Abstract
Hyperspectral images (HSIs) provide invaluable information in both spectral and spatial domains for image classification tasks. In this paper, we use semantic representation as a middle-level feature to describe image pixels’ characteristics. Deriving effective semantic representation is critical for achieving good classification performance. Since different image descriptors depict characteristics from different perspectives, combining multiple features in the same semantic space makes semantic representation more meaningful. First, a probabilistic support vector machine is used to generate semantic representation-based multifeatures. In order to derive better semantic representation, we introduce a new adaptive spatial regularizer that well exploits the local spatial information, while a nonlocal regularizer is also used to search for global patch-pair similarities in the whole image. We combine multiple features with local and nonlocal spatial constraints using an extended Markov random field model in the semantic space. Experimental results on three hyperspectral data sets show that the proposed method provides better performance than several state-of-the-art techniques in terms of region uniformity, overall accuracy, average accuracy, and Kappa statistics.
Highlights
In the past few decades, hyperspectral images (HSIs) have been frequently used in earth observation
We have proposed a novel method which could obtain semantic representation of each pixel with more detailed information and less noise for hyperspectral image classification
The probabilistic support vector machine (SVM) was used to map these features which lie on different spaces to the same semantic space
Summary
In the past few decades, hyperspectral images (HSIs) have been frequently used in earth observation. A variety of spectral pixel-wise classifiers [5]-[8] have been proposed to solve this problem Among these established methods, support vector machine (SVM) classifiers [5], to some degree, have shown promising success in HSI classification and gained large attention due to their robust performance in a high-dimensional feature space with the ability of dealing with a small number of training samples. There are many ways to impose spatial information, such as post-processing techniques [9], [10], composite kernel [11]-[13], joint sparsity model [14]-[16], and Markov random fields [17]-[23] These methods can significantly enhance the classification accuracy in the applications
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callMore From: IEEE Transactions on Geoscience and Remote Sensing
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
