Title :
Cosine transform based preconditioners for total variation deblurring
Author :
Chan, Raymond H. ; Chan, Tony F. ; Wong, Chiu-Kwong
Author_Institution :
Dept. of Math., Chinese Univ. of Hong Kong, Shatin, Hong Kong
fDate :
10/1/1999 12:00:00 AM
Abstract :
In PDE image restoration problems, one has to invert operators which is a sum of a blurring operator and an elliptic operator with highly varying coefficient. We present a preconditioner for such operators, which can be used with the conjugate gradient (CG) method, and compare it with Vogel and Oman´s (see SIAM J. Sci. Stat. Comput., vol.17, p.227-38, 1996, and IEEE Trans. Image Processing, vol.7, p.813-24, 1998) product preconditioner
Keywords :
conjugate gradient methods; discrete cosine transforms; image restoration; optimisation; PDE image restoration problems; blurring operator; conjugate gradient method; cosine transform based preconditioners; elliptic operator; general matrix; optimal cosine transform approximation; product preconditioner; total variation deblurring; varying coefficient; Anisotropic magnetoresistance; Character generation; Circuits; Image edge detection; Image processing; Image restoration; Laplace equations; Mathematics; Neural networks; Signal processing;
Journal_Title :
Image Processing, IEEE Transactions on
