Title :
Frequency and Space Domain Features for Image Classification Using Gaussian Mixture Models
Author :
Bin Fu ; Ren, Zhen
Author_Institution :
Pattern Recognition Lab., Harbin Eng. Univ., Harbin
Abstract :
This paper presents an effective combination of wavelet-based features and SIFT features, both of them have the frequency domain and space domain information characteristic. For the combined feature patches extracted from images we then adopt the PCA transformation to reduce the dimensionality of their feature vectors. And the reduced vectors are used to train Gaussian mixture models (GMMs) in which the mixture weights are adjusted iteratively. We experiment on Caltech datasets using this enhanced method, and the results comparing with several other methods show that the combination of salient feature vectors and GMM gives a much better improvement in image classification.
Keywords :
Gaussian processes; feature extraction; image classification; principal component analysis; wavelet transforms; Caltech datasets; Gaussian mixture models; PCA transformation; frequency domain features; image classification; salient feature vectors; space domain features; wavelet-based features; Computer vision; Detectors; Educational institutions; Feature extraction; Image classification; Image recognition; Laboratories; Object recognition; Pattern recognition; Principal component analysis; Feature extraction; Gaussian mixture models; Image Classification;
Conference_Titel :
Embedded Software and Systems Symposia, 2008. ICESS Symposia '08. International Conference on
Conference_Location :
Sichuan
Print_ISBN :
978-0-7695-3288-2
DOI :
10.1109/ICESS.Symposia.2008.33