| have been
; areas. The
mation about
. as about
P for a
a community
ise is
-ise about 4%
ped for the
t users of the
jernment
sriculture.
1d limitations
also to be
ohotographs
sidential and
mmercial)
ardens
lic grounds
hotographs
ually inter-—
» categories
| with the aid
1:10'000.
Afterwards a regular sampling grid of a width of 100 meters
was placed on the map. Land use is determined at every
sampling point and coded with a number. The data were entered
on a record sheet, transferred on punch cards and processed by
computer. The time needed for the data collection is shown in
table 2, the results obtained are shown in table 5.
Table 2:
Time spent on photo interpretation, transfer of land use
boundaries to map, and sampling with grid of 100 m.
photo interpretation with stereoscope 1 n. 20 min.
delineating of land use units
data transfer with ZOOM TRANSFER SCOPE i-2 h.
drawing and coloring of map 50 min.
sampling 1h. 20 min.
punching data 30 min.
TOTAL 5-6 h.
4.2 Transfer of a regular sampling grid to the aerial photo-
graphs
Instead of transferring the land use boundaries to a topo-
graphic map, the sampling grid can be transferred to the
aerial photographs. This point transfer can be achieved
analytically by computing the image coordinates of every
sample point. For this purpose the orientation elements of the
photographs and the elevation of the sample points are needed.
The orientation elements can be determined with the aid of
some control points for each photo, the elevations were taken
from a digital terrain model of a grid width of 250 m. As this
point distance did not coincide with the sampling pattern, a
height interpolation was required (10). After computing the
image coordinates of the sample points the sample grid can be
plotted with an automatic tracing table on a transparent foil,
which finally can be placed on the photos.
Compared with the procedure described before, this method has
several advantages:
- besides measuring the control points the creation of the
sample grid is fully automatic.