|
Automatic digitalization workflow is considered for
an example of a topographical map on the 1:24000 scale
being stored
in a foreign fund of digital data.
To "reduce all the data to a common denominator",
all rasters of the fund were normalized, i.e. they
have the same 18-color palette. Raster resolution is
254 dpi.
Due to such a radical palette reduction, most objects
in the raster are represented by sets of the same colors
with predominance of one or another color.
Trying to extract relief isolines from the raster,
you capture inevitably some other objects - inscriptions,
grid lines, etc.
A similar situation can result from low printing quality
of source materials, interpenetration of colors, tint
forming by color dilution with white dots, etc. Till
recently, such rasters could be processed semi-automatically
only, and the manual labor share was great.
|