An Efficient Method for Supervised Hyperspectral Band Selection
Citations
444 citations
Cites background from "An Efficient Method for Supervised ..."
...[22] select the bands with an incremental manner....
[...]
359 citations
Cites methods from "An Efficient Method for Supervised ..."
...In [34], FS is treated as a task to make the class mean signatures the best endmembers (a concept used in spectral unmixing [91])....
[...]
246 citations
Cites methods from "An Efficient Method for Supervised ..."
...Similarly, in [49], using class spectral signatures, the minimum estimation abundance covariance (MEAC) algorithm incrementally selected dissimilar bands to preserve the desired information for classification through minimization of the trace of abundance covariance matrix as...
[...]
195 citations
174 citations
Cites background from "An Efficient Method for Supervised ..."
...4) Extensive empirical studies with four publicly available data sets are conducted on three important applications, such as hyperspectral classification [48], anomaly detection [49], and target detection [50] (Section IV)....
[...]
References
11,697 citations
"An Efficient Method for Supervised ..." refers background in this paper
...According to [9], the stochastic features of α̂ include...
[...]
...According to [9], the first- and second-order moments of α̂ become...
[...]
10,148 citations
3,532 citations
3,104 citations
"An Efficient Method for Supervised ..." refers methods in this paper
..., sequential forward selection (SFS) and sequential forward floating selection (SFFS), can be used [7]....
[...]
1,570 citations
"An Efficient Method for Supervised ..." refers methods in this paper
...In order to evaluate the amount of class information and class separability in the selected bands, a supervised classification algorithm, such as orthogonal subspace projection (OSP) [14], constrained linear discriminant analysis (CLDA) [15], or SVM, can be applied....
[...]
...The OSP classifier was chosen for soft classification, and spatial correlation coefficient was calculated between the corresponding classifications maps using all the original bands and the selected bands....
[...]
...This is doable only if no training or test is required for classification (e.g., OSP and CLDA)....
[...]