DocumentCode :
3382243
Title :
Design method for CNN Gabor-type filters
Author :
Matei, Radu
Author_Institution :
Fac. of Electron. & Telecommun., Tech. Univ. of Iasi, Iasi
fYear :
2008
fDate :
Aug. 31 2008-Sept. 3 2008
Firstpage :
320
Lastpage :
323
Abstract :
A class of widely used tools for image processing and computer vision applications are Gabor filters. In this paper analog implementation of these filters using cellular neural networks is approached. Some template design methods for Gabor filters are proposed, based on rational approximations of the frequency response, and their accuracy and efficiency is discussed comparatively.
Keywords :
Gabor filters; approximation theory; cellular neural nets; computer vision; frequency response; network synthesis; CNN Gabor filters; cellular neural networks; computer vision; frequency response; image processing; Cellular neural networks; Design methodology; Feature extraction; Filtering; Frequency response; Gabor filters; Image processing; Motion analysis; Nonlinear filters; Transfer functions;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronics, Circuits and Systems, 2008. ICECS 2008. 15th IEEE International Conference on
Conference_Location :
St. Julien´s
Print_ISBN :
978-1-4244-2181-7
Electronic_ISBN :
978-1-4244-2182-4
Type :
conf
DOI :
10.1109/ICECS.2008.4674855
Filename :
4674855
Link To Document :
بازگشت