Table 1 Number of training and testing samples for IP, UP, and HU datasets.

From: Fusion of circulant singular spectrum analysis and multiscale local ternary patterns for effective spectral-spatial feature extraction and small sample hyperspectral image classification

No.

Indian Pines (1%)

Pavia University (1%)

Houston2013 (1%)

Class

Training

Test

Class

Training

Test

Class

Training

Test

1

Alfalfa

1

53

Asphalt

67

6564

Healthy grass

13

1238

2

Corn-notill

15

1419

Meadows

187

18462

Stressed grass

13

1241

3

Corn-mintill

9

825

Gravel

21

2078

Synthetic grass

7

690

4

Corn

3

231

Trees

31

3033

Trees

13

1231

5

Grass-pasture

5

492

Metal Sheets

14

1331

Soil

13

1229

6

Grass-trees

8

739

Bare-Oil

51

4978

Water

4

321

7

Grass-pasture-mowed

1

25

Bitumen

14

1316

Residential

13

1255

8

Hay-windrowed

5

484

Bricks

37

3645

Commercial

13

1231

9

Oats

1

19

Shadows

10

937

Road

13

1239

10

Soybean-notill

10

958

   

Highway

13

1214

11

Soybean-mintill

25

2443

   

Railway

13

1222

12

Soybean-clean

7

607

   

Parking lot1

13

1220

13

Wheat

3

209

   

Parking lot2

5

464

14

Woods

13

1281

   

Tennis court

5

423

15

Buildings-Grass-Trees

4

376

   

Running track

7

653

16

Stone-Steel-Towers

1

94

      
 

Total

111

10255

Total

432

42344

Total

158

14871