The combination of IR and visible imaging in the problems of identification of mobile ground targets on board UAVs

Technical cybernetics. Information technology. Computer facilities


Kazbekov B. V.

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



Algorithms and methods for automated digital image processing in the visible and infrared ranges obtained from the board of unmanned aerial vehicles (UAVs) searching for moving ground targets.
The superposition of infrared images with images in the visible range in the tasks of identification of moving ground targets from UAV`s board.
Purpose of the paper is development of complex algorithm of moving ground targets identification from UAV`s board. To achieve this purpose it is necessary to develop an algorithm that can operate under the condition of incomplete a priori information about the object of identification.
The main features of the algorithm are:

  • universality of the algorithm (ability to recognize different classes of mobile ground targets);
  • stability against affine corruption of image;
  • operation in the near real time mode.

Image processing received from the UAV board becomes complicated by the following factors:

  • onboard UAV video probe moves in space with the UAV, moreover, its spatial orientation can be changed by an operator with the help of a positioning device;
  • video probe is affected by the vibration impact of different nature;
  • environmental influence (rain, snow, fog, wind, etc.).

In this study the author used methods of mathematical modeling, computer vision, and the theory of probability and mathematical statistics.
The results can be implemented in solution of tasks of mobile ground targets identification. These tasks arise when UAVs are used for military purposes: for reconnaissance, for observation of a battlefield, for monitoring results of enemy toll under destruction fire, etc.
Usage of combined analysis of IR images can reduce almost 3 times the number of false alarms what was confirmed experimentally. It is also important to note that at this stage of research combined joint analysis of the infrared image along with image in the visible range is not aimed at the object type definition (it is defined whether an object moves or not). This approach allows for quick and efficient search of moving objects from the UAVs` board without detailed analysis of the object itself.


pattern recognition, key features of the image, IR images, identification of moving targets


