om the
e field
ng was
1domly
z result
It from
ch with
er 242
val
id haze
1g. The
good,
which
date of
moved
00nm).
—
oval
image
oved
) good,
which
date of
on
among
ell as
except
use of
caused
misclassification. The LC class “Settlements” was extracted
with relatively more manual editing and hence the omission and
commission errors for this class are relatively low.
Settle- | Other Row Omission
ments | Land Error (in %
(Pixels)
115 18.3
32 12.5
21.6
24.0
12.5
30.8
Cropland Wetlands
Land
Overall
80
Sample Calculation:
Omission Error of Forest Land = 21/115 = 18.3%
Commission Error of Forest Land = 4/98 = 4.1%
Overall Accuracy = 186 / 230 = 80.9%
Table 1. Accuracy Assessment of LC Result from ALOS
AVNIR-2
LC pattern among the analyzed 3 epochs is similar, Figure 5.
This is also supported from the result of pixel tracking between
1995 and 2009 LC result so presented in Table 2 which
indicates vey less change in forest change for the tracked pixels.
Landsat 5 TM (1990) Image and Classified LC
Figure 5. Satellite Image of 3 Epochs (bands 4, 3,2 as R, G, B)
and Classified LC
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Volume XXXIX-B7, 2012
XXII ISPRS Congress, 25 August — 01 September 2012, Melbourne, Australia
Med Land Cover 2009 | Forest | | Settte. | Orher
bacon Land : Cropland | Grassiand | Wetlands ments | Land
5 hd Bixci Court
Laud Cover 1985 : al = : 38 17 33 i 3 | i
Forest Land 158 & 3 it b i ; f
Cropland 25 m 7 1 o | 0
Grassiand 18 è A 0 o i i
Wettands as à e A: ü n
Settlements | 3 0: f o f | x 9
Other Land 3 6: 1 2 oj ef à
Total Number of Pixels Analyzed = 242
Table 2. Result of Analyzed Pixels for LC Change from 1995 to
2009
Comparatively, the change between cropland and grassland is
more and mis-classification may be a reason for this. Moreover,
it shall be noted that, spatial resolution for used satellite data are
different. The spatial resolution is 30m for Landsat TM and
ETM+, and 10m for ALOS AVNIR-2. Also, the “Unsupervised
classification” was carried out with 6 bands for Landsat and
ETM+ data and 4 bands for ALOS AVNIR-2 data.
Moreover, in the analyzed area, there is location specific
change of LC and can be traced out only upon detail analysis,
such as at location presented in Figure 6.
Landsat TM (1995) Landsat ETM+ (2002) : ALOS AVNIR-2 (2009)
Image and Classified IC Image and Classified LC fmage and Classified LC ;
Figure 6. An Example of Land Cover Change (1995 to 2009)
5. CONCLUSIONS
The approach devised for cloud/shadow and haze removal and
LC change detection has good results. The accuracy of LC
classification is also promising one. The program used for
processing also considers classification of large number of
imageries in the shortest possible time with required number of
LC classes, specifically for REDD+ application with [PCC
standard. This involves less human intervention and can
classify imagery for extracting LC in wide variety of landscape
and different dated data; thus can be considered as semi-
automatic.
Thus, this semi-automated methodology is expected to be suited
for diversified conditions and allow processing of large number
of imageries in comparatively less time, which will be very
useful in analysis of successive change/monitoring of tropical
rain forest in Africa.