Image modeling based on a 2-D stochastic subspace system identification algorithm

Fitting a causal dynamic model to an image is a fundamental problem in image processing, pattern recognition, and computer vision. In image restoration, for instance, the goal is to recover an estimate of the true image, preferably in the form of a parametric model, given an image that has been degraded by a combination of blur and additive white Gaussian noise. In texture analysis, on the other hand, a model of a particular texture image can serve as a tool for simulating texture patterns. Finally, in image enhancement one computes a model of the true image and the residuals between the image and the modeled image can be interpreted as the result of applying a de-noising filter. There are numerous other applications within the field of image processing that require a causal dynamic model. Such is the case in scene analysis, machined parts inspection, and biometric analysis, to name only a few. There are many types of causal dynamic models that have been proposed in the literature, among which the autoregressive moving average and state-space models (i.e., Kalman filter) are the most commonly used. In this paper we introduce a 2-D stochastic state-space system identification algorithm for fitting a quarter plane causal dynamic Roesser model to an image. The algorithm constructs a causal, recursive, and separable-in-denominator 2-D Kalman filter model. The algorithm is tested with three real images and the quality of the estimated images are assessed.

[1]  A. Habibi.  Two-dimensional Bayesian estimate of images , 1972 .

[2]  Paulo J. Lopes dos Santos,et al.  Subspace system identification of separable-in-denominator 2-D stochastic systems , 2011, IEEE Conference on Decision and Control and European Control Conference.

[3]  Calyampudi R. Rao.  The use and interpretation of principal component analysis in applied research , 1964 .

[4]  Cishen Zhang,et al.  H2 and mixed H2/Hinfinity control of two-dimensional systems in Roesser model , 2006, Autom..

[5]  Gene H. Golub,et al.  Matrix computations , 1983 .

[6]  Donald D. Givone,et al.  Multidimensional Linear Iterative Circuits - General Properties , 1972, IEEE Trans. Computers.

[7]  T. Kaczorek.  Two-Dimensional Linear Systems , 1985 .

[8]  B. Suresh,et al.  New results in two-dimensional Kalman filtering with applications to image restoration , 1981 .

[9]  Tao Ding,et al.  Robust Identification of 2-D Periodic Systems with Applications to Texture Synthesis and Classification , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.

[10]  Herbert Werner,et al.  Identification of spatially interconnected systems , 2009, Proceedings of the 48h IEEE Conference on Decision and Control (CDC) held jointly with 2009 28th Chinese Control Conference.

[11]  Zhe Wu,et al.  Multidimensional state-space model Kalman filtering with application to image restoration , 1985, IEEE Trans. Acoust. Speech Signal Process..

[12]  Leonard M. Silverman,et al.  Recursive estimation of two-dimensional processes with application to image processing , 1986 .

[13]  Rangasami L. Kashyap,et al.  Image data compression using autoregressive time series models , 1979, Pattern Recognit..

[14]  Mahmood R. Azimi-Sadjadi,et al.  Two-dimensional block Kalman filtering for image restoration , 1987, IEEE Trans. Acoust. Speech Signal Process..

[15]  Eric Rogers,et al.  Analysis of Linear Iterative Learning Control Schemes - A 2D Systems/Repetitive Processes Approach , 2000, Multidimens. Syst. Signal Process..

[16]  Guillaume Mercère,et al.  Subspace algorithms for identifying separable-in-denominator 2D systems with deterministic–stochastic inputs , 2016, Int. J. Control.

[17]  J. Mendel,et al.  Modeling and recursive state estimation for two-dimensional noncausal filters with applications in image restoration , 1987 .

[18]  M. Strintzis,et al.  Comments on "Two dimensional Bayesian estimate of images" , 1976 .

[19]  Bart De Moor,et al.  Subspace Identification for Linear Systems: Theory ― Implementation ― Applications , 2011 .

[20]  J. Woods,et al.  Kalman filtering in two dimensions: Further results , 1981 .

[21]  Aydin Kizilkaya,et al.  Estimation of 2-D ARMA model parameters by using equivalent AR approach , 2005, J. Frankl. Inst..

