A comparative study of new HOS-based estimators for moving objects in noisy video sequence

The need for motion estimation (ME) arises quite often in many areas such as computer vision, target tracking, medical imaging, robotic vision. A five new estimators for frame-to-frame image ME are described in this paper. The new ME estimators exploit the higher-order statistics (HOS) characteristics of the received images, and various frequency weighting functions are used to prefilter the received images before calculating the generalized cross-cumulant function and, therefore, suppress the Gaussian noise effect. The estimators of interest are the HOS-ROTH impulse response, the HOS-phase transform, the HOS-smoothed coherence transform, the HOS-maximum likelihood and the HOS-Wiener estimators. Since the performances of the HOS-based estimators are considerably degraded by the signal-to-noise ratio level, this factor has been taken as a prime factor in benchmarking the different estimators. For robust ME it has been found that the HOS-Wiener estimator is particularly suited to this purpose. The accuracy of the estimators is also discussed.

[1]  Jean-Marc Le Caillec,et al.  Comparison of statistical indices using third order statistics for nonlinearity detection , 2004, Signal Process..

[2]  A. Enis Çetin,et al.  An iterative algorithm for signal reconstruction from bispectrum , 1991, IEEE Trans. Signal Process..

[3]  Vania V. Estrela,et al.  Pel-Recursive Motion Estimation Using the Expectation-Maximization Technique and Spatial Adaptation , 2004, WSCG.

[4]  El Mehdi Ismaili Alaoui,et al.  A Robust hierarchical motion estimation algorithm in noisy image sequences in the bispectrum domain , 2009, Signal Image Video Process..

[5]  Nader M. Namazi,et al.  Nonuniform image motion estimation from noisy data , 1990, IEEE Trans. Acoust. Speech Signal Process..

[7]  B. Boashash,et al.  Pattern recognition using invariants defined from higher order spectra: 2-D image inputs , 1997, IEEE Trans. Image Process..

[8]  K. S. Lii,et al.  Cross-Bispectrum Computation and Variance Estimation , 1981, TOMS.

[9]  Horia Cucu,et al.  Applying the Accumulation of Cross-Power Spectrum Technique for Traditional Generalized Cross-Correlation Time Delay Estimation , 2013 .

[10]  Bernd Girod,et al.  Motion Analysis and Image Sequence Processing , 1993 .

[11]  Janne Heikkilä.  Image scale and rotation from the phase-only bispectrum , 2004, ICIP.

[12]  Javier Rodríguez Fonollosa,et al.  Motion estimation using higher order statistics , 1996, IEEE Trans. Image Process..

[13]  Georgios B. Giannakis,et al.  Noise insensitive image motion estimation using cumulants , 1991, [Proceedings] ICASSP 91: 1991 International Conference on Acoustics, Speech, and Signal Processing.

[14]  Alan C. Bovik,et al.  The Essential Guide to Image Processing , 2009, J. Electronic Imaging.

[15]  Georgios B. Giannakis,et al.  Image motion estimation algorithms using cumulants , 1995, IEEE Trans. Image Process..