Radiotekhnika
Publishing house Radiotekhnika

"Publishing house Radiotekhnika":
scientific and technical literature.
Books and journals of publishing houses: IPRZHR, RS-PRESS, SCIENCE-PRESS


Тел.: +7 (495) 625-9241

 

Method for moving object area identification in image sequence

Keywords:

P.V. Smirnov – Post-graduate Student, Ulyanovsk State Technical University A.G. Tashlinskii – Dr. Sc. (Eng.), Professor, Head of Department «Radio Engineering», Ulyanovsk State Technical University


The paper carries out a comparative analysis of various approaches to identify area of moving object in image sequence using pixel-by-pixel estimation of inter-frame shifts of all nodes of the reference image. It studies projections and polar parameters of shift vectors of reference image points corresponding to the nodes as their estimated parameters. The paper proposes the method for estimating shift vectors’ field. The method finds two sets of shift estimates of all nodes of the reference image using stochastic gradient descent algorithm and then processes them jointly. The joint processing allows compensating inertia of the recursive estimation. The paper studies two methods to find a set of shift estimates. The first method uses stochastic gradient descent algorithm to sequentially process all nodes of the image row-by-row. It processes each row bidirectionally (from the left to the right and from the right to the left). However this method does not consider correlation between adjacent rows and processes images as one-dimensional signals. Therefore the second method uses the correlation between adjacent rows to increase the efficiency of moving object area identification. The method processes rows one after the other with change in direction after each row and performs the joint processing of the adjacent rows’ estimates obtained in opposite directions. This method shows significantly better results. The joint processing of sets of shift estimates uses correlation coefficient maximum as a criterion of shift vectors’ field formation. It provides not only low mean value and variance of estimation error both in motion area and outside it but the oscillations are also low.
References:

 

  1. Elhabian Sh.Y., El-Sayed Kh.M., Ahmed S.H. Moving Object Detection in Spatial Domain using Background Removal Techniques // Recent Patents on Computer Science. 2008. V. 1. P. 32−54.
  2. Karasulu B., Korukoglu S. Performance Evaluation Software: Moving Object Detection and Tracking in Videos // SpringerBriefs in Computer Science. 2013. P. 76.
  3. Wang L., Yung N.H.C. Extraction of Moving objects from their Background based on mulitple adaptive threshold and boundary evaluation // IEEE Trans. Intelligenttransportationsystems. 2010. V. 11. P. 40−51.
  4. Kucov R.V., Trifonov A.P. Algoritmy obnaruzhenija dvizhushhegosja obekta na izobrazhenii // Izvestija RAN. Teorija i sistemy upravlenija. 2006. № 3. S. 129−138.
  5. Grishin S.V., Vatolin D.S., Lukin A.S. Obzor blochnykh metodov ocenki dvizhenija v cifrovykh video signalakh // Tematicheskijj sb. «Programmnye sistemy i instrumenty». 2008. T. 9. S. 50−62.
  6. Tashlinskijj A.G., Smirnov P.V. Algoritm kompensacii ehffekta smaza izobrazhenija dvizhushhegosja obekta po posledovatelnosti kadrov // Radiotekhnika. 2014. № 7. S. 81−87.
  7. Zolotykh N.JU., Kustikova V.D., Meerov I.B. Obzor metodov poiska i soprovozhdenija transportnykh sredstv na potoke videodannykh // Vestnik Nizhegorodskogo universiteta im. N.I. Lobachevskogo. 2012. № 5(2). S. 348−358.
  8. Tashlinskijj A.G. Ocenivanie parametrov prostranstvennykh deformacijj posledovatelnostejj. Uljanovsk: Izd‑vo UlGTU. 2000. 132 s.
  9. Tashlinskii A.G. Computational expenditure reduction in pseudo-gradient image parameter estimation // Lecture Notes in Computer Science. 2003. V. 2658. P. 456−462.
  10. Tashlinskii A.G.,Kurbanaliev R.M., Zhukov S.S. Method for detecting instability and recovery of signal shape under intense noise // Pattern recognition and image analysis. 2013. V. 23. № 3. P. 425−428.
  11. Tashlinskii A.G.The Specifics of Pseudogradient Estimation of Geometric Deformations in Image Sequences // Pattern Recognition and Image Analysis. 2008. V. 18. № 4. P. 701−706.
  12. Tashlinskijj A.G., KHoreva A.M., Smirnov P.V. Vybor konechnykh raznostejj pri nakhozhdenii psevdogradienta celevojj funkcii v procedurakh ocenivanija mezhkadrovykh deformacijj izobrazhenijj // Radiotekhnika. 2012. № 9. S. 56−60.
  13. Tashlinskijj A.G., Smirnov P.V. Popikselnoe ocenivanie mezhkadrovykh geometricheskikh deformacijj izobrazhenijj pri vydelenii oblasti podvizhnogo obekta // Avtomatizacija processov upravlenija. 2015. № 1(39). S. 41−49.
  14. Tashlinskijj A.G., Levchukov D.A. Ocenivanie parametrov mezhkadrovykh geometricheskikh deformacijj izobrazhenijj po matrice deformacijj // Kontinualnye algebraicheskie logiki i nejjroinformatika v nauke i tekhnike: trudy mezhdun. konf. Uljanovsk. 2005. T. 2. S. 149.
  15. Kaveev I.N., Tashlinskijj A.G. Ocenivanie parametrov affinnojj modeli privjazki izobrazhenijj po soprjazhennym tochkam // Informatika, modelirovanie, avtomatizacija proektirovanija: sb. nauchnykh trudov. Uljanovsk: UlGTU. 2009. S. 109−111.

 

© Издательство «РАДИОТЕХНИКА», 2004-2017            Тел.: (495) 625-9241                   Designed by [SWAP]Studio