ed in
and
than
nsion
hand,
s for
d.
),000
; and
with
arent
t 2-3
affin
nsed
ich is
etc.
lying
on of
on.
ases,
Itain
atum
maps
data
natic
OMs
enes,
o the
ry to
Trors
data
table
000 -
ation
| and
part
with
rk, a
to be
d by
ellite
nt to
erent
the
ve to
e Updating of existing digitized data. This method is efficient
when the revisions are limited. It should be noted that the
editing of data bases is highly time consuming.
e Interpretation of recent satellite images as the main source
of information. In some cases it could be necessary to use
existing maps as support for the interpretation.
e The use of already digitized data in cases when
interpretation of the object from the images is not possible,
such as contour lines and administrative boundaries.
Areas. A preferable way to start a project of the current
dimension is to separate the simplest area features from each
other, land vs. water. The following order of priority is adopted
for the features included in the projects.
1. Water, interpreted mainly from satellite images.
2. Agricultural areas, from satellite images with some help
from maps.
3. Urban areas and settlements, from a combination of satellite
images and maps.
4. Exploited areas, from satellite images and maps in
combination.
5. Mires and wetlands, from satellite images and maps in
combination.
All remaining information are classified as forest.
Lines are more time consuming to interpret than area
information owing to the need for external information to
achieve a satisfactory result. Road information, for example,
could be extracted from road maps, including different classes
of the object. Finally, field studies could be necessary to
investigate missing information.
Symbols. Most of the symbols are extracted from external
sources and then correlated to areas and lines interpreted from
the satellite images in combination with existing maps.
2.5.4 Materials and technique. Lines and areas are handled on
separate originals. This could create problems when a line, like
a road, coincides with a field boundary. In the data base these
two lines have to be identical, otherwise problems will occur
when finally creating a printed product from the data base.
Two main methods could be used during the collection and
conversion of analogue information to digital data.
* Digitizing and coding of the features from the original
manuscripts.
* Scanning the original, vectorizing and coding of the
features.
These methods are suggested to reduce the need for computer
power. A third method, including screen editing and
digitalization of objects, using a geometrically corrected
satellite image as background, requires powerful computers to
visualize the images.
2.6 Data Base and Map Design
One of the main advantages of storing map information in
digital data bases is the flexible post-processing of the data. It is
not possible to make any major changes easily in the layout of
traditionally created maps. Objects from digital bases could be
created in an endless number of designs, all due to the
customers’ requirements. It is also possible to select elements of
interest for the specific map product.
The development of printing equipment and the reduced
"fitting" problems between different printing originals has made
It possible to use the 4-colour printing technique. These
originals are usually combined with a separate original for the
black text to be included in the map.
2.6.1 Data base. A geographic data base traditionally includes
information digitized from maps. Due to this fact, it is
Important that the data base handler includes satisfactory
solutions on the following criteria.
457
* Flexible area selection from the base, map sheet
independent. Area boundaries selected by polygons or edge
coordinates.
Extraction of features of editing or mapping interest.
e Lock information for other users during the editing.
* Log files, describing feature events during the data base
history.
Elements included in the data bases produced by the Baltic
states are coded in accordance with the Swedish standard, KF-
> where all codes contain seven digits divided into three
evels
XX XX XXX
Title Object group Object
Example
Highway 30 11 000
Main road 30 13 000
Local road 30 24 000
Field/Forest road 30 27 000
Bridge 30 xx 230
All interpreted digital data are checked before being stored in
the Data Base.
2.6.2 Map Design. The final map design is a result of a
iterative work with employees from SSC Satellitbild, LM-
Kartor and most of all with representatives from the respective
Clients in the Baltic countries. A number of prototype maps
was produced, not only to evaluate single elements but also to
find a cartographic balance in the final map product. This
balance is of special interest in this case because the resulting
map includes an image background produced as a mosaic from
satellite images. The satellite background mosaic leaves the
interpretation to the end user, thereby compensating the reduced
number of features superimposed on the map.
Owing to the printing technique it was decided to present
contour lines in gray. The traditional colour of contour lines is
brown, but the 4-colour printing method would possibly have
caused matching problems between the printing originals
yellow, magenta and cyan.
All text has been printed in black, except names of water which
are presented in blue. Some names are rastered in the same
value as the boundary they correspond to, for example parish
names.
Due to the need for some special characters it was decided to
use PostScript format, which allows the user to create personal
letters, not included in US standard.
2.6.3 Image Background. Three versions of the black and
white satellite image are produced to optimize the visualization
of forests, agricultural areas and build-up areas. These features
are presented in different colours in the final product.
The 4-colour printing technology is based on the use of RGB-
complement colours in the subtractive process. This fact
explains the need for a program converting RGB colours to the
CMY(K) system allowing printing on paper. The software
market offers such facilities, but the current extreme case with
just green colour, in different brightness, presenting forest areas
must be handled as a special case of transformation between
these two colour systems. A simple but efficient method was
investigated and established to allow optimal vizualisation in
forest and agricultural areas.
Instead of transforming the whole "colour body", each channel
is converted from RGB, or more correctly the channels included
in the black and white RGB image, to either Cyan Magenta or
Yellow. This conversion is done by using a Look Up Table
(LUT) which describes the relationship between input and
output values of the gray levels in the image to be processed.
The information from 3*256 gray-levels in the RGB system has
to be inverted to the CMY system according to the different
ways to create colours. For example, O intensity is represented
as black in RGB and as 100 percent in the inverted CMY. This
step is included in the LUT used for the transformation from
RGB to CMY.
International Archives of Photogrammetry and Remote Sensing. Vol. XXXI, Part B4. Vienna 1996
