Title :
Multiscale pursuit scheme for image approximation
Author :
He, Yanmin ; Gan, Tao ; Wang, Houjun
Author_Institution :
Sch. of Autom. Eng., Univ. of Electron. Sci. & Technol., Chengdu
Abstract :
In this paper, a new multiscale matching pursuit algorithm for image approximation is presented. The dictionary consisting of geometric atoms is used to provide the approximation ability. To speed up the matching process, the pursuit is performed at different image scales. Experimental results show that compared to the latest matching pursuit approach, the proposed algorithm achieves a significant speed improvement while maintaining the high approximation quality.
Keywords :
approximation theory; image representation; iterative methods; approximation quality; image approximation; multiscale matching pursuit algorithm; Approximation algorithms; Dictionaries; Gallium nitride; Greedy algorithms; Helium; Image coding; Matching pursuit algorithms; Pursuit algorithms; Signal processing algorithms; Signal representations;
Conference_Titel :
Communications, Circuits and Systems, 2008. ICCCAS 2008. International Conference on
Conference_Location :
Fujian
Print_ISBN :
978-1-4244-2063-6
Electronic_ISBN :
978-1-4244-2064-3
DOI :
10.1109/ICCCAS.2008.4657871
