9 vu e
—
4
V4
— Wd
"4
Automatic classification Visual interpretation
E high density urban arable land EB forest
low density urban pastures WE water
industrial heterogeneous 0 5 10km
L—NE.— N
Figure 2: Result of automatical classification compared to CORINE land-cover map (ÖK 50-49 Wels)
CORINE Land-use model
LI I. artificial surfaces II. agricultural areas III. natural areas IV V Y %
LII ri 12 I3 L4 II.1 0.2 0.3 0.4 mi m2 | 3.1 vi V. X %
I 11 26 6 0 0 0 0 0 0 0 0 0 0 0 33 | 798 252 |-81.5
12 7 116 1 8 17 4 0 9 2 0 0 0 4 170 | 68.6
L3 0 3 4 2 0 0 0 0 2 0 0 0 2 13} 276
L4 1 7 0 23 2 0 0 2 0 0 0 0 1 36 | 627
I O1 0 18 0 61 >715 4 2 65 16 2 0 0 1 829 | 86.2 | 1531 | 89.1
na 0 1 0 0 8 81 0 5 2 3 0 1 1 102 | 792
IL3 0 5 0 0 13 1 81 60 32 4 0 3 2] 200 | 405
I4 0 11 0 0 76 3 40 | 21 52 2 0 1 21° 398 | 530
IH | mm. 0 3 0 1 20 1 13 29 | 818 21 2 0 2 | 910 | 898 | 1067 | 93.1
m2 0 0 0 0 0 0 3 0 48 33 1 0 0 84 | 39.0
1.3 0 0 0 0 0 0 0 15 30 26 0 0 73 | 41.6
IV | IV.1 0 0 0 0 1 2 0 1 0 3 0 76 3 86 | 88.4 86 | 88.4
V V.1 0 1 0 0 0 0 0 1 3 0 0 2 154 161 | 952 161 | 95.2
> 35 172 5 40 | 853 96 139 | 382] 991 98 20 83 173 | 3097
% 76.0 | 67.7 | 65.6 | 56.6 | 83.8 | 842 | 583 | 55.2 | 82.6 33.5 | 915 | 913 | 888 76.3
E 253 1471 1118 83 173 3097
% 81.3 92.8 88.9 91.3 | 88.8 90.2
Table 5: Confusion matrix for Level I and Level II land-use classes (elements = absolute numbers / 100)
845
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996