SciELO - Scientific Electronic Library Online

vol.14 issue4Model Based Testing for Workflow Enabled ApplicationsTrajectory Optimization for Systems Under Nonholonomic Constraints author indexsubject indexsearch form
Home Pagealphabetic serial listing  

Services on Demand




Related links

  • Have no similar articlesSimilars in SciELO


Computación y Sistemas

Print version ISSN 1405-5546

Comp. y Sist. vol.14 n.4 México Apr./Jun. 2011




Una estrategia para la selección dinámica de características aplicada a la estabilización de secuencias de imágenes


An Strategy for the Dynamic Selection of Features Applied to the Stabilization of Image Sequences


Hugo Jiménez Hernández1 y Joaquín Salas Rodríguez2


Centro de Investigación en Ciencia Aplicada y Tecnología Avanzada del IPN;


Artículo recibido en Marzo 10, 2010
Aceptado en Junio 11, 2010



En este trabajo se presenta una estrategia para la discriminación entre las características pertenecientes a objetos fijos y móviles de una escena observada desde una cámara sujeta a vibración. Nuestra estrategia selecciona como características fijas aquellas que minimizan el error de la proyección de la homográfica entre las imágenes, siendo tolerante a oclusiones de regiones y cambios lumínicos en la escena. Una posible aplicación de este resultado es la estabilización de secuencias de imágenes. En una etapa experimental, utilizando distintos escenarios en exteriores, mostramos los resultados y evidencia que los niveles de precisión obtenidos son mejores, que los obtenidos por propuestas eficientes basadas en la selección aleatoria de características, tal como la de RANSAC.

Palabras clave: Selección dinámica de características, estimación de la homografía, método no supervisado.



This work introduces an algorithm to discriminate between either moving or static features of a given scene as observed from a fixed camera, which is under the effects of vibration,. In our strategy, we select the features minimizing the registration error from one image to the next one. The process discards the features corresponding to moving objects and untrackable regions. The algorithm is applied to the task of stabilizing an image sequence. In our experiments, we benchmark our approach with several images sequences and match the results with a randomized strategy known as RANSAC.

Key words: Dynamic selection of features, estimation of the copying, unsupervised method.





1. Anderson, M. P. & Woessner, W. W. (1992). Applied Groundwater Modeling: Simulation of Flow and Advective Transport, San Diego, California, Academic Press.         [ Links ]

2. Collins, R. T. & Lui, Y. (2003). On–Line Selection of Discriminative Tracking Features, Ninth IEEE International Conference on Computer Vision, Nice, France, 346–352.         [ Links ]

3. Cham, T. & Rehg, J. (1999). Dynamic Feature Ordering for Efficient Registration. Seventh IEEE International Conference on Computer Vision, 2, 1084–1091.         [ Links ]

4. Chum, O. & Matas, J. (2008). Optimal Randomized RANSAC. IEEE Transactions on Pattern Analysis and Machine Intelligence, 30(8), 1472–1482.         [ Links ]

5. Dempster, A., Laird, N. & Rubin, D. (1997). Maximum Likelihood from Incomplete Data via the EM Algorithm. Journal of the Royal Statistical Society, vol. 1, 1–38.         [ Links ]

6. Eadie, W., [et. all] (1971). Statistical Methods in Experimental Physics. Amsterdam: North–Holland.         [ Links ]

7. Fischler, M. & Bolles, R. (1981). Random Sample Consensus: A Paradigm for Model Fitting with Applications to Image Analysis and Automated Cartography. Communications of the ACM, 24(6), 381 –395.         [ Links ]

8. Fleuret, F. (2004). Fast Binary Feature Selection with Conditional Mutual Information. Journal of Machine Learning Research, 5, 1531–1555.         [ Links ]

9. Gil, S. Milanese, R. & Pun, T. (1994). Feature Selection for Object Tracking in Traffic Scenes. (tr–94–060) International Computer Science Institute, California, USA.         [ Links ]

10. Hwann–Tzong, Ch., Tyng–Luh, L., & Chiou–Shann F. (2004). Probabilistic Tracking with Adaptive Feature Selection. 17th International Conference on Pattern Recognition, 2, 736–739.         [ Links ]

