Recognition of Situations on Set of Moving Objects using Fuzzy Finite State Machines and Dynamic Programming

Authors: Devyatkov V.V., Lychkov I.I. Published: 02.08.2017
Published in issue: #4(115)/2017  
DOI: 10.18698/0236-3933-2017-4-64-78

Category: Informatics, Computer Engineering and Control | Chapter: System Analysis, Control, and Information Processing  
Keywords: computer vision, moving objects, time series, recognition of situations, dynamic programming

Recognition of situations on set of moving objects is a crucial task for human security in transport and public places. This work improves time series approach to recognition of situations based on the analysis of time series of moving object coordinates. Hidden Markov model, dynamic time warping and finite state machines are the most studied time series methods for recognition of situations. Methods based on hidden Markov model and dynamic time warping were originally developed for recognition of situations in presence of noise in time series; however, they require laborious programming by examples. Methods based on finite state machine are training free, however, they require extra tools for noise eliminations, e.g. filters, and lose accuracy under noisy conditions. This work proposes a new time-series method for recognition of situations which is training free and resistant to faulty readings.


[1] Turaga P., Chellappa R., Subrahmanian V.S., Udrea O. Machine recognition of human activities: a survey. IEEE Transactions on Circuits and Systems for Video Technology, 2008, vol. 18, no. 11, pp. 1473-1488. DOI: 10.1109/TCSVT.2008.2005594 Available at: http://ieeexplore.ieee.org/document/4633644

[2] Lin C.Y., Wang S.M., Hong J.W., Kang L.W., Huang C.L. Vision-based fall detection through shape features. Proc. IEEE Int. Conf. on Multimedia Big Data, 2016, pp. 237-240. DOI: 10.1109/BigMM.2016.22 Available at: http://ieeexplore.ieee.org/document/7545029

[3] Wang B., Hu Y., Gao J., Sun Y., Yin B. Laplacian LRR on product Grassmann manifolds for human activity clustering in multi-camera video surveillance. IEEE Transactions on Circuits and Systems for Video Technology, 2017, vol. 27, no. 3, pp. 554-566. DOI: 10.1109/TCSVT.2016.2609760 Available at: http://ieeexplore.ieee.org/document/7569105

[4] Shatalin R.A., Ovchinnikov P.E. [Abnormal situation detection algorithm in observation problem based on method of principal components]. Perspektivnye informatsionnye tekhnologii: Sbornik trudov Mezhdunarodnoy nauchno-tekhnicheskoy konferentsii. T.1 [Cutting-edge informational technologies: proc. Int. sci.-tech. conf. Vol. 1]. 2015, pp. 240-244 (in Russ.).

[5] Tokarev V.L., Abramov D.A. Methods of allocation abnormal situations in the information-measuring system surveillance. Izvestiya TulGU. Tekhnicheskie nauki [News of the Tula state university. Technical sciences], 2015, no. 11-1, pp. 258-265 (in Russ.).

[6] Seo H.J., Milanfar P. Action recognition from one example. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2011, vol. 33, no. 5, pp. 867-882. DOI: 10.1109/TPAMI.2010.156 Available at: http://ieeexplore.ieee.org/document/5557879

[7] Everts I., Van Gemert J.C., Gevers T. Evaluation of color STIPs for human action recognition. Proc. IEEE Conf. on Computer Vision and Pattern Recognition, 2013, pp. 2850-2857. DOI: 10.1109/CVPR.2013.367 Available at: http://ieeexplore.ieee.org/document/6619211

[8] Soomro K., Zamir A.R. Action recognition in realistic sports videos. In: Moeslund T.B. et al. (eds.). Computer vision in sports. Switzerland, Springer International Publishing, 2014, pp. 181-208.

[9] Noorit N., Suvonvorn N. Human activity recognition from basic actions using finite state machine. Proc. Int. Conf. on Advanced Data and Information Engineering, 2014, pp. 379-386.

[10] Ji X., Wang C., Li Y., Wu Q. Hidden Markov model-based human action recognition using mixed features. Journal of Computational Information Systems, 2013, vol. 9, no. 9, pp. 3659-3666.

[11] Pham C.H., Le Q.K., Le T.H. Human action recognition using dynamic time warping and voting algorithm. VNU Journal of Science: Computer Science and Communication Engineering, 2014, vol. 30, no. 3, pp. 22-30. Available at: http://www.jcsce.vnu.edu.vn/index.php/jcsce/article/view/15

[12] Fang C. From dynamic time warping (DTW) to hidden Markov model (HMM). University of Cincinnati, 2009. 19 p.

[13] Kumar S.K., Kant L.K., Shachi S. HMM based enhanced dynamic time warping model for efficient hindi language speech recognition system. In: Das V.V., Chaba Y. (eds.). Mobile communication and power engineering. Springer Berlin Heidelberg, 2013, pp. 200-206.

[14] Bhuyan M.K. FSM-based recognition of dynamic hand gestures via gesture summarization using key video object planes. International Journal of Computer and Communication Engineering, 2012, vol. 1, no. 6, pp. 248-259.

[15] Seto S., Zhang W., Zhou Y. Multivariate time series classification using dynamic time warping template selection for human activity recognition. IEEE Symp. Series on Computational Intelligence, 2015, pp. 1399-1406. DOI: 10.1109/SSCI.2015.199 Available at: http://ieeexplore.ieee.org/document/7376775

[16] Trinh H., Fan Q., Jiyan P., Gabbur P., Miyazawa S., Pankanti S. Detecting human activities in retail surveillance using hierarchical finite state machine. Proc. IEEE Int. Conf. on Acoustics, Speech and Signal Processing, 2011, pp. 1337-1340. DOI: 10.1109/ICASSP.2011.5946659 Available at: http://ieeexplore.ieee.org/document/5946659

[17] Kerr W., Tran A., Cohen P. Activity recognition with finite state machines. Proc. Int. Joint Conf. on Artificial Intelligence, 2011, vol. 22, no. 1, pp. 1348-1353. DOI: 10.5591/978-1-57735-516-8/IJCAI11-228 Available at: https://www.ijcai.org/Proceedings/11/Papers/228.pdf

[18] Bertsekas D.P. Dynamic programming and optimal control. Vol. 1. 3rd Edition. Belmont (USA), Athena Scientific, 2005. 558 p.

[19] Allen B.L., Shin B.T., Cooper D.J. Analysis of traffic conflicts and collision. Journal of the Transportation Research Board, 1978, vol. 667, pp. 67-74.

[20] Lychkov I.I. [Optimal moving object tracking in video stream]. Sovremennye dostizheniya i razrabotki v oblasti tekhnicheskikh nauk. Sbornik nauchnykh trudov po itogam mezhdunarodnoy nauchno-prakticheskoy konferentsii [Present-day achievements and developments in the area of technical sciences. Proc. Int. sci.-practice conf.]. Orenburg, Evansys Publ., 2016, pp. 5-13 (in Russ.).

[21] Abbey Road - Crossing Webcam. abbeyroad.com: website. Available at: http://www.abbeyroad.com/crossing (accessed 18.09.2016)