Table 6 Quantitative comparison of different methods on the Pavia_Center dataset.

From: Cross-range self-attention single hyperspectral image super-resolution method based on U-Net architecture

Models

d

HR_image_size

\(\text {PSNR}\uparrow\)

\(\text {SSIM}\uparrow\)

\(\text {SAM}\downarrow\)

\(\text {ERGAS}\downarrow\)

\(\text {CC}\uparrow\)

Bicubic

2

256 × 256

32.8517

0.9314

7.3982

3.9635

0.9617

GDRRN

2

256 × 256

32.9161

0.9314

7.3518

3.9410

0.9619

Deep_hs_Prior

2

256 × 256

33.1241

0.9389

6.8965

3.7564

0.9685

SSPSR

2

256 × 256

34.7190

0.9521

5.9274

3.2507

0.9740

PDE-Net

2

256 × 256

34.0184

0.9455

6.4350

3.4572

0.9703

PUDPN

2

256 × 256

33.9447

0.9442

6.4663

3.5147

0.9696

CST

2

256 × 256

34.4367

0.9495

6.1245

3.3439

0.9724

Cs_Unet (Ours)

2

256 × 256

34.9608

0.9538

5.7643

3.1872

0.9749

Bicubic

4

256 × 256

29.3110

0.8081

13.8957

8.2676

0.8899

GDRRN

4

256 × 256

29.6292

0.8018

13.5647

7.9813

0.8957

Deep_hs_Prior

4

256 × 256

29.4749

0.8164

13.6542

8.0125

0.8924

SSPSR

4

256 × 256

30.1413

0.8306

12.6442

7.6061

0.9082

PDE-Net

4

256 × 256

30.0437

0.8297

12.8160

7.8977

0.9040

PUDPN

4

256 × 256

30.2815

0.8366

12.4569

7.4260

0.9107

CST

4

256 × 256

30.8411

0.8585

11.7536

7.1354

0.9213

Cs_Unet (Ours)

4

256 × 256

30.8925

0.8568

11.7257

6.9656

0.9226

Bicubic

8

256 × 256

22.8723

0.4486

23.7054

12.1303

0.6208

GDRRN

8

256 × 256

23.4299

0.4461

22.4375

11.0773

0.6325

Deep_hs_Prior

8

256 × 256

23.3895

0.4454

22.5234

11.1597

0.6395

SSPSR

8

256 × 256

23.3531

0.4462

22.2301

11.2486

0.6254

PDE-Net

8

256 × 256

23.2059

0.4478

23.0538

11.7969

0.6160

PUDPN

8

256 × 256

23.4761

0.4521

22.2113

11.1386

0.6392

CST

8

256 × 256

23.0664

0.4118

23.4727

11.7914

0.5962

Cs_Unet (Ours)

8

256 × 256

23.5198

0.4625

22.1365

11.0315

0.6430

Back to article page