  1. Hörster E., Lienhart R., Slaney M. Proc. 6th ACM int. conf. on Image and Video Retrieval, Canada, 2007, pp. 17–24.
  2. Potapov А.S. Izvestiya vuzov. Priborostroenie, Moscow, 2008, no. 51, pp. 3-7.
  3. Marr D. Zrenie. Informatsionnyi podkhod k izucheniyu predstavleniya i obrabotki zritel'nykh obrazov (Vision. Information approach to the study of representation and processing of visual images), Moscow, Radio i svjaz', 1987, 400 p.
  4. Chan T.F., Shen J., Vese L. Variational PDE models in image processing, Notice Amer. Math. Soc., 2003, vol. 50, pp. 14–26.
  5. Lei B.J., Hendriks E.A., Reinders M.J.T. On feature extraction from images, Technical Report, Deliverable, MCCWS project, Information and Communication Theory Group. TU Delft, 1999, 57 p.
  6. Baker S. Design and Evaluation of Feature Detectors:PhD thesis. Columbia University, 1998, 167 p.
  7. Furman Ya.A. Vvedenie v konturnyi analiz i ego prilozheniya k obrabotke izobrazhenii i signalov (Introduction to circuit analysis and its applications to image and signal processing), Moscow, Fizmatlit, 2002, 592 p.
  8. Brown L.G. A survey of image registration techniques, ACM Computing surveys, 1992. vol. 24, pp. 325–376.
  9. Lutsiv V.R., Malyshev I.A., Pepelka V., Potapov A.S. Target independent algorithms for description and structural matching of aerospace photographs, Proc. SPIE, 2002, vol. 4741, pp. 351–362.
  10. Nacken P. Image analysis methods based on hierarchies of graphs and multi-scale mathematical morphology: PhD thesis, Univ. of Amsterdam, 1994, 176 p.
  11. Rozenfel'd A., Deivis L.S. Segmentation and image model. IEEE, 1979, no. 5, pp. 71–81.
  12. Abdu I.E., Prett. Quantative calculation detector circuits, based on highlighting the differences of brightness with the subsequent thesholding. IEEE, 1979, no. 5, pp. 59–70.
  13. Roberts L. Avtomaticheskoe vospriyatie trekhmernykh stsen (Automatic perception of three-dimensional scenes), Moscow, MIR, 1973, 421 p.
  14. Prewitt J.M.S. Picture Processing and Psychopictorics, New York, Academic Press, 1970, pp. 75–149.
  15. Duda R.,Khart P. Raspoznavanie obrazov i analiz stsen (Recognition and scene analysis), Moscow, MIR, 1976, 511 p.
  16. Canny J.F. A computational approach to edge detection, IEEE Trans. Pattern Analysis and Machine Intelligence. 1986, vol. 8, no. 6, pp. 679–698.
  17. Deriche R. Optimal edge detection using recursive filtering, Proc. 1st Int. Conf. Computer Vision, 1987,pp. 501–505.
  18. Lindeberg T. Edge detection and ridge detection with automatic scale selection, Int. J. of Computer Vision, 1998, vol. 30, pp. 117–156.
  19. Park R.-H., Yoon K.S., Choi W.Y. Eight-point discrete Hartley transform as an edge operator and its interpretation in the frequency domain, Pattern Recognition Letters, 1998, vol. 19. pp. 569–574.
  20. Chanda B., Kundu M.K., Padmaja Y.V. A multi-scale morphologic edge detector, Pattern Recognition, 1998, vol. 31, no. 10, pp. 1469–1478.
  21. Lutsiv V., Malyshev I., Potapov A. Hierarchical structural matching algorithms for registration of aerospace images, Proc. SPIE, 2003 vol. 5238, pp. 164–175.
  22. Thomas P. and Vernon D. Image registration by differential evolution, Proc. Irish Machine Vision and Image Processing Conference, 1997, pp. 221–225.
  23. Jerebko A., Barabanov N., Luciv V., Allinson N. Neural net based image matching, Proc. SPIE, 2000, vol. 3962, pp. 128–137.
  24. Thevenaz P. et al. A pyramid approach to subpixel registration based on intensity, IEEE Trans. Image Processing, 1998, vol. 7, no. 1, pp. 27–41.
  25. Olson C.F. Improving the generalized Hough transform through imperfect grouping, Image and Vision Computing, 1998, vol. 16. P. 627–634.
  26. Castleman K.R. Digital Image Processing: New York, Prentice Hall Press, 1996, 667 p.
  27. Petrou M., Kadyrov A. Affine invariant features from the trace transform, IEEE Trans. on Pattern Analysis and Machine Intelligence. 2004.vol. 26. № 1. P. 30–44.
  28. Lowe D. Object recognition from local scale-invariant features, Proc. Int. Conf. on Computer Vision, 1999, pp.1150–1157.
  29. Razin I.V., Teterin V.V. Opticheskii zhurnal, 2001, no. 11, pp. 33–37.
  30. Kruizinga P., Petkov N. Nonlinear operator for oriented texture, IEEE Trans. on Image Processing, 1999,vol. 8, no.1, pp. 1395–1407.
  31. Portilla J., Simoncelli E.P. A parametric texture model based on joint statistics of complex wavelet coefficients, Int. J. of Computer Vision, 2000, vol. 40, no.1,pp.49–71.
  32. Baumberg A. Reliable feature matching across widely separated views, Conf. on Computer Vision and Pattern Recognition, 2000, pp. 774–781.
  33. Rao C., Guo Y., Sawhney H.S., Kumar R. A heterogeneous feature-based image alignment method, Int. Conf. on Pattern Recognition, ICVR06, 2006, pp. 345–350.
  34. Zavorin I., LeMoigne J. Use of multiresolution wavelet feature pyramids for automatic registration of multisensor imagery, IEEE Trans. Image Processing, 2005, vol.14, no.6, pp. 770–782
  35. Kazbekov B.V., Maksimov N.A., Purtov I.S., Sincha D.P. Nauchno-tekhnicheskii vestnik Povolzh'ya, 2011, no. 5, pp. 20-26.

Download — informational site MAI

Copyright © 2000-2020 by MAI