data file coflation?
robert sechrist
the files to a common format is a solved problem, as is removal of duplicate
records. We are, however, seeking to solve the problem of records representing
the same features but with slightly different coordinates.
A simple example involves a street segment. The first file identifies its
location as x1,y1 x2,y2. the second file identifies its location as x1,y1
x3,y3. The from points are identical but the to points are different by
some small amount.
Right now we are attempting to remove these near duplicate records by hand
IS THERE A BETTER WAY?
bob sechrist
new user
Bob,
It was with some interest I read your e-mail. This is a common problem
that many GIS users are looking to solve. Intergraph has a new product
called MGE Dynamic Data Integrator that addresses this need by automating
conflation workflows.
For those unfamiliar with the term, conflation refers to the merging
of spatially coincident data sets into a single file. You may have
for instance two files, one with accurate geometry but no attributes,
and a second with valuable attribution but innaccurate graphics. The
merging of these two datasets to come up with a single file with
accurate graphics and attribution is refered to as conflation. Imagine
having to copy the links from each graphic element in the innacurate
file to is corresponding feature in the accurate file.
The potential is huge for all application areas and is especially
suited to TIGER data where the linework was never designed for
cartographic purposes but the attributes are extremely useful.
For example, transportation agencies need to merge TIGER's address ranges
and census attributes to existing high quality map graphics. They do
not want to use TIGER's inaccurate graphics.
Metropolitan area local and municipal governments need to merge addresses
for dispatch management, sanitary sewer management, water utilities, and
bus and transit customer service.
Commercial companies like utility companies, phone, gas electric, need to merge
addresses onto existing Department of Transportation Maps for GIS.
By federal law all airports need this functionality to manage the
businesses and census data information around the airport's noise zones.
These examples only address the market for TIGER data. Conflation
also provides an excellent solution map accuracy, maintenance and update costs.
For example, you have a municipal parcel database that was digitized off
inaccurate parcel map sheets and have collected new more accurate information
through resurvey, GPS, aerial photos and want to update your existing
database without sacrificing data.
Or perhaps you have data in different vendor formats that need to be
combined into a single file...
If you or anyone is interested I have a paper that describes in more
detail the process entitled:
Conflation: the Merging of Cartographic and Attribute Data--
A Case Report
by Eric Swanson and Virginia Peterman
Peter Crosbie
Product Marketing Manager
GIS Analysis Products Phone: (205) 730-7534
Mapping Sciences Division Fax : (205) 730-1263
Intergraph Corportion Mail Path: ptcr...@ingr.com
Roy Chung
please send me a copy of the Conflation report mentioned inyour
recent e-mail. Thank you
Roy Chung
Geography Dept., Univ. of Northern Iowa
Cedar Falls, IA 50614-0406
Pat McClanahan
I must have missed the original post on this topic. I think (:-) I
am interested in the Conflation report too. Please contact me
so I can find out more about the report.
Thanks.
Pat McClanahan Internet:mccl...@dlgeo.cr.usgs.gov
EROS Data Center mccl...@edcserver1.cr.usgs.gov
Sioux Falls, SD
605-336-4601
