An Improved Multispectral Palmprint Recognition System Using Autoencoder with Regularized Extreme Learning Machine

Table 1

Parameters setup.

Network Model	Parameters
AE	A number of hidden nodes of AE is .
	Encoder and Decoder transfer function is a logistic sigmoid function.
	Maximum epochs = 10.
	L2WeightRegularization = 0.004.
	A loss function is a mean squared error function.
	Training algorithm is a conjugate gradient descent.
RELM	A number of hidden nodes of RELM, is
	A regularization parameter () =
	An activation function is a nonlinear sigmoid function,