DocumentCode :
1699855
Title :
Rotation invariant features of wavelet transform for texture retrieval
Author :
Caglar, Faruk ; Cavusoglu, Bulent
Author_Institution :
Tercan Vocational Sch., Erzincan Univ., Erzincan, Turkey
fYear :
2013
Abstract :
Wavelet transform is both sensitive to translation and rotation. This feature of the transform diminishes the discriminative power of wavelet coefficients among different classes where rotated versions of textures are present. We analyze statistical behavior of wavelet coefficients and show that some features of wavelets are more robust than others against translation and rotation. We also show that using higher order moments increase the overall performance. A new parameter is also proposed to determine the discriminative features out of a possible feature set. This new parameter is derived based on statistics of wavelet transform and called as effective discriminative power (EP). Based on EP, reduced subband feature set is proposed and applied on the modified Brodatz database for texture retrieval and shown that the reduced feature set have superior performance compared to the one which just includes mean and standard deviations of all the subbands. The reduced feature set is also computationally less expensive since it eliminates rotation-variant features and results in less number of features with better performance.
Keywords :
feature extraction; image retrieval; image texture; statistical analysis; visual databases; wavelet transforms; Brodatz database; EP; discriminative features; effective discriminative power; higher order moments; rotation invariant features; rotation-variant features; subband feature set reduction; texture retrieval; wavelet coefficients discriminative power; wavelet coefficients statistical behavior; wavelet transform; wavelet transform statistics; Abstracts; Approximation methods; Databases; Wavelets; effective discriminative power; reduced feature set; texture retrieval;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Signal Processing and Information Technology(ISSPIT), 2013 IEEE International Symposium on
Conference_Location :
Athens
Type :
conf
DOI :
10.1109/ISSPIT.2013.6781909
Filename :
6781909
Link To Document :
بازگشت