Neural Computing and Applications

, Volume 28, Supplement 1, pp 439–459 | Cite as

Constrained self-organizing feature map to preserve feature extraction topology

  • Jorge Azorin-LopezEmail author
  • Marcelo Saval-Calvo
  • Andres Fuster-Guillo
  • Jose Garcia-Rodriguez
  • Higinio Mora-Mora
Original Article


In many classification problems, it is necessary to consider the specific location of an n-dimensional space from which features have been calculated. For example, considering the location of features extracted from specific areas of a two-dimensional space, as an image, could improve the understanding of a scene for a video surveillance system. In the same way, the same features extracted from different locations could mean different actions for a 3D HCI system. In this paper, we present a self-organizing feature map able to preserve the topology of locations of an n-dimensional space in which the vector of features have been extracted. The main contribution is to implicitly preserving the topology of the original space because considering the locations of the extracted features and their topology could ease the solution to certain problems. Specifically, the paper proposes the n-dimensional constrained self-organizing map preserving the input topology (nD-SOM-PINT). Features in adjacent areas of the n-dimensional space, used to extract the feature vectors, are explicitly in adjacent areas of the nD-SOM-PINT constraining the neural network structure and learning. As a study case, the neural network has been instantiate to represent and classify features as trajectories extracted from a sequence of images into a high level of semantic understanding. Experiments have been thoroughly carried out using the CAVIAR datasets (Corridor, Frontal and Inria) taken into account the global behaviour of an individual in order to validate the ability to preserve the topology of the two-dimensional space to obtain high-performance classification for trajectory classification in contrast of non-considering the location of features. Moreover, a brief example has been included to focus on validate the nD-SOM-PINT proposal in other domain than the individual trajectory. Results confirm the high accuracy of the nD-SOM-PINT outperforming previous methods aimed to classify the same datasets.


Self-organizing feature map Topology preservation Human behaviour analysis 



This study was supported in part by the University of Alicante, Valencian Government and Spanish government under grants GRE11-01, GV/2013/005 and DPI2013-40534-R.


