Super-pixel as an object-oriented segmentation method has been widely used in image analysis. SLIC super-pixel utilizes the self-similarity of pixel color feature in natural scene as well as the edge information constraint between different regions, segments image through iteratively clustering, and obtains the corresponding homogeneous region. But for typical terrain classification, the color and intensity of a pixel do not have the consistency, this may lead to a problem of misclassification. Compared with the color and intensity features of pixels, the local texture features have a better description for the class attribute of pixels. In this paper, we use the random projection to extract texture feature from local image patches, combined with the advantage of compact structure and excellent homogeneity of SLIC segmentation, we propose a random projection super-pixel segmentation method, then utilize K-means to cluster the segmented patches in order to implement image classification. Experiments show that compared with the SLIC superpixel segmentation, the patches segmented by the random projection super-pixel segmentation is better to reflect the texture information of images, and the classification accuracy has been greatly improved.
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.