Title :
A maximum entropy Kalman filter for signal reconstruction
Author :
David, A. ; Aboulnasr, T.
Author_Institution :
Sch. of Inf. Technol. & Eng., Ottawa Univ., Ont., Canada
Abstract :
In this paper, we propose a Maximum Entropy Kalman Filter (MEKF) and its application in image recovery. The proposed 2-D MEKF employs Maximum Entropy (ME) as its optimization criterion to identify the appropriate parameters of a standard Kalman filter. The strength of the ME based filters is due to the fact that these filters make no assumptions regarding the unobserved data, and avoid the over-smoothing that is associated with the Mean Square Error (MSE) based algorithms. Furthermore, we address the issues of ME 2-D separable filter expansion and the finite constraint bound on the reconstructed pixels
Keywords :
Kalman filters; filtering theory; image reconstruction; matrix algebra; maximum entropy methods; optimisation; 2D separable filter expansion; finite constraint bound; image recovery; maximum entropy Kalman filter; optimization criterion; reconstructed pixels; signal reconstruction; Autocorrelation; Cost function; Entropy; Filtering; Filters; Laboratories; Optimization methods; Signal processing; Signal reconstruction; Stochastic processes;
Conference_Titel :
Circuits and Systems, 1999. ISCAS '99. Proceedings of the 1999 IEEE International Symposium on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-5471-0
DOI :
10.1109/ISCAS.1999.779964
