Abstract
Spectral-spatial classification for hyperspectral imagery has been receiving much attention, since the detailed spectral and rich spatial information of hyperspectral images can be fully exploited to improve the classification accuracy. However, when the original hyperspectral images have very noisy bands, these bands may have an unfavorable impact on the classification, and are often discarded in advance based on expert knowledge. In this study, a spectral-spatial conditional random field classification algorithm integrating band selection (CRFBS) is developed for hyperspectral imagery with severe noise bands. The proposed algorithm integrates band selection based on the relative utility of the spectral bands for classification. Consequently, negative effects of severe noise bands are eliminated and the need for high-quality image data is substantially reduced. In addition, the CRFBS algorithm makes comprehensive use of both the spectral and the spatial cues to improve the classification performance. The spectral cues are formulated by integrating the support vector machine and random forest algorithms to improve the spectral discriminative ability in the unary potentials, and the spatial information are modeled to consider the interactions between pixels in pairwise potentials. The experiments using different airborne and UAV-borne hyperspectral data verified the effectiveness of the CRFBS method. The CRFBS algorithm can achieve accurate interpretation of the various classification categories and a more than 3% improvement in classification accuracy, compared with the method using the original hyperspectral image with severe noise bands.
Highlights
H YPERSPECTRAL imagery is a very important data source for deriving detailed thematic information on the earth surface, since it contains hundreds of narrow spectral channels to distinguish the subtle spectral difference of various materials [1], [2]
For the object-oriented classification method, a multi-resolution segmentation algorithm implemented in eCognition 8.0 (FNEA) was used to obtain segmentation objects, and a majority voting strategy was applied to obtain the objectoriented classification map based on the pixel-wise Support Vector Machine (SVM) classification map
The deep learning approach used as a comparison algorithm was the spectral-spatial attention network (SSAN) [32], which extracts spectral-spatial features based on a spectral attention bi-directional recurrent neural network branch and a spatial attention convolutional neural network (CNN) branch
Summary
H YPERSPECTRAL imagery is a very important data source for deriving detailed thematic information on the earth surface, since it contains hundreds of narrow spectral channels to distinguish the subtle spectral difference of various materials [1], [2]. In the classification process there is a strong correlation between hundreds of narrow spectral bands, which results in redundant information content and high spectral dimensionality. This can lead to a high-dimensional processing problem, the so-called Hughes phenomenon [5], if only a limited number of training samples are available. Feature extraction creates new features in a feature space with lower dimensionality while satisfying certain criteria regarding the original spectral features [7], [8]. Such techniques comprise linear discriminant analysis and principal component analysis, among others. Examples are the band selection method based on saliency bands and scale selection (SBSS) [11] and the salient band selection method based on manifold ranking [12]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
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 call
