Table 8 Performance metrics comparison across methodologies using the MUUFL 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

0.9675

0.9946

0.9572

0.9679

0.9840

0.9573

0.9474

0.9178

C2

0.7440

0.1717

0.7937

0.3445

0.8270

0.8762

0.7509

0.8821

C3

0.8334

0.9914

0.8100

0.9889

0.8907

0.8885

0.9655

0.8602

C4

0.8777

0.8762

0.9526

0.8649

0.9720

0.9391

0.9740

0.9720

C5

0.9482

0.9716

0.9520

0.9619

0.9694

0.9387

0.9075

0.9299

C6

0.8000

0.8798

0.9639

0.9809

0.0286

0.9385

0.9920

0.9833

C7

0.7045

0.9901

0.9075

0.9522

0.9055

0.9155

0.8495

0.9636

C8

0.9434

0.9881

0.9765

0.9635

0.9856

0.9545

0.9543

0.9624

C9

0.5012

0.9170

0.6079

0.8741

0.2799

0.3436

0.3186

0.8540

C10

0.5818

0.6721

0.0265

0.7818

0.0000

0.0000

0.0476

0.7333

C11

0.8889

0.8885

0.9909

0.9959

0.0000

0.8571

0.9074

0.9959

OA

0.8981

0.9163

0.6922

0.9128

0.9195

0.9172

0.9076

0.9166

AA

0.7991

0.8492

0.8126

0.8797

0.6621

0.7826

0.7831

0.9141

Kappa

0.8646

0.8889

0.6275

0.8850

0.8934

0.8910

0.8787

0.8906

  1. Bold denotes optimal values.
Back to article page