provides greater situational awareness beyond mere firearm detection. Utilizing Graph-LSTM, we capture spatial and temporal information. As a result, our proposed algorithm is lighter and more accurate than the CNN-LSTM model used as a baseline, achieving test F1-scores of 82.04 % on our collected data.

키워드

과제정보

이 논문은 2024년도 정부(과학기술정보통신부)의 재원으로 정보통신기획평가원의 지원(No. RS-2023-00228996, 우주상황인식을 위한 실-가상 연동형 국방 메타버스 기반 기술 개발) 및 2024년도 정부(과학기술정보통신부)의 재원으로 한국연구재단의 지원(No. RS-2024-00419657, 우주 현지 자원을 활용한 건축 기술)을 받아 수행된 연구임.

참고문헌

  1. Kim, J., and Choi, S., "Performance Improvement of a Deep Learning-based Object Recognition using Imitated Red-green Color Blindness of Camouflaged Soldier Images," J. KIMS Technol., Vol. 23, No. 2, pp. 139-146, 2020.
  2. Lee, H., and Park, J., "A Study on Maritime Object Image Classification Using a Pruning-Based Lightweight Deep-Learning Model," J. KIMS Technol., Vol. 27, No. 3, pp. 346-354, 2024.
  3. Kim, M., and Lee, S., "Experimental Research on the Optimal Surveillance Equipment Allocation Using Geo-spatial Information," J. KIMS Technol., Vol. 9, No. 1, pp. 72-79, 2006.
  4. Gelana, F., and Yadav, A., "Firearm Detection from Surveillance Cameras Using Image Processing and Machine Learning Techniques," Smart Innovations in Communication and Computational Sciences: Proceedings of ICSICCS-2018, Springer Singapore, pp. 203-215, 2019.
  5. de Azevedo Kanehisa, R. F., and de Almeida Neto, A., "Firearm Detection using Convolutional Neural Networks," ICAART (2), pp. 45-52, 2019.
  6. Si, C., Chen, W., Wang, W., Wang, L., and Tan, T., "An Attention Enhanced Graph Convolutional LSTM Network for Skeleton-Based Action Recognition,"  Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp. 1227-1236, 2019.
  7. Zhu, W., Lan, C., Xing, J., Zeng, W., Li, Y., and Shen, L., "Co-Occurrence Feature Learning for Skeleton Based Action Recognition Using Regularized Deep LSTM Networks," Proceedings of the Thirtieth AAAI Conference on Artificial Intelligence, pp. 3697-3703, 2016.
  8. Zhang, H., Zhu, X., Yao, T., Li, Y., and Mei, T., "Graph Convolutional LSTM Model for Skeleton-Based Action Recognition," Proceedings of the IEEE International Conference on Multimedia and Expo (ICME), pp. 645-650, 2019.
  9. Jocher, G., et al., "YOLOv8: Open Source Pose Model," Available at: https://github.com/ultralytics/yolov8, 2023.
  10. Velickovic, P., et al., "Graph Attention Networks," arXiv preprint arXiv:1710.10903, 2017.
  11. Salazar-Gonzalez, A., et al., "Simulation of Attacks Dataset," Journal of XYZ Research, Vol. 12, No. 4, pp. 123-134, 2020.
  12. Brooks, M., "COCO Annotator," Available at: https://github.com/jsbroks/coco-annotator, 2019.
  13. Paszke, A., et al., "Pytorch: An Imperative Style, High-Performance Deep Learning Library," Advances in Neural Information Processing Systems, Vol. 32, pp. 8024-8035, 2019.
  14. Wang, M., et al., "Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks," arXiv preprint arXiv:1909.01315, 2019.
  15. Donahue, J., Hendricks, L. A., Venugopalan, S., Guadarrama, S., Saenko, K., Rohrbach, M., and Darrell, T., "Long-term Recurrent Convolutional Networks for Visual Recognition and Description," Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition(CVPR), pp. 2625-2634, 2015.