Hello,

I am building Garmin map of Greenland with contour using OSM data and data from http://www.viewfinderpanoramas.org/ for creating contours.
On many places in northern part of Greenland data from these two sources doesn't match.

Screenshot:
https://plus.google.com/photos/104247758975882866759/albums/5804312979465166881

Jonathan de Ferranti (creator of this excellent contours data source) is convinced that his data are right and discrepancies seen on map are caused by placement errors in the OSM coastline data.

How can I check accuracy of OSM coastline data in this area and possibly fix it?

Many thanks for help.

mira

asked 31 Oct '12, 22:46

miramikes's gravatar image

miramikes
26224
accept rate: 0%

hey mira, did you manage to sucessfully create that map? i would be interested in it. could you share it? thankyou

(18 Dec '14, 19:28) xox

Hey xox,

I build this map 2 years ago and there were a lot of updates since in OSM data that time. For e.g. problem described in this thread has been fixed ;) Also some improvements has been done in contour data. In other words it would be good to build this map again with fresh data. BTW if I remember well my Greenland map with contours has been about 1.3 GB. If you want do it yourself you can use my "styles" if you wish. In README file you will find simple "how-to" https://github.com/miramikes/garmin_hiking_map

regards

mira

(20 Dec '14, 12:00) miramikes

thank you, i already tried it before, but failed. with your instructions, styles and data there is hope! i'll play around with it in the next days… merry christmas!

(22 Dec '14, 11:04) xox

hey xox, if there will some problem have look in OSM Garmin forum. http://forum.openstreetmap.org/viewforum.php?id=26 I have learned everything about Garmin maps there myself ... many helpful people there and more appropriate place to discus Garmin related topics. merry christmas ;)

(22 Dec '14, 11:30) miramikes

For checking the placement accuracy of the osm-data I would suggest to compare your contour data at a location, that is well mapped, has some gps-tracks and high-res Bing coverage, e.g. here. It is indeed well possible that large amounts of greenlands coastline data in osm has some accuracy issues, since it is imported from pgs.

Fixing it will require quite some work. Where Bing and gps-tracks are available you may use this source. If the license of viewfinderpanoramas is compliant with osm's odbl (which I do not know) you may use this as a source as well. Otherwise you may take your gps, a really warm down jacket and head out for a long, long walk on the coast...

permanent link

answered 01 Nov '12, 08:36

FischersFritz's gravatar image

FischersFritz
19117
accept rate: 0%

Yes I did check other places too and 95% of Grenland map is totally fine for me. Problems start from 81°N and going more to the north are worse.

(02 Nov '12, 11:30) miramikes

The imagery Bing has of the region in question is quite poor (as it is on Google, too). The Landsat imagery doesn't even reach to this latitude. But the OSM coastlines too have note=may be inaccurate on them.

An other issue is the projection. With my little understanding I can just tell that the Google Mercator projection distorts land masses when it comes close to the poles. It could be possible that the the contour lines are projected with some other parameters then OSM. So it could be that a comparison more southwards cannot tell you if it is in the north just as good. But that's just me; I hope some people more familiar with that issue can say something.

A comparison of OSM and that contour lines on the Islands south of Cape Horn would be interesting, too

permanent link

answered 02 Nov '12, 02:37

malenki's gravatar image

malenki
4.7k24683
accept rate: 6%

I there will be somebody who has a skills to fix this kind of issue I can provide same "test map" for southern areas close to south pole too.

(02 Nov '12, 11:36) miramikes

I've just found this thread where somebody fixed coastline by using contour data.
http://forum.openstreetmap.org/viewtopic.php?id=4075
(last message in thread)

I think I could manage to provide data where will be just 0m contour for given area which should be actually coastline. Not sure how to convert to "pseudo-gps-track" and how to use the for fixing OSM data.

mira

permanent link

answered 02 Nov '12, 11:50

miramikes's gravatar image

miramikes
26224
accept rate: 0%

Screenshot of coastline (0m contour)
Here with base map. https://plus.google.com/u/0/photos/104247758975882866759/albums/5806262493089876753

Would be possible to use "0m contour" for fixing OSM coastline?

mira

(02 Nov '12, 17:09) miramikes
1

Just a note: viewfinderpanorama contours are very unlikely to be a suitable data source for OSM. The data is gathered from a wide variety of sources, many of which are probably copyrighted (e.g. former Soviet maps which are based on other topo maps).

(18 Dec '14, 20:27) SK53 ♦

Hi,

I am not a lawyer, but I am not sure if manually convert paper maps to digital data is illegal.

mira

(20 Dec '14, 13:15) miramikes
Your answer
toggle preview

Follow this question

By Email:

Once you sign in you will be able to subscribe for any updates here

By RSS:

Answers

Answers and Comments

Markdown Basics

  • *italic* or _italic_
  • **bold** or __bold__
  • link:[text](http://url.com/ "title")
  • image?![alt text](/path/img.jpg "title")
  • numbered list: 1. Foo 2. Bar
  • to add a line break simply add two spaces to where you would like the new line to be.
  • basic HTML tags are also supported

Question tags:

×533
×289
×59
×23
×22

question asked: 31 Oct '12, 22:46

question was seen: 5,653 times

last updated: 22 Dec '14, 11:30

powered by OSQA