2004 
  
lected 
dness 
1010S: 
Wing 
high 
s the 
as, 
It and 
were 
, this 
uring 
ficult, 
on of 
| and 
some 
d, 
ected 
perly 
using 
The 
d by 
were 
some 
tures 
ation 
ee of 
The 
The 
and 
some 
o the 
ature 
International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, Vol XXXV, Part B4. Istanbul 2004 
  
was really under the vector. Also opening/closing of 
the vector data takes too much time, 
3. Deletion of the inexistent data takes too much time, 
also there were hardware problems arousing from 
high dimension of the ortophoto and vector data, 
4. Definition of the type of the roads were very difficult, 
but this difficulty will be overcome by time, 
5. Detection of stony places, arks, bridges, vents, 
perennial and intermittent streams were very difficult, 
6. It was nearly impossible to detect the features like 
walls, hedges, wire fences whereas they were still 
difficult in stereo models, 
7. There were nearly 25 and 20 meters displacement in 
both X and Y directions respectively. 
The following results were obtained from the comparison of 
orthophoto against stereo compilation: 
|. The difficulties met are nearly the same as 2.2.1 1 and 
2, 
An advantage of the method was that if the revision 
period was short, so there were few changes. But, in 
our map for some watering project, the area was 
changed greatly. The watering system could not be 
detected. 
n2 
2.3.2 Editing of the Second Map with a Hardness Degree 
of 2 by Using Orthophotos from 25K Aerial Photos: The 
compilation by editing was completed in 90 hours. The 
following problems have aroused during the compilation: 
|l. There were a lot of difficulties due to the lack of 3d 
dimension, 
2. There were 1-25 meters displacements in cartographic 
features, 
3. Editing of these displaced features takes too much 
time, 
4. Some high buildings hindered the some details, such 
as roads, 
5. The detection of electric lines was very difficult 
except great ones, 
6. There were some discrepancies between cartographic 
features and photogrammetric features, 
7. "The compilation of streams in wooded areas was very 
difficult, 
8. The superhighways in orthophotos show cut places 
due to the discrepancy of digital elevation model of 
the orthophoto, 
9. The data was very intense that it took much time to 
pan the view. 
The following results were obtained from the comparison of 
orthophoto against stereo compilation: 
I. The difficulties met are nearly the same as 2.2.1 1 and 
2 
“; 
2. An advantage of the method was that if the revision 
period was short, so there were few changes. But, in 
our map for some watering project, the area was 
changed greatly. The watering system could not be 
detected, 
3. By the assumption of the operator that cartographic 
features were in correct placement, even necessary 
editing was not done. 
687 
2.4 Method 3: Editing by Using Stereo Models 
In this method, stereo models were produced in Softplotter 
software automatically by using the results of the triangulation. 
The data to control was prepared by the Cartography 
Department in ArcInfo format. Then they were imported into 
KDMS format. Two dimension data were converted into 3D by 
using existing digital elevation model. The resulting map was 
converted into MicroStation DGN. 
2.4.1 Editing of the First Map with a Hardness Degree of 
2 by Using Stereo Models from 35K Aerial Photos: The 
compilation by editing was completed in 20 hours. The 
following problems have aroused during the compilation: 
1. There were some problems because of the conversion, 
2. Some 3D problems in the features converted from 2D 
appeared during stereo editing due to discrepancy of 
the existing digital elevation model. 
The following results were obtained from the comparison of 
editing against stereo compilation: 
1. The problems aroused in other methods related with 
detection were not met, 
2. The method was successful except some 3D faults. 
2.4.2 Editing of the Second Map with a Hardness Degree 
of 4 by Using Stereo Models from 35K Aerial Photos: The 
compilation by editing was completed in 45 hours. The 
problems aroused were the same as 2.4.1 during the 
compilation. 
The following results were obtained from the comparison of 
editing against stereo compilation: 
1. The problems aroused in other methods related with 
detection were not met, 
2. The method was successful except some 3D faults. 
3. By the assumption of the operator that cartographic 
features were in correct placement, even necessary 
editing was not done. 
2.4.3 Editing of the Second Map with a Hardness Degree 
of 4 by Using Stereo Models from 25K Aerial Photos: The 
compilation by editing was completed in 72 hours. The 
following problems have aroused during the compilation: 
1. The features had nearly 20 meters displacements 
especially in rapidly changing areas, 
An area of 7 km? was compiled from the beginning, 
25K scale made easy the detection in revision. 
Ww N 
The results are the same as the 2.4.2 in comparison of editing 
against sterco compilation. 
2.5 Method 4: The Compilation by Using 25K Stereo 
Models 
In this method, stereo models were produced in Softplotter 
software automatically by using the results of the triangulation. 
The compilation was completed in 75 hours. Here are the some 
results: