Title :
A finite element approach to local nonlinear image restoration
Author :
Veldhuizen, Todd L. ; Jernigen, M.E.
Author_Institution :
Dept. of Syst. Design Eng., Waterloo Univ., Ont., Canada
Abstract :
We consider finite element basis functions for local nonlinear image restoration. The high dimensionality of the problem requires symmetry assumptions to make the filter design process computationally feasible. In contrast to polynomial filters, finite element filters are fast to evaluate and train, and handle outliers well. Preliminary results are superior to Lee´s local adaptive filter
Keywords :
filtering theory; finite element analysis; image restoration; nonlinear filters; piecewise-linear techniques; filter design process; finite element basis functions; finite element filters; local nonlinear image restoration; symmetry assumptions; Adaptive filters; Degradation; Design engineering; Finite element methods; Image restoration; Nonlinear filters; Pixel; Polynomials; Systems engineering and theory; Wiener filter;
Conference_Titel :
Electrical and Computer Engineering, 1997. Engineering Innovation: Voyage of Discovery. IEEE 1997 Canadian Conference on
Conference_Location :
St. Johns, Nfld.
Print_ISBN :
0-7803-3716-6
DOI :
10.1109/CCECE.1997.608268