  1. 1.
    Anjum N, Cavallaro A (2007) Single camera calibration for trajectory-based behavior analysis. In: IEEE conference on advanced video and signal based surveillance, 2007. AVSS 2007. IEEE, pp 147–152Google Scholar
  2. 2.
    Anjum N, Cavallaro A (2008) Multifeature object trajectory clustering for video analysis. IEEE Trans Circ Syst Video Technol 18(11):1555–1564CrossRefGoogle Scholar
  3. 3.
    Anjum N, Cavallaro A (2010) Trajectory clustering for scene context learning and outlier detection. In: Schonfeld D, Shan C, Tao D, Wang L (eds) Video search and mining, vol 280. Springer, Berlin, pp 33–51. doi: 10.1007/978-3-642-12900-1_2 CrossRefGoogle Scholar
  4. 4.
    Antonakaki P, Kosmopoulos D, Perantonis SJ (2009) Detecting abnormal human behaviour using multiple cameras. Signal Process 89(9):1723–1738CrossRefzbMATHGoogle Scholar
  5. 5.
    Azorin-Lopez J, Saval-Calvo M, Fuster-Guillo A, Garcia-Rodriguez J (2013) Human behaviour recognition based on trajectory analysis using neural networks. In: International joint conference in neural networks, 2013Google Scholar
  6. 6.
    Azorin-Lopez J, Saval-Calvo M, Fuster-Guillo A, Garcia-Rodriguez J (2015) A novel prediction method for early recognition of global human behaviour in image sequences. Neural Process Lett. doi: 10.1007/s11063-015-9412-y Google Scholar
  7. 7.
    Azorin-Lopez J, Saval-Calvo M, Fuster-Guillo A, Oliver-Albert A (2014) A predictive model for recognizing human behaviour based on trajectory representation. In: 2014 International joint conference on neural networks, IJCNN 2014, Beijing, China, July 6–11, 2014, pp 1494–1501Google Scholar
  8. 8.
    Blunsden S, Fisher RB (2010) The BEHAVE video dataset: ground truthed video for multi-person behavior classification. Ann BMVA 2010(4):1–12Google Scholar
  9. 9.
    Brown M, Lowe DG (2005) Unsupervised 3d object recognition and reconstruction in unordered datasets. In: IEEE fifth international conference on 3-D digital imaging and modeling, 2005, 3DIM 2005, pp 56–63Google Scholar
  10. 10.
    Cho NG, Kim YJ, Park U, Park JS, Lee SW (2015) Group activity recognition with group interaction zone based on relative distance between human objects. Int J Pattern Recognit Artif Intell 29:1555007. doi: 10.1142/S0218001415550071 MathSciNetCrossRefGoogle Scholar
  11. 11.
    Fisher R, Santos-Victor J, Crowley J (2005) CAVIAR hidden semi-Markov model behaviour recognition.
  12. 12.
    Fisher RB (2004) The PETS04 surveillance ground-truth data sets. In: Sixth IEEE international workshop on performance evaluation of tracking and surveillance (PETS04), pp 1 – 5Google Scholar
  13. 13.
    Fritzke B (1994) Growing cell structures self-organizing network for unsupervised and supervised learning. Neural Netw 7(9):1441–1460CrossRefGoogle Scholar
  14. 14.
    Fritzke B (1995) Growing grid self-organizing network with constant neighborhood range and adaptation strength. Neural Process Lett 2(5):9–13CrossRefGoogle Scholar
  15. 15.
    Fritzke B et al (1995) A growing neural gas network learns topologies. Adv Neural Inf Process Syst 7:625–632Google Scholar
  16. 16.
    Hu W, Xie D, Tan T, Maybank S (2004) Learning activity patterns using fuzzy self-organizing neural network. IEEE Trans Syst Man Cybern Part B Cybern 34(3):1618–1626. doi: 10.1109/TSMCB.2004.826829 CrossRefGoogle Scholar
  17. 17.
    Juan L, Gwun O (2009) A comparison of sift, pca-sift and surf. Int J Image Process IJIP 3(4):143–152Google Scholar
  18. 18.
    Kangas JA, Kohonen TK, Laaksonen JT (1990) Variants of self-organizing maps. IEEE Trans Neural Netw 1(1):93–99CrossRefGoogle Scholar
  19. 19.
    Kim Y, Cho N, Lee S (2014) Group activity recognition with group interaction zone. In: 22nd international conference on pattern recognition, ICPR 2014. IEEE, pp 3517–3521Google Scholar
  20. 20.
    Kohonen T (1982) Clustering, taxonomy, and topological maps of patterns. In: Proceedings of the 6th international conference on pattern recognition, IEEE, pp 114–128Google Scholar
  21. 21.
    Lavee G, Rivlin E, Rudzsky M (2009) Understanding video events: a survey of methods for automatic interpretation of semantic occurrences in video. IEEE Trans Syst Man Cybern Part C Appl Rev 39(5):489–504CrossRefGoogle Scholar
  22. 22.
    Lee ACD, Rinner C (2015) Visualizing urban social change with self-organizing maps: Toronto neighbourhoods, 1996–2006. Habitat Int 45(Part 2):92–98. doi: 10.1016/j.habitatint.2014.06.027 CrossRefGoogle Scholar
  23. 23.
    Li X, Hu W, Hu W (2006) A coarse-to-fine strategy for vehicle motion trajectory clustering. In: 18th international conference on pattern recognition, 2006. ICPR 2006, vol 1. IEEE, pp 591–594Google Scholar
