Table 6 Performance metrics comparison across methodologies using the Houston2013 dataset.

From: ELDGG: an end-to-end LiDAR-dynamic-guided GAN for hyperspectral image hierarchical reconstruction and classification

Class

SVM

FusAtNet

CCRNet

MAHiDFNet

GLT

MFT

ExViT

ELDGG

C1

0.8747

0.9856

0.9796

0.9813

0.9565

0.9710

0.9221

0.9796

C2

0.9947

0.9928

0.9646

0.9911

0.9973

0.9930

0.9980

0.9575

C3

0.9904

0.9957

0.9984

1.0000

1.0000

0.9964

0.9946

0.9968

C4

0.9866

0.9871

0.9973

1.0000

0.9982

0.9799

1.0000

0.9991

C5

0.9973

0.9847

1.0000

1.0000

1.0000

0.9990

1.0000

1.0000

C6

0.8469

0.7754

1.0000

0.9966

0.8191

0.8736

0.9923

0.9966

C7

0.8816

0.9890

0.9851

0.9842

0.9799

0.9734

1.0000

1.0000

C8

0.8606

0.9598

0.9670

0.9098

0.9982

0.9016

0.9197

0.9434

C9

0.8112

0.9832

0.9414

0.9911

0.9397

0.8433

0.9631

0.9885

C10

0.8315

0.9658

0.9258

0.8234

0.9937

0.7721

0.9949

0.9810

C11

0.8356

0.9652

0.9658

0.9317

1.0000

0.9899

0.9777

0.9982

C12

0.8108

0.9416

0.9649

0.9883

0.9486

0.9686

0.9838

0.9892

C13

0.3759

0.7377

0.9594

0.9716

0.9551

0.8966

0.9521

0.9882

C14

0.9766

0.9089

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

C15

0.9899

0.9924

1.0000

1.0000

1.0000

1.0000

1.0000

1.0000

OA

0.8833

0.9636

0.9731

0.9662

0.9791

0.9437

0.9781

0.9859

AA

0.8710

0.9443

0.9766

0.9713

0.9724

0.9439

0.9799

0.9879

Kappa

0.8737

0.9606

0.9709

0.9635

0.9774

0.9391

0.9764

0.9847

  1. Bold denotes optimal values.
Back to article page