[GIS] the best way to extract the geometry of a road from OSM data

geometrygeoprocessingopenstreetmap

What is the best way to extract the geometry of a road from OSM data? The goal is to display segments of highway that are closed due to weather. The weather data consists of positions (in KM measured along the road) from major cities, and length of the affected stretch of road. The main problem is clipping the road based on these distances; afterwards rendering seems straightforward.

Best Answer

You definitely want to get that data into Postgis using something like osm2pgsql or imposm (if all you have is an .osm extract) or shp2pgsql (if you have a shapefile). Once your data is in Postgis, you can use some of the linear referencing functions explained here, esp. ST_Line_Interpolate_Point() and ST_Line_Locate_Point():

http://postgis.net/docs/reference.html#Linear_Referencing

I imagine that the error in the linear referencing is going to be a major problem. If you have long distances in km and the OSM road is a different geometry to what the weather service uses, then you you'll see errors magnified the further you go.

Related Solutions

[GIS] Assessing OpenStreetMap coverage for international routing

There's no statistics on routing quality in different countries available, as far as I'm aware. The best thing you can do is try to measure things indirectly.

I organised a project to do this in the USA on OpenStreetMap during 2009. For full details see the 250 cities project on the OpenStreetMap wiki

First we measured how many of the 250 largest cities in the country were routable, in a large matrix of every possible start/finish point. Secondly, we found places where the outward/backward routes were significantly different lenghts. Then we looked at routes where the distance was greater than 1.4x of the great-circle route, as an indication of likely routing problems.

You could use a similar approach to estimate routing quality in your countries of interest. Your assumption that major cities and major roads are mapped first generally holds true for OpenStreetMap, in my experience.

[GIS] Why is sum of st_length() of segments 20% too big

You are not using geodesic functions to calculate the length, which means that for a point there is an error factor of:

cos( LATITUDE * pi() / 180 )

If you then multiply the calculated lenght by the error factor you should obtain a value pretty close to the actual trail length. For instance the Old Beechy Rail Trail is close to Melbourne, which has a latitude of ~ 37° which means:

cos( 37 * pi() / 180 ) ~ 0.798635510047293

in turn:

60 * 0.798635510047293 = 47.91 Km

which, depending on your needs and given the brutal rounding of Melbourne's lat that I took in the calculations, could be considered close enough to the reported official length of 46 Km.

This question has very useful information on the topic of geodesic measurements.

Best Answer

Related Solutions

[GIS] Assessing OpenStreetMap coverage for international routing

[GIS] Why is sum of st_length() of segments 20% too big

Related Question