  24. 24.
    Madokoro H, Honma K, Sato K (2012) Classification of behavior patterns with trajectory analysis used for event site. In: The 2012 international joint conference on neural networks (IJCNN), pp 1–8. doi: 10.1109/IJCNN.2012.6252565
  25. 25.
    Martinetz T, Schulten K (1991) A “neural-gas” network learns topologies. Artif Neural Netw l:397–402Google Scholar
  26. 26.
    Martinez-Contreras F, Orrite-Urunuela C, Herrero-Jaraba E, Ragheb H, Velastin Sa (2009) Recognizing human actions using silhouette-based HMM. In: 2009 Sixth IEEE international conference on advanced video and signal based surveillance, pp 43–48. doi: 10.1109/AVSS.2009.46
  27. 27.
    Moeslund TB, Hilton A, Krüger V (2006) A survey of advances in vision-based human motion capture and analysis. Comput Vis Image Underst 104(2–3):90–126CrossRefGoogle Scholar
  28. 28.
    Morris B, Trivedi M (2008) A survey of vision-based trajectory learning and analysis for surveillance. IEEE Trans Circ Syst Video Technol 18(8):1114–1127. doi: 10.1109/TCSVT.2008.927109 CrossRefGoogle Scholar
  29. 29.
    Morris B, Trivedi M (2011) Trajectory learning for activity understanding: unsupervised, multilevel, and long-term adaptive approach. IEEE Trans Pattern Anal Mach Intell 33(11):2287–2301. doi: 10.1109/TPAMI.2011.64 CrossRefGoogle Scholar
  30. 30.
    Münch D, Michaelsen E, Arens M (2012) Supporting fuzzy metric temporal logic based situation recognition by mean shift clustering. In: Glimm B, Krüger A (eds) KI 2012: Advances in artificial intelligence. Springer, Berlin, Heidelberg, pp 233–236CrossRefGoogle Scholar
  31. 31.
    Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) Smote: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357zbMATHGoogle Scholar
  32. 32.
    Naftel A, Khalid S (2006) Classifying spatiotemporal object trajectories using unsupervised learning in the coefficient feature space. Multimed Syst 12(3):227–238. doi: 10.1007/s00530-006-0058-5 CrossRefGoogle Scholar
  33. 33.
    Owens J, Hunter A (2000) Application of the self-organising map to trajectory classification. In: Proceedings of the third IEEE international workshop on visual surveillance, 2000. IEEE, pp 77–83Google Scholar
  34. 34.
    Parisi G, Wermter S (2013) Hierarchical som-based detection of novel behavior for 3d human tracking. In: The 2013 international joint conference on neural networks (IJCNN), pp 1–8. doi: 10.1109/IJCNN.2013.6706727
  35. 35.
    Saul H, Kozempel K, Haberjahn M (2014) A comparison of methods for detecting atypical trajectories. Urban Transp XX 138:393CrossRefGoogle Scholar
  36. 36.
    Saval-Calvo M, Azorin-Lopez J, Fuster-Guillo A, Mora-Mora H (2015) \(\mu\)-mar: multiplane 3d marker based registration for depth-sensing cameras. Expert Syst Appl 42(23):9353–9365CrossRefGoogle Scholar
  37. 37.
    Schreck T, Bernard J, von Landesberger T, Kohlhammer J (2009) Visual cluster analysis of trajectory data with interactive Kohonen maps. Inf Vis 8(1):14–29. doi: 10.1057/ivs.2008.29 CrossRefGoogle Scholar
  38. 38.
    Turaga P, Chellappa R, Subrahmanian V, Udrea O (2008) Machine recognition of human activities: a survey. IEEE Trans Circ Syst Video Technol 18(11):1473–1488CrossRefGoogle Scholar
  39. 39.
    Tweed D, Fisher R, Bins J, List T (2005) Efficient hidden semi-markov model inference for structured video sequences. In: 2nd joint IEEE international workshop on visual surveillance and performance evaluation of tracking and surveillance, 2005, pp 247–254Google Scholar
  40. 40.
    Uriarte EA, Martín FD (2005) Topology preservation in som. Int J Appl Math Comput Sci 1(1):19–22Google Scholar
  41. 41.
    Villmann T, Der R, Herrmann M, Martinetz TM (1997) Topology preservation in self-organizing feature maps: exact definition and measurement. IEEE Trans Neural Netw 8(2):256–266CrossRefGoogle Scholar
  42. 42.
    Yin Y, Yang G, Man H (2013) Small human group detection and event representation based on cognitive semantics. In: 2013 IEEE seventh international conference on semantic computing, pp 64–69. doi:10.1109/ICSC.2013.20.
  43. 43.
    Zhang C, Yang X, Lin W, Zhu J (2012) Recognizing human group behaviors with multi-group causalities. In: 2012 IEEE/WIC/ACM international conferences on web intelligence and intelligent agent technology, pp 44–48. doi:10.1109/WI-IAT.2012.162.

Copyright information

© The Natural Computing Applications Forum 2016

Authors and Affiliations

  1. 1.Department of Computer TechnologyUniversity of AlicanteAlicanteSpain

Personalised recommendations