Local and Global Geometric Structure Preserving and Application to Hyperspectral Image Classification

Table 1

Training set for Indian Pines.

ID	Class name
1	Alfalfa (46/16, 34.78%)^*
2	Corn-no-till (1428/121, 8.47%)
3	Corn-min-till (830/72, 8.67%)
4	Corn (237/31, 13.08%)
5	Grass-pasture (483/48, 9.94%)
6	Grass-trees (730/75, 10.27%)
7	Grass-pasture-mowed (28/16, 57.14%)
8	Hay-windrowed (478/48, 10.04%)
9	Oats (20/15, 75.00%)
10	Soybean-no-till (972/80, 8.23%)
11	Soybean-min-till (2455/234, 9.53%)
12	Soybean-clean (593/52, 8.77%)
13	Wheat (205/34, 16.59%)
14	Woods (1265/114, 9.01%)
15	Buildings-grass-trees-drives (386/53, 13.73%)
16	Stone-steel-towers (93/20, 21.51%)
Total	10249/1029

Numerical value in each row refers to number of total samples, number of training samples, and percentage of training samples in each class, respectively.