Video sequences retrieval algorithm

Mathematical support and software for computers, complexes and networks


Lukin V. N.*, Nikitin I. K.**

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia



The article focuses on the algorithms of the event detection in content-based video retrieval. Video has a complex structure and can express the same idea in different ways. This makes the task of searching for video more complicated. Video titles and text descriptions cannot give the whole information about objects and events in the video. This creates a need for content-based video retrieval. There is a semantic gap between low-level video features, that can be extracted, and the users’ perception. The task of event detection is reduced to the task of video segmentation. Complex content-based video retrieval can be regarded as the bridge between traditional retrieval and semantic-based video retrieval. The properties of video as a time series are described. The concept of anomalies in the video is introduced. A method for event detection based on comparing moving averages with windows of different sizes is proposed. According to the classification given at the beginning of this article, our method refers to statistical methods. It differs from other methods of low computational complexity and simplicity. The video stream processing language is proposed for function-based description of video handling algorithms. So, our method is formulated in the form of a declarative description on an interpreted programming language. Unfortunately, most of the existing video processing methods use exclusively imperative approach, which often complicate its understanding. Examples of this language implementation are given. Its grammar is described either. As it was shown by the experiments, the implementation of the proposed video events retrieval method, unlike their counterparts, can work for video streams as well with a real-time and potentially infinite frame sequences. Such advantages within low computational requirements make implementation of the method helpful in aviation and space technology. The algorithm has some disadvantages due to necessity of parameter selection for particular task classes. The theorem on near-duplicates of video is formulated at the end of the article. It asserts the near-duplicate videos express the same sequence of phenomena.


discord detection, video segmentation, video duplicates, moving average score, video streaming


  1. Boreczky J.S., Wilcox L.D. A hidden Markov model framework for video segmentation using audio and image features. Proceedings of the International Conference on Acoustics, Speech, and Signal Processing (Seattle, WA), 12 May 1998, vol. 6, pp. 3741–3744.

  2. Cernekova Z., Pitas I., Nikou C. Information theory-based shot cut/fade detection and video summarization. IEEE Transactions on Circuits and Systems for Video Technology, 2006, vol. 16, no. 1, pp. 82–91.

  3. Cooper Matthew, Liu Ting, Rieffel Eleanor G. Video Segmentation via Temporal Pattern Classification. IEEE Transactions on Multimedia, 2007, vol. 9, no. 3, pp. 610–618.

  4. Jinhui Yuan, Huiyi Wang, Lan Xiao et al. A Formal Study of Shot Boundary Detection. IEEE Transactions on Circuits and Systems for Video Technology, 2007, vol. 17, no. 2, pp. 168–186.

  5. Xue Ling, Li Chao, Xiong Zhang, Li Huan. A General Method for Shot Boundary Detection. Multimedia and Ubiquitous Engineering, International Conference on, 2008, pp. 394-397, doi:10.1109/MUE.2008.102, URL:

  6. Hanjalic Alan. Shot-boundary detection: unraveled and resolved?. IEEE Transactions on Circuits and Systems for Video Technology, 2002, vol. 12, no. 2, pp. 90–105.

  7. Hoi Steven C.H., Wong Lawson, Lyu Albert. Chinese University of Hong Kong at TRECVID 2006: Shot Boundary Detection and Video Search. Int. TREC Video Retrieval workshop (TRECVID’06), 2006, URL:

  8. Liu Chunxi, Liu Huiying, Jiang Shuqiang et al. JDL at Trecvid 2006 Shot Boundary Detection, 2006, URL:

  9. Wan-Lei Zhao, Chong-Wah Ngo, Hung-Khoon Tan, Xiao Wu. Near-Duplicate Keyframe Identification With Interest Point Matching and Pattern Learning. IEEE Transactions on Multimedia, 2007, vol. 9, no. 5, pp. 1037–1048.

  10. Ngo Chong-Wah. A Robust Dissolve Detector by Support Vector Machine. Proceedings of the Eleventh ACM International Conference on Multimedia, MULTIMEDIA ’03, New York, USA, 2003, pp. 283–286.

  11. Quénot Georges, Moraru Daniel, Besacier Laurent. CLIPS at TRECvid: Shot Boundary Detection and Feature Detection. TRECVID 2003 Workshop Notebook Papers, Gaithersburg, MD, USA, 2003, pp. 18–21

  12. Kazunori Matsumoto, Masaki Naito, Keiichiro Hoashi, Fumiaki Sugaya. SVM-Based Shot Boundary Detection with a Novel Feature. IEEE International Conference on Multimedia and Expo, 2006, Toronto, ON, Canada, pp. 1837 – 1840.

  13. Xiaomeng Wu, Masao Takimoto, Shin’ichi Satoh, Jun Adachi. Scene Duplicate Detection Based on the Pattern of Discontinuities in Feature Point Trajectories. Proceedings of the 16th ACM International Conference on Multimedia (MM), New York, NY, USA, ACM, 2008, pp. 51–60.

  14. G Camara-Chavez, F Precioso, M Cord, S Phillip-Foliguet, A de A Araujo. Shot boundary detection by a hierarchical supervised approach. Systems, Signals and Image Processing, 2007 and 6th EURASIP Conference focused on Speech and Image Processing, Multimedia Communications and Services. 14th International Workshop on, June 2007, pp. 197–200.

  15. Prem Kumar Kalra, Shmuel Peleg. Computer Vision, Graphics and Image Processing. 5th Indian Conference, ICVGIP 2006, Madurai, India, December 13-16, 2006, Proceedings. Lecture Notes in Computer Science 4338, Springer 2006, URL: icvgip2006.html.3

  16. van Rossum Guido, Drake Fred L. The Python Language Reference Manual. Bristol, United Kingdom, Network Theory Ltd., 2003. 144 p.

  17. Gusev V.Yu. Krapivenko A.V. Trudy MAI, 2012, no. 50, available at:

  18. Marchuk V.I., Tokareva S.V. Sposoby obnaruzhenia anomal’nykh znachenii pri analize nestatsionarnykh sluchainykh signalov (Methods for detecting the abnormal value in the analysis of non-stationary random signals), Shakhty, IuRGUS, 2009, 209 p.

  19. Prutov I.S. Sincha D.P. Trudy MAI, 2012, no. 52, available at:

Download — informational site MAI

Copyright © 2000-2022 by MAI