Texture features are recognized to be a special hint in images, which represent the
spatial relations of the gray pixels. Nowadays, the applications of the texture analysis in image
classification spread abroad. Combined with wavelet multi-resolution analysis or support vector
machine statistical learning theory, texture analysis could improve the quality of classification
increasingly. In this paper, we focus on the land cover for the Three Gorges reservoir using remote
sensing data SPOT-5, a new classification method, wavelet-SVM classifier based on texture
features, is employed for this study. Compare to the traditional maximum likelihood classifier and
SVM classifier only use spectrum feature, this method produces more accurate classification
results. According to the real environment of the Three Gorges reservoir land cover, a best texture
group is selected from several texture features. Decompose the image at different levels, which is
one of the main advantage of wavelet, and then compute the texture features in every sub-image,
and the next step is eliminating the redundant, every texture features are centralized on the first
principal components using principal component analysis. Finally, with the first principal
components inputted, we can get the classification result using SVM in every decomposition scale,
but what the problem we couldn't overlook is how to select the best SVM parameters. So an
iterative rule based on the classification accuracy is induced, the more accuracy, the proper
parameters.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.