30 October 2009 Contourlet spectral histogram for texture retrieval of remotely sensed imagery
Author Affiliations +
Proceedings Volume 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications; 74981R (2009) https://doi.org/10.1117/12.833964
Event: Sixth International Symposium on Multispectral Image Processing and Pattern Recognition, 2009, Yichang, China
Abstract
In this paper, a progressive texture retrieval algorithm for remotely sensed imagery based on Contourlet spectral histogram is proposed. Contourlet transform is applied to extract texture features of remotely sensed imagery from different scales and different directions. Decomposed low-pass subband and high-pass subbands are used to realize coarse and fine retrieval respectively. The proposed algorithm not only utilizes the advantages of Contourlet on multiscale and multi-direction feature representation and extraction, but also utilizes the efficiency of spectral histogram on distributed statistical feature description. Experimental results prove that Contourlet Spectral Histogram provides a powerful tool for texture retrieval of remotely sensed imagery.
© (2009) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Qimin Cheng, Guangxi Zhu, Xianqiang Zhu, "Contourlet spectral histogram for texture retrieval of remotely sensed imagery", Proc. SPIE 7498, MIPPR 2009: Remote Sensing and GIS Data Processing and Other Applications, 74981R (30 October 2009); doi: 10.1117/12.833964; https://doi.org/10.1117/12.833964
PROCEEDINGS
6 PAGES


SHARE
Back to Top