The International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences. Vol. XXXVII. Part B4. Beijing 2008
486
completed after applying cartographic and topologie correction
and control on the processed data by the operators.
The workflow of feature extraction and the time spent in
workflow steps for any of the 1:25.000 scaled maps with the
mean difficulty level in the sense of production are depicted in
Table 1.
Feature Extraction Steps
Duration
(hour)
Creating stereo models (approximately 9 pieces)
1
Feature extraction
88
Control and correction of DECM data
16
Controlling the extracted and edited data on the
stereo models
4
Format conversion
1
Cartographic and topologic editing of extracted
data
40
Total
150
Table 1: The workflow and time spent table for
photogrammetric feature extraction and data editing procedures
in the 1:25.000 scaled map production.
As it is seen from Table 1, the revision of a 1:25.000 scaled map
lasts for approximately 19 days. This period doesn’t include the
field study after the feature extraction and the cartographic
works. It can easily be deduced that this process is quite
expensive for any organization producing maps. The 1:25.000
scaled map, produced after the workflow given in Table 1 in the
Figure 1 : The 1:25.000 scaled map produced in the present
production system.
3. A NEW 1:25.000 SCALED PHOTOGRAMMETRIC
MAP PRODUCTION SYSTEM
In this study, for the aim of reducing the production duration,
operator interaction and costs, a new production system will be
introduced as an alternative for the current photogrammetric
feature extraction system.
In this proposed system, it is aimed primarily to extract features
(roads, lakes etc.) that can be extracted semi-automatically from
orthophoto images via semi-automatic data collection software
developed and completion by operator of missing point features
and line features (narrow tracks, hedges etc.) that can not be
extracted automatically.
To achieve this aim, DEM was generated from 1:25.000 scaled
DECM of the same map and this DEM data was modified by
editing on the stereo models generated from exterior orientation
parameters determined from bundle block adjustment and
dry/wet streams were added to DEM after feature extraction as
breaklines. By using DEM data, orthophoto images with a
resolution of 1 m. were generated from aerial photographs which
cover the map sheet. Besides, contour lines at 10 m. intervals
were generated automatically from the same elevation data and
auxiliary contour lines were added via editing on stereo models.
Photogrammetric production process of a 1:25.000 scaled map
is planned to be completed after combining all data extracted
automatically, semi-automatically and manually, then checking
the data by means of topology and cartography and making
necessary corrections interacted with operator and software.
Photogrammetric processes and time spent for updating
1:25.000 scaled map sheet via production/update system that is
suggested as alternative are shown at Table 2.
Feature Extraction Steps
Duration
(hour)
Creating stereo models (approximately 9 pieces)
1
Making DTM from DECM, Control of the DTM
data on stereo models and addition of the vertex
lines.
8
Orthophoto image and making mosaic
4
Vector data generation with semi-automatic data
generation software which build on mosaic.
60
Control of the generated data and completion of
missing data.
4
Generating contours from DTM automatically,
controlling contours on stereo models, completion
of missing data and correction of errors.
8
Controlling the extracted and edited features on
stereo models.
4
Format Conversions
1
Cartographic and topologic edits of generated
data.
24
Total
114
Table.2. Photogrammetric generation and data edit processes of
a 1:25.000 scaled map revision and process times in available
workflow.
As seen on table2, revision of a 1:25.000 scaled map with
suggested revision system takes approximately 14 work days
(excluding topographic and cartographic works).
1:25.000 scaled map produced with suggested production
system and the process steps listed on Table 2 is shown in
Figure 2.
Summary information about semi-automatic data generation
software which is used in suggested map production method is
presented below.