Skip to main content

Table 2 ERGAS values of state-of-the-art SR methods. The bold number indicates the best performance

From: Single-frame super-resolution for remote sensing images based on improved deep recursive residual network

Scene

Scale

Bicubic

SRCNN

FSRCNN

DRCN

VDSR

EDSR

IDRRN

Airplane

Ă—2

3.1462

2.6994

2.4515

2.3379

2.2613

2.2075

2.1987

Ă—3

2.9942

2.6364

2.3988

2.4103

2.3015

2.2889

2.2044

Ă—4

2.6828

2.4634

2.4077

2.3345

2.1613

2.0481

1.9985

Bridge

Ă—2

6.5783

6.0234

5.9902

5.6359

5.2230

5.5834

5.1681

Ă—3

5.7809

5.5293

5.4437

5.2569

5.1571

4.9849

4.9901

Ă—4

4.8862

4.6293

4.6128

4.5153

4.3820

4.4121

4.3815

Harbor

Ă—2

9.4446

8.1335

8.3982

9.0341

7.0818

7.9539

6.3731

Ă—3

11.2215

10.0231

9.9025

8.9336

8.0004

8.0349

8.1795

Ă—4

12.9885

11.3769

10.5539

10.2495

10.5594

10.5104

9.5517

Inter-section

Ă—2

14.2147

11.2275

11.0060

12.4054

11.1685

11.7313

10.8496

Ă—3

13.4601

12.7299

12.0232

11.9441

11.5986

12.7035

11.0287

Ă—4

15.1112

13.0236

12.7750

13.6313

11.0426

11.0405

11.4467

Parking_lot

Ă—2

7.7762

7.4731

7.5994

6.9825

6.9904

7.5398

6.8247

Ă—3

6.3143

6.1019

5.7282

5.5627

5.1651

5.0154

5.3081

Ă—4

5.8401

4.6099

4.5310

6.7772

5.5812

4.8773

4.5238