Broad Learning System with Locality Sensitive Discriminant Analysis for Hyperspectral Image Classification

Algorithm 1

BLS-LSDA training algorithm.

(i)	Require: Training samples , and the corresponding labels ; the group of mapped nodes in BLS n.
(ii)	Ensure: The weights of the output layer, .
(1)	Extract the Gabor feature and AWF feature of sample ;
(2)	Feed and into the layer which implement LSDA, to produce dimensional reduced features and ;
(3)	Concatenate weighted and by equation (15), yielding ;
(4)	for i = 1; i < n; i++ do
(5)	Assign a random value to and ;
(6)	Calculate the mapped feature values .
(7)	end for
(8)	Concatenate the mapped feature values to get a mapped feature group ;
(9)	Assign and with random values;
(10)	Calculate the enhancement nodes with ;
(11)	Apply LSDA to each in and in another LSDA layer to get and ;
(12)	Concatenate and to produce ;
(13)	Calculate the connection weights between the BLS’s hidden layer and an output layer with .