Title :
Isomap Algorithm Based on 2D Gabor Wavelets and 2DPCA
Author :
Qin, Chanchan ; Xu, Hongbo ; Wang, Lihua ; Zhang, Huijuan ; Li, Guoqing
Author_Institution :
Coll. of Phys. Sci. & Technol., HuaZhong Normal Univ., Wuhan, China
Abstract :
Compared with gray feature, the Gabor feature is more effective for facial image representation. But in practice, the dimension of a Gabor feature is so high that the computation is prohibitively large. In this paper, We use the two-dimensional principal component analysis (2DPCA) to reduce the dimension and then replace the original gray feature with the Gabor feature vector of reduced dimension as the front-end input of traditional Isometric Mapping(Isomap) algorithm. The validity of this method can be verified by experimental results.
Keywords :
face recognition; feature extraction; image representation; principal component analysis; vectors; wavelet transforms; 2D Gabor wavelets; Gabor feature vector; Isomap algorithm; facial image representation; gray feature; isometric mapping; principal component analysis; Algorithm design and analysis; Classification algorithms; Convolution; Covariance matrix; Face; Image edge detection; Principal component analysis; 2D Gabor wavelets; 2D principal component analysis; Isomap algorithm;
Conference_Titel :
Information Science and Management Engineering (ISME), 2010 International Conference of
Conference_Location :
Xi´an
Print_ISBN :
978-1-4244-7669-5
Electronic_ISBN :
978-1-4244-7670-1
DOI :
10.1109/ISME.2010.257