[22]  J. Y. Zhang,et al.  A very fast Kalman filter for image restoration , 1990, IEEE International Symposium on Circuits and Systems.

[23]  A.K. Jain,et al.  Advances in mathematical models for image processing , 1981, Proceedings of the IEEE.

[24]  Leonard M. Silverman,et al.  Approximation of 2-D separable in denominator filters , 1983 .

[25]  Rangaraj M. Rangayyan,et al.  An algorithm for direct computation of 2-D linear prediction coefficients , 1993, IEEE Trans. Signal Process..

[26]  Tohru Katayama,et al.  Subspace Methods for System Identification , 2005 .

[27]  R. Roesser.  A discrete state-space model for linear image processing , 1975 .

[28]  Michel Verhaegen,et al.  Filtering and System Identification: Frontmatter , 2007 .

[29]  N.K. Bose,et al.  Problems and progress in multidimensional systems theory , 1977, Proceedings of the IEEE.

[30]  Ahmed Alenany,et al.  Subspace algorithms for identifying separable-in-denominator two-dimensional systems with deterministic inputs , 2011 .

[31]  Victor Sreeram,et al.  Identification and model reduction of 2-D systems via the extended impulse response Gramians , 1998, Autom..

[32]  Christian Olivier,et al.  Choice of a 2-D causal autoregressive texture model using information criteria , 2003, Pattern Recognit. Lett..

[33]  J. Kurek,et al.  The general state-space model for a two-dimensional linear digital system , 1985 .

[34]  J. Magnus,et al.  Matrix Differential Calculus with Applications in Statistics and Econometrics (Revised Edition) , 1999 .

[35]  Gene H. Golub,et al.  Matrix computations (3rd ed.) , 1996 .

[36]  V. Verdult,et al.  Filtering and System Identification: A Least Squares Approach , 2007 .

[37]  Ahmet H. Kayran,et al.  New Efficient 2-D Lattice Structures for General Autoregressive Modeling of Random Fields , 2014, IEEE Transactions on Signal Processing.

[38]  J. Kurek,et al.  Iterative learning control synthesis based on 2-D system theory , 1993, IEEE Trans. Autom. Control..

[39]  Giorgio Picci,et al.  Realization of stochastic systems with exogenous inputs and subspace identification methods , 1999, Autom..

[40]  M. Kosaka,et al.  Recursive filtering algorithm for a two-dimensional system , 1979 .

[41]  Jason Sheng-Hong Tsai,et al.  Modeling of Variable Coefficient Roesser Model for Systems Described by Second-Order Partial Differential Equation , 2003 .

[42]  M. Verhaegen,et al.  A spatial canonical approach to multidimensional state-space identification for distributed parameter systems , 2005, The Fourth International Workshop on Multidimensional Systems, 2005. NDS 2005..

[43]  Eero P. Simoncelli,et al.  Image quality assessment: from error visibility to structural similarity , 2004, IEEE Transactions on Image Processing.

[44]  Michel Verhaegen,et al.  Identification of the deterministic part of MIMO state space models given in innovations form from input-output data , 1994, Autom..

[45]  S. Attasi.  Modelling and Recursive Estimation for Double Indexed Sequences , 1976 .

[46]  M. Moonen,et al.  On- and off-line identification of linear state-space models , 1989 .

[47]  M. Morf,et al.  New results in 2-D systems theory, part II: 2-D state-space models—Realization and the notions of controllability, observability, and minimality , 1977, Proceedings of the IEEE.

[48]  J. Ramos,et al.  A subspace algorithm for identifying 2-D separable in denominator filters , 1994 .

[49]  Masayuki Kawamata,et al.  Decomposition of 2-D separable-denominator systems: Existence, uniqueness, and applications , 1987 .

[50]  Rama Chellappa,et al.  A system identification approach for video-based face recognition , 2004, ICPR 2004.

[51]  Ahmet H. Kayran,et al.  Spectrum estimation using new efficient 2-D AR lattice modeling of random fields , 2013, 2013 Africon.

[52]  Ettore Fornasini,et al.  Doubly-indexed dynamical systems: State-space models and structural properties , 1978, Mathematical systems theory.

[53]  Krzysztof Galkowski.  Higher order discretization of 2-D systems , 2000 .