DocumentCode :
3795848
Title :
On a parameter estimation method for Gibbs-Markov random fields
Author :
M.I. Gurelli;L. Onural
Author_Institution :
Dept. of Electr. & Electron. Eng., Bilkent Univ., Ankara, Turkey
Volume :
16
Issue :
4
fYear :
1994
Firstpage :
424
Lastpage :
430
Abstract :
This correspondence is about a Gibbs-Markov random field (GMRF) parameter estimation technique proposed by Derin and Elliott (1987). We will refer to this technique as the histogramming (H) method. First, the relation of the H method to the (conditional) maximum likelihood method is considered. Second, a bias-reduction based modification of the H method is proposed to improve its performance, especially in the case of small amounts of image data.
Keywords :
"Parameter estimation","Pattern recognition","Sonar","Error analysis","Convergence","Eigenvalues and eigenfunctions","Iterative algorithms","Timing","Multidimensional systems","Vectors"
Journal_Title :
IEEE Transactions on Pattern Analysis and Machine Intelligence
Publisher :
ieee
ISSN :
0162-8828
Type :
jour
DOI :
10.1109/34.277597
Filename :
277597
Link To Document :
بازگشت