Robust multivariate density estimation under Gaussian noise

Observation of random variables is often corrupted by additive Gaussian noise. Noise-reducing data processing is time-consuming and may introduce unwanted artifacts. In this paper, a novel approach to description of random variables insensitive with respect to Gaussian noise is presented. The proposed quantities represent the probability density function of the variable to be observed, while noise estimation, deconvolution or denoising are avoided. Projection operators are constructed, that divide the probability density function into a non-Gaussian and a Gaussian part. The Gaussian part is subsequently removed by modifying the characteristic function to ensure the invariance. The descriptors are based on the moments of the probability density function of the noisy random variable. The invariance property and the performance of the proposed method are demonstrated on real image data.

[1]  Jan Flusser,et al.  Degraded Image Analysis: An Invariant Approach , 1998, IEEE Trans. Pattern Anal. Mach. Intell..

[2]  F. Comte,et al.  Adaptive estimation of linear functionals in the convolution model and applications , 2009, 0902.1443.

[3]  Emmanuel J. Candès,et al.  The curvelet transform for image denoising , 2002, IEEE Trans. Image Process..

[4]  James R. Schott,et al.  Kronecker product permutation matrices and their application to moment matrices of the normal distribution , 2003 .

[5]  Gabriele Steidl,et al.  Denoising by second order statistics , 2012, Signal Process..

[6]  P. Hall,et al.  Optimal Rates of Convergence for Deconvolving a Density , 1988 .

[7]  Dietrich von Rosen,et al.  Moments for matrix normal variables , 1988 .

[8]  Ramin Zabih,et al.  Histogram refinement for content-based image retrieval , 1996, Proceedings Third IEEE Workshop on Applications of Computer Vision. WACV'96.

[9]  Claire Lacour,et al.  Data driven density estimation in presence of unknown convolution operator , 2011 .

[10]  Johanna Kappus,et al.  Adaptive density estimation in deconvolution problems with unknown error distribution , 2013 .

[11]  Majid Ahmadi,et al.  Wavelet-Domain Blur Invariants for Image Analysis , 2012, IEEE Transactions on Image Processing.

[12]  William Puech,et al.  Digital image restoration by Wiener filter in 2D case , 2007, Adv. Eng. Softw..

[13]  Ville Ojansivu,et al.  Image Registration Using Blur-Invariant Phase Correlation , 2007, IEEE Signal Processing Letters.

[14]  Jianqing Fan,et al.  Deconvolution with supersmooth distributions , 1992 .

[15]  W. Bar,et al.  Useful formula for moment computation of normal random variables with nonzero means , 1971 .

[16]  Jean-Michel Morel,et al.  A Review of Image Denoising Algorithms, with a New One , 2005, Multiscale Model. Simul..

[17]  H. W. Gould,et al.  Double Fun with Double Factorials , 2012 .

[18]  R. Carroll,et al.  Deconvolving kernel density estimators , 1987 .

[19]  P. Lions,et al.  Image recovery via total variation minimization and related problems , 1997 .

[20]  Iickho Song,et al.  Explicit formulae for product moments of multivariate Gaussian random variables , 2015 .

[21]  Glenn Healey,et al.  Using Zernike moments for the illumination and geometry invariant classification of multispectral texture , 1998, IEEE Trans. Image Process..

[22]  Huazhong Shu,et al.  Blurred Image Recognition by Legendre Moment Invariants , 2010, IEEE Transactions on Image Processing.

[23]  Kostas Triantafyllopoulos,et al.  On the central moments of the multidimensional Gaussian distribution , 2003 .

[24]  Claire Lacour,et al.  Data‐driven density estimation in the presence of additive noise with unknown distribution , 2011 .

[25]  Jan Flusser,et al.  Blur Invariant Translational Image Registration for $N$-fold Symmetric Blurs , 2013, IEEE Transactions on Image Processing.

[26]  IV CyrilHöschl,et al.  Robust histogram-based image retrieval , 2016, Pattern Recognit. Lett..

[27]  J. Johannes.  DECONVOLUTION WITH UNKNOWN ERROR DISTRIBUTION , 2007, 0705.3482.

[28]  Rama Chellappa,et al.  A Blur-Robust Descriptor with Applications to Face Recognition , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[29]  Jean-Michel Morel,et al.  A non-local algorithm for image denoising , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[30]  Jitendra Malik,et al.  Scale-Space and Edge Detection Using Anisotropic Diffusion , 1990, IEEE Trans. Pattern Anal. Mach. Intell..

[32]  Tien D. Bui,et al.  Multivariate statistical approach for image denoising , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[33]  Jan Flusser,et al.  Moment Forms Invariant to Rotation and Blur in Arbitrary Number of Dimensions , 2003, IEEE Trans. Pattern Anal. Mach. Intell..

[34]  A. Meister.  Deconvolution Problems in Nonparametric Statistics , 2009 .

[35]  R. Blacher.  Multivariate quadratic forms of random vectors , 2003 .

[36]  B. Vidakovic,et al.  Adaptive wavelet estimator for nonparametric density deconvolution , 1999 .

[37]  L. Isserlis.  ON A FORMULA FOR THE PRODUCT-MOMENT COEFFICIENT OF ANY ORDER OF A NORMAL FREQUENCY DISTRIBUTION IN ANY NUMBER OF VARIABLES , 1918 .

[38]  Guangyi Chen,et al.  Wavelet-based denoising: A brief review , 2013, 2013 Fourth International Conference on Intelligent Control and Information Processing (ICICIP).

[39]  Rakhi C. Motwani,et al.  Survey of Image Denoising Techniques , 2004 .

[40]  Michael J. Swain,et al.  Color indexing , 1991, International Journal of Computer Vision.

[41]  Jan Flusser,et al.  Projection Operators and Moment Invariants to Image Blurring , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[42]  Bart De Moor,et al.  Deconvolution in nonparametric statistics , 2012, ESANN.