Table 12 Training time for PCA, ICA, Naïve AE and D-Path-AE in seconds.

From: Enhancing feature learning of hyperspectral imaging using shallow autoencoder by adding parallel paths encoding

Dataset

PCA

ICA

Naïve

3D ResAE

D-Path-AE

KSC

0.01671

0.5325

80.4124

250.2428

79.9325

Salinas

0.5658

3.4779

190.2503

341.3912

178.9444

Pavia Center

0.2035

1.1143

229.3928

402.4124

200.5259

Back to article page