11. Jianbo, S. & Tomasi, C. (1994). Good Features to Track. IEEE Computer Society Conference on Computer Vision and Pattern Recognition, Seattle, Washington, USA, 593–600.         [ Links ]

12. Julier, S. J. & Uhlmann, J. K. (1997). A New Extension of the Kalman Filter to Nonlinear Systems. 11th International Symposium on Aerospace/Defense Sensing, Simulation and Controls Orlando,Florida, USA, 182–193.         [ Links ]

13. Junlan, Y.J., Schonfeld, D., Chong, Ch.. & Mohamed, M. (2006). Online Video Stabilization Based on Particle Filters. IEEE International Conference on Image Processing, Atlanta, Georgia, USA, 1545–1548.         [ Links ]

14. Kalman, R. E. (1960) A New Approach to Linear Filtering and Prediction Problems. Transactions of the ASME Journal of Basic Engineering, 82 (Series D), 35–45.         [ Links ]

15. Kariya, T. & Kurata, H. (2004). Generalized Least Squares, Hoboken, NJ: Wiley.         [ Links ]

16. Knot J., Jurisica L. & Hubinsky P. (2006). Feature Based Object Tracking for Oscillation Detection. 16th International Conference Radioelectronika 2006, 316319.         [ Links ]

17. Lacey, A.J., Pinitkarn, N. & Thacker, N.A (2000). An Evaluation of the Performance of RANSAC Algorithms for Stereo Camera Calibrarion. 11th BritishMachine Vision Conference, Bristol U.K.         [ Links ]

18. Lucas, B.D. & Kanade, T. (1981). An Iterative Image Registration Technique with an Application to Stereo Vision. DARPA Image Understanding Workshop, San Francisco, CA, USA 121–130.         [ Links ]

19. Matsushita, Y., Ofek, E., Weina, G., Xiaoou, T. & Heung–Yeung, S. (2006). Full–Frame Video Stabilization with Motion Inpainting. IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 28(7), 1150 –1163.         [ Links ]

20. Mikolajczyk, K. & Schmid, C. (2005). A performance evaluation of local descriptors. IEEE Transactions on Pattern Analysis and Machine Intelligence, 27(10),1615–1630.         [ Links ]

21. Salas, J., Jiménez, H., Gonzalez, J. & Hurtado, J. (2007). Detecting Unusual Activities at Vehicular Intersections". IEEE International Conference on Robotics and Automation, Roma, Italy, 864–869.         [ Links ]

22. Tan, Y., Kulkarni, S. & Ramadge, P. (1996). Extracting Good Features for Motion Stimation, International Image Proccesing, Lausanne, Switzerland, 117 – 120.         [ Links ]

23. Tang, F. & Tao, H. (2005). Object Tracking with Dynamic Feature Graph, 2nd Joint IEEE International Workshop on Visual Surveillance and Performance Evaluation of Tracking and Surveillance, Beijin, China25–32.         [ Links ]

24. Thomas, G. B. & Finney, R. L. (1992). Calculus and Analytic Geometry. (8th Ed.). Reading Mass: Addison Wesley.         [ Links ]

25. Tomasi C. (1993). Input Redundancy and Output Observability in the Analysis of Visual Motion. Sixth International Symposium on Robotics Research, Ithaca, NY, EUA, 213–222.         [ Links ]

26. Tomasi, C. (2004). Cs 296.1Mathematical Modelling of Continuos Systems. Durham, NC:Duke University.         [ Links ]

27. Tomasi, C. & Manduchi, R. (1998). Stereo Matching as a Nearest–Neighbor Problem. IEEE Transactions on Pattern Analysis and Machine Intelligence,20(3), 333–340.         [ Links ]

28. Trucco, E. & Verri, A. (1998). Introductory Techniques for 3–D Computer Vision. Upper Saddle River, NJ: Prentice Hall.         [ Links ]

29. Wald, A. (1945). Sequential Tests of Statistical Hypotheses. Annals of Mathematical Statistics, 16(2), 117–186.         [ Links ]

30. ZuWhan, K. (2008). Real Time Object Tracking Based on Dynamic Feature Grouping with Background Subtraction. IEEE Conference on Computer Vision and Pattern Recognition, Anchorage, Alaska, USA 1–8.         [ Links ]

Creative Commons License All the contents of this journal, except where otherwise noted, is licensed under a Creative Commons Attribution License