In: Wagner W., Székely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
434 
Fig3: Classified print of the NigeriaSat-1 image 
Land Cover Type 
Built-up Area 
BBI Bare rock 
Farm land 
BB Sec forest regrowth 
BH Water body 
Kilometers 
10 O 10 20 
Fig4: Classified print of the SPOT image 
The SPOT image had been rectified but was resampled to 32m in 
order to have the three imageries in the same spatial resolution. 
4.3 Land use characterization of the study 
Table 4.3 Shows the area covered in hectares, by the five 
landuse classes’ namely built-up areas, bare rocks, farmland and 
secondary forest regrowth and water body. On the Landsat image, 
“built-up area” covered 4,388.15 hectares which represented 9.6% 
of the total area of land under study. “Bare rock” covered 8,925.8 
hectares (19.4%); “farmland” 9,461.96 hectares (20.6%); 
“secondary forest regrowth” 19,881.9 hectares (43.3%) and water 
body 3,241.56 hectares (7.1%). On NigeriaSat-1, “built-up area” 
covered an area of 4,547.69 hectares which represented 10% of the 
total land area under study. “Bare rock” was 8,016.2 hectares 
(18%) “farmland” 6,682.6 hectares (15%); “secondary forest 
regrowth” 23,266.60 hectares (52%) and “water body” 2,326.32 
hectares (5%). 
With respect to SPOT image, “built-up area” covered 4,891.85 
hectares which was 12% of the total land area under study. “Bare 
rock” was 12,887.70 hectares (30%); “farmland” 7,565.11 hectares 
(18%), “secondary forest regrowth” 14,665.7 hectares (34%) and 
“water body” covered 2,642.84 hectares (6%). In general, the land 
cover characterizations of the three imageries are over comparable 
area cover, despite the differences in the spatial resolutions of the 
images. “Secondary forest regrowth” is the largest in the three 
images and farmlands are comparable in size. The main difference 
is with respect to bare rock which was shown to cover a larger area 
in SPOT. The difference may be related to the time of imaging. 
Images taken in dry seasons would reveal more bare surfaces than 
those taken in the wet season. 
Figure . 5 Histogram showing the three imageries with their 
respective landuses types. 
4.4. Discussion: Variation in land use characterization and 
Accuracy 
The most basic output in remote sensing applications to 
landuse/landcover studies is land use characterization. This is 
information about the sizes of various landuse. When such 
information is available over a long period of time, it allows an 
assessment of landuse dynamics. Such characterizations are 
invaluable for the monitoring and managing of land resources and 
are increasingly vital for natural and regional development in 
Africa. 
Landuse characterization has been most used particularly in the 
developing countries, (Adesina and Amamoo 1994; Oyinloye and 
Adesina 2006). Differences in the spatial and spectral 
characteristics of features make a differentiation of 
landuse/landcover types possible. 
As a result of fundamental differences in the characteristics of 
the images used for this research, the study has proceeded with the 
assumptions that land use characterization will some how be 
different from one image to the other. Thus, SPOT, because of its 
higher resolution, would be more able to capture information than 
the two others. Also, Landsat would generate more details than 
NigeriaSat-1. 
In general, the point raised above was demonstrated in this 
study. For example SPOT image had higher correlation of 
important wave bands with each other than both Landsat and 
NigeriaSat-1. This implies that, SPOT captures more information 
than either of Landsat and NigeriaSat-1. Also with respect to 
landuse recognition, built-up areas were more efficiently identified 
and defined on the SPOT image than on the two others. However, 
the three satellite images were relatively inefficient in defining the 
other land use categories. It appears from the results that the 
superiority of enhanced spatial resolution declines with the size of 
the land use category to be defined. 
Table. 7 Areas of Landuse categories on the image 
LANUSE/LAND 
COVER 
CATEGORIES 
LANDSAT 
NIGERIASAT-1 
SPOT 
AREA 
(Ha) 
AREA 
% 
AREA 
(Ha) 
AREA 
% 
AREA 
(Ha) 
AREA 
% 
BUILT-UP AREA 
4388.15 
9.56 
4547.69 
10.14 
4891.85 
11.47 
BARE ROCK 
8925.80 
19.44 
8016.18 
17.88 
12887.70 
30.22 
FARM LAND 
9461.96 
20.62 
6682.62 
14.90 
7565.11 
17.74 
SEC. FOREST 
REGROWTH 
19881.90 
43.31 
23266.60 
51.89 
14665.70 
34.38 
WATER BODY 
3241.56 
7.07 
2326.32 
5.19 
2642.84 
6.20 
TOTAL 
45899.37 
100 
44839.41 
100 
42653.20 
100