655 
In: Wagner W„ Szflcely, B. (eds.): ISPRS TC VII Symposium - 100 Years ISPRS, Vienna, Austria, July 5-7, 2010, IAPRS, Vol. XXXVIII, Part 7B 
land cover classification. The capability of narrow bands in the 
hyperion data will be assessed to map the land cover in general 
and the biodiversity of tropical plants in particular. 
2. MATERIALS 
2.1 Study Area 
The study area for this research is in Bogor Botanical Garden 
area and surrounding, Indonesia (Figure 1). Bogor Bitanical 
Garden is managed by the Ministry of Forestry in Indonesia and 
contains more than one thousand plant species. The area covers 
about 1.2 X 1.2 kilometer squares, and this is equivalent to 
about 1600 pixels in the Hyperion image. From Figure 1 it can 
be seen that the only homogenous type of vegetation covering at 
least one pixel in Hyoerion is the grass. Therefore, in this 
exercise biodiversity will focus on extracting limited vegetation 
cover, but will be extended to cover more land cover such as 
manmade (asphalt), residential/roof, water (freshwater), etc. 
698000 698500 699000 699500 
9271000 
9270500 
9270000 
9269500 
Figure 1. Bogor Botanical Garden, viewed from IKONOS 
image, West Java, Indonesia (UTM Zone 48S). 
2.2 Hyperion Data 
The hyperion data were acquired on 13 September 2002 with 
less than 10% of cloud cover. Unfortunately, some cloud covers 
some little parts of the botanical garden. The image can be seen 
in Figure 2. The zoomed image display in RGB for VNIR and 
SWIR bands can be seen in Figure 3 and 4. 
Figure 3. Zoomed Hyperion Image of Bogor Botanical Garden 
and surrounding areas using RGB bands 29, 23 and 
16 (VNIR). 
Figure 4. Zoomed Hyperion Image of Bogor Botanical Garden 
and surrounding areas using RGB bands 204, 150 
and 93 (SWIR). 
9330000 
9320000 
9310000 
9300000 
9290000 
9280000 
9270000 
9260000 
9250000 
9240000 
_ -9270000 
-I- . .9260000 
680000 690000 700000 710000 720000 730000 
Figure 2. Whole scene of Hyperion Image of Bogor, Jakarta and 
surrounding areas using RGB combination of bands 
24, 12 and 8 (Botanical Garden in red circle) Image 
was obtained from http://eol.usgs.gov. 
3. METHODS 
The general procedures for classifying the Hyperion data can be 
seen in Figure 5. The procedures mainly consist of data 
preprocessing that include conversion of data into absolute 
radiance, atmospheric correction to get reflectance value and 
classification using the spectral angle mapper method. 
Figure 5. Classification procedures using spectral angle mapper 
from Hyperion satellite data.