Journal of the Korea Institute of Information and Communication Engineering (한국정보통신학회논문지)

The Korea Institute of Information and Commucation Engineering (한국정보통신학회)
권호 표지
DOI QR코드

DOI QR Code

Behavior Pattern Analysis and Design of Retrieval Descriptor based on Temporal Histogram of Moving Object Coordinates

이동 객체 좌표의 시간적 히스토그램 기반 행동패턴 분석 및 검색 디스크립터 설계

  • Lee, Jae-kwang (Department of Information and Communications Eng., Daejeon University) ;
  • Lee, Kyu-won (Department of Information and Communications Eng., Daejeon University)
  • Received : 2016.11.16
  • Accepted : 2016.11.28
  • Published : 2017.04.30
Download PDF
⟨ Previous Next ⟩

Abstract

A behavior pattern analysis algorithm based on descriptors consists of information of a moving object and temporal histogram is proposed. Background learning is performed first for detecting, tracking and analyzing moving objects. Each object is identified using an association of the center of gravity of objects and tracked individually. A temporal histogram represents a motion pattern using positions of the center of gravity and time stamp of objects. The characteristic and behavior of objects are figured out by comparing each coordinates of a position history in the histogram. Behavior information which is comprised with numbers of a start and end frame, and coordinates of positions of objects is stored and managed in the linked list. Descriptors are made with the stored information and the video retrieval algorithm is designed. We confirmed the higher retrieval accuracy compare with conventional methods.

이동 객체 정보로 이루어진 디스크립터 및 시간적 히스토그램 기반 움직임 패턴 분석 알고리즘을 제안한다. 이동객체의 검출, 추적 및 분석을 위하여 배경으로부터 이동객체를 분리하는 배경학습을 수행한다. 무게중심의 좌표연관성을 이용하여 객체를 식별한 후 객체별로 추적한다. 시간적 히스토그램은 객체의 무게중심의 위치와 시간 정보를 이용해 움직임 특징 패턴을 정의한 것으로서 시간적 히스토그램으로부터 각 객체의 좌표정보와 비교하여 움직임특징 및 행동정보를 파악한다. 검출된 각 객체의 시작프레임, 종료프레임, 위치 등 행동정보를 연결리스트에 저장하여 관리한다. 저장된 정보들을 바탕으로 디스크립터를 작성하고 비디오 검색 알고리즘을 설계한다. 다양한 객체 이동 비디오에 대한 검색 실험을 통해 기존 방법보다 높은 검색 정확도를 보임을 확인하였다.

Keywords

References

  1. Y. J. Song, "Trend of video contents retrieval technology," Journal of Korea Contents Association, vol. 7, no. 1, pp. 46-52, July 2009.
  2. N. Dimitrova and F. Golshani, "Motion recovery for video content classification," ACM Transaction of Information Systems, vol. 13, no. 4, pp. 408-439, April 1995. https://doi.org/10.1145/211430.211433
  3. D. S. Messing, P. van Beek and J.H Errico, "The MPEG-7 colour structure descriptor: image description using colour and local spatial information," in Proceedings of International Conference on Image Processing, vol. 1, pp. 670-673, 2001.
  4. J. D. Jeon, M. J. Lee, J. H. Kim, D. G. Kim and B. D. Kang, "Object detection system using Eigen-background and Clustering," Journal of Korea Multimedia Society, vol. 13, no. 1, pp. 45-57, Jan. 2010.
  5. J. H. Han, Y.J.Kim, S.W.Yoo, S.H.Lee and J.I.Park, "Effective extraction of moving objects in a dynamic environment," in Proceedings of Korea HCI Association Conference, pp. 631-636, 2009.
  6. K. H. Kang, Y. S. Park, Y. I. Yun, J.S.Choi and D.W.Kim "Image search using spatial information and color change rate," Journal of Korea Multimedia Society, vol. 11, no.1, pp. 23-33, Nov. 2008.
  7. Byung-man Jung and Kyu-won Lee. "A Descriptor Design for the Video Retrieval Combining the Global Feature of an Image and the Local of a Moving Object," Journal of Korea Institute of Information and Communication Engineering, vol. 18, no. 1, pp. 142-148, Jan. 2014. https://doi.org/10.6109/jkiice.2014.18.1.142
  8. H. B. Kang and C.S.Won, "Image search using combination of MPEG-7 descriptors," Journal of Broadcasting Engineering, vo. 8, no. 1, pp. 91-100, Aug. 2003.
  9. H.Y.Lim and D.S.Kang, "A Study on Improved Object Tracking Algorithm Using Color Probability Distribution for Object Tracking System," Journal of Korea Information Science Society, vol. 8, no. 2, pp. 153-158, Aug. 2010.
  10. R. Y. Si, H. R. Kim, T. H. Kim and Y. H. Ju, "Multiple moving object detection and tracking algorithm," in Proceedings of Summer Conference of the Korean Institute of Electrical Engineers, pp. 18-20, 2012.