DOI QR코드

DOI QR Code

Surveillance Video Summarization System based on Multi-person Tracking Status

다수 사람 추적상태에 따른 감시영상 요약 시스템

  • 유주희 (덕성여자대학교 컴퓨터학과) ;
  • 이경미 (덕성여자대학교 컴퓨터학과)
  • Received : 2015.08.28
  • Accepted : 2015.12.07
  • Published : 2016.02.15

Abstract

Surveillance cameras have been installed in many places because security and safety has become an important issue in modern society. However, watching surveillance videos and judging accidental situations is very labor-intensive and time-consuming. So now, requests for research to automatically analyze the surveillance videos is growing. In this paper, we propose a surveillance system to track multiple persons in videos and to summarize the videos based on tracking information. The proposed surveillance summarization system applies an adaptive illumination correction, subtracts the background, detects multiple persons, tracks the persons, and saves their tracking information in a database. The tracking information includes tracking one's path, their movement status, length of staying time at the location, enterance/exit times, and so on. The movement status is classified into six statuses(Enter, Stay, Slow, Normal, Fast, and Exit). This proposed summarization system provides a person's status as a graph in time and space and helps to quickly determine the status of the tracked person.

현대사회는 보안과 안전이 중요해지면서 감시카메라들이 여러 곳에 설치되어 있다. 하지만 감시영상을 보고 상황을 파악하는 것은 여전히 사람의 몫으로 인력과 시간이 소모된다. 그래서 자동으로 감시영상을 분석하여 주요 사건 중심으로 요약해 주는 연구의 필요성이 커지고 있다. 본 논문에서는 감시영상에서 존재하는 다수의 사람을 추적하고, 추적을 통해 얻은 정보를 이용하여 감시영상을 요약하는 방법을 제안한다. 제안하는 감시영상 요약 시스템은 조명보정을 적용하여 배경제거한 후 다수의 사람을 추출하고, 추출된 사람의 추적 정보를 상태 데이터베이스에 저장한다. 추적을 통해 얻은 정보로 추적 대상들의 추적 경로, 움직임 상태, 지체시간, 카메라 안으로의 출입시간 등을 사용한다. 또 사람의 움직임에 따라 6 가지(Enter, Stay, Slow, Normal, Fast and Exit)로 움직임 상태를 분류하였고, 움직임 상태를 시간별, 공간별로 요약 그래프로 나타내 추적대상의 움직임 상태를 빠르게 파악할 수 있다.

Keywords

References

  1. P. Banerjee and S. Sengupta, "Human motion detection and tracking for video surveillance," Proc. of the National Conference on Communication, pp. 88-92, 2008.
  2. H. Patel and M. Wankhade, "Human tracking in video surveillance," Advances in Computing and Information Technology, pp. 749-756, 2012.
  3. J. Han, T. Tan, L. Chen, and D. Zhang, "Moving objects representation for object based surveillance video retrieval system," International Journal of Security and Its Applications, Vol. 8, No. 2, pp. 315-322, 2014. https://doi.org/10.14257/ijsia.2014.8.2.32
  4. G. Shu, A. Dehghan, O. Oreifej, E. Hand, and M. Shah, "Part-based multiple-person tracking with partial occlusion handling," Proc. of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 1815-1821, 2012.
  5. B. Cancela, M. Ortega, and M. G. Penedo, "Multiple human tracking system for unpredictable trajectories," Machine Vision and Applications, Vol. 25, No. 2, pp. 511-527, 2014. https://doi.org/10.1007/s00138-013-0544-7
  6. W. Choi and S. Savarese, "Multiple target tracking in world coordinate with single, minimally calibrated camera," Proc. of the European Conference on Computer Vision, pp. 553-567, 2010.
  7. H. Xuan, F. Croquer, and W. Zeng, "Surveillance video retrieval based on moving objects detection," Journal of Multimedia, Vol. 8, No. 6, pp. 762-769, 2013.
  8. H. Kwon and K. Lee, "Activity-based key-frame detection and video summarization in a wide-area surveillance system," Journal of Korean Society for Internet Information, Vol. 9, No. 3, pp. 169-178, 2008. (in Korean)
  9. K. Lee, "Tracking multi-person robust to illumination changes and occlusions," Proc. of the 14th International Conference on Artificial Reality and Telexistence, pp. 429-432, 2004.