Satellite Image Classification Using Cascaded Architecture of Neural Fuzzy Network - Advanced Techniques in Knowledge Discovery and Data Mining - page 226

Gray values of (x,y) y = 1 y = 2 y = 3

x = 1 316 316 316

x = 2

297 297 297

x = 3

316 297 297

BTN CON IDM ENT
1/569 95/5: 1/::9 1/544

LL LH HL HH
Energy 51318 :/389 27/1: 2/528
Entropy 1/:63 1/6:8 1/824 1/7:

SOM SOM SOM

2D coordinates

2D coordinates 2D coordinates

C

C

C

Neural fuzzy network, SONFIN

Class 3

Fig. 8.11. An illustrative example of the CNFM.

Table 8.1. Classification results with all types of inputs.

Class 1 Class 2 Class 3 Class 4 Class 5 Overall Average

Class 1 495 0 0 4 1 99 %

C lass 2 0 512 0 0 0 100 %

96.5 %

Class 3 0 0 470 0 0 100 %

Class 4 7 2 1 424 28 92.2 %

Class 5 25 5 0 13 462 91.5%