Title :
Dynamic Textures Retrieval with Surfacelet Transform
Author :
Zhang, Jiuwen ; Sun, Xianqiang
Author_Institution :
Sch. of Inf. Sci. & Eng., Lanzhou Univ., Lanzhou, China
Abstract :
Surfacelet transform has characteristics of multidirectional and multiresolution decomposition. In this paper, we use it to extract features of dynamic textures, and then build a retrieval system of dynamic textures. We compare the retrieval results at different scales and directions, and test the anti-noise and anti-rotation ability of this approach. The experimental results show that this approach has a better effect compared to 3D DWT.
Keywords :
feature extraction; image resolution; image retrieval; image sequences; image texture; transforms; dynamic texture retrieval; feature extraction; image sequence; moving scene; multidirectional decomposition; multiresolution decomposition; surfacelet transform; Data mining; Feature extraction; Geometrical optics; Image motion analysis; Image sequence analysis; Optical filters; Optical surface waves; Sea surface; Spatiotemporal phenomena; Surface texture;
Conference_Titel :
Image and Signal Processing, 2009. CISP '09. 2nd International Congress on
Conference_Location :
Tianjin
Print_ISBN :
978-1-4244-4129-7
Electronic_ISBN :
978-1-4244-4131-0
DOI :
10.1109/CISP.2009.5304594
