Posts: 5
Threads: 1
Joined: Jun 2019
24.06.2019, 13:20
(This post was last modified: 24.06.2019, 13:21 by oniregoc.)
Hello,
It looks like RouteConverter cant handle 1 million point.
for a topography project I need to retrieve 1 million elevation point on each km2 (1000x1000 points) of the map I am trying to build...I need all those point to build a 1m resolution map raster of a very large area...of course no map out of US get that resolution but what I a trying to do is an "interpolation map" on existing map.
But ! even 250000 points take ages to load and make the process unusable on large scale. It's working perfectly if i limit the points numbers to 100000 but means i have to do 10 files on each tile....and I have 400 tiles ! It seems that it's because the application try to display all the necessary points on the map. Is ther any way to get all those elevation data without the display of the points on the maps ? (it should reduce drastically the necessary memory to do the process)
Is there a trick I missed ? Any help ?
Thanks.
Posts: 1,033
Threads: 58
Joined: Jan 2011
(24.06.2019, 13:20)oniregoc Wrote: Is there a trick I missed ?
No.
(24.06.2019, 13:20)oniregoc Wrote: Any help ?
Make yourself familiar with GIS Software, e.g. QGIS, ArcGIS, Global Mapper and the like. RouteConverter is not suitable for what you are trying to achieve.
Apart from that the highest freely available resolution for elevation data is 30 m by 30 m => aprox. 1000 points per square km.
Grüße
Hans
Posts: 5
Threads: 1
Joined: Jun 2019
(24.06.2019, 14:48)nordlicht Wrote: (24.06.2019, 13:20)oniregoc Wrote: Is there a trick I missed ?
No.
(24.06.2019, 13:20)oniregoc Wrote: Any help ?
Make yourself familiar with GIS Software, e.g. QGIS, ArcGIS, Global Mapper and the like. RouteConverter is not suitable for what you are trying to achieve.
Apart from that the highest freely available resolution for elevation data is 30 m by 30 m => aprox. 1000 points per square km.
Thanks...but I know that except that the displayed maps in google maps, ArcGis topo are much more precise than the supposed 30x30 m, probably because their algorithm are doing some good interpolation. I use RouteConverter just to get the data I missed to do my raster afterword : elevation. So my real question was in fact: how can I make it work on a very large list without cutting my list in 10 ?! I tryed to use Qgis but I didn't find the way (if you have some tutorial)... also the last version is buggy which make the thing not that easy ! Thanks anyway for answering.
Posts: 7,532
Threads: 230
Joined: Aug 2007
(24.06.2019, 13:20)oniregoc Wrote: It looks like RouteConverter cant handle 1 million point.
I've optimized RouteConverter with a track of 380000 points. But you needs several GB of RAM for this.
(24.06.2019, 13:20)oniregoc Wrote: Is there a trick I missed ? Any help ?
Take a debugger and find the bottle necks ;-)
--
Christian
Posts: 7,532
Threads: 230
Joined: Aug 2007
Or have a look a the source code and build the elevation retrieval and storage (as GPX?) as a command line client.
--
Christian
Posts: 48
Threads: 5
Joined: Apr 2017
nordlicht Wrote:Make yourself familiar with GIS Software, e.g. QGIS, ArcGIS, Global Mapper and the like. RouteConverter is not suitable for what you are trying to achieve.
Correct. For things like that I'd also use GlobalMapper and QGIS that can handle such data amounts with ease.
Quote:Apart from that the highest freely available resolution for elevation data is 30 m by 30 m => aprox. 1000 points per square km.
Not true. Meanwhile there are many sources out there derived from LIDAR that have a much better resolution.
See e.g. https://www.data.gv.at/katalog/dataset/l...irolgelnde ,
https://data.opendataportal.at/dataset/dtm-germany ,
https://fbinter.stadt-berlin.de/fb/index.jsp (AKTIS DGM, 1m !)
...
Posts: 5
Threads: 1
Joined: Jun 2019
(27.06.2019, 13:40)SaschaT Wrote: nordlicht Wrote:Make yourself familiar with GIS Software, e.g. QGIS, ArcGIS, Global Mapper and the like. RouteConverter is not suitable for what you are trying to achieve.
Correct. For things like that I'd also use GlobalMapper and QGIS that can handle such data amounts with ease.
Quote:Apart from that the highest freely available resolution for elevation data is 30 m by 30 m => aprox. 1000 points per square km.
Not true. Meanwhile there are many sources out there derived from LIDAR that have a much better resolution.
See e.g. https://www.data.gv.at/katalog/dataset/l...irolgelnde ,
https://data.opendataportal.at/dataset/dtm-germany ,
https://fbinter.stadt-berlin.de/fb/index.jsp (AKTIS DGM, 1m !)
...
Hello SaschaT ! any chance you can provide a tutorial to do this in QGIS ? I have made a hudge search over the net and didn't find a solution...some exist but are using plugins that do not exist anymore.
I fact I think the real resolution is not a problem : when you look the Opentopomap they have done probably an "interpolation" map that gives the feeling it's very detailled ( about elevation I mean ) ... and the zone I would like is for sure not in public domain with a 1m detail like in us maps...so the idea is to do a "simulation map" from the exiting one. The one I get for now is quite good for what I would like to do but I am sure I can get much better if I can collect the necessary 1 000 000 points by km2, then glue the 529 needed tiles all together to make the .R16 file I need.
Posts: 1,033
Threads: 58
Joined: Jan 2011
04.07.2019, 07:31
(This post was last modified: 04.07.2019, 07:50 by nordlicht.)
(27.06.2019, 13:40)SaschaT Wrote: ....
https://data.opendataportal.at/dataset/dtm-germany ,
...
Sonny's data sets are scaled to SRTM1 resolution which is still the commonly available resolution for large areas, despite higher resolution data available locally.
So far the OP has not indicated which area he is looking at, so no chance to point him to higher resolution data.
My choice of tool would be Global Mapper as it has powerful and comfortable elevation data handling capabilities.
Grüße
Hans
Posts: 5
Threads: 1
Joined: Jun 2019
(26.06.2019, 14:18)routeconverter Wrote: (24.06.2019, 13:20)oniregoc Wrote: It looks like RouteConverter cant handle 1 million point.
I've optimized RouteConverter with a track of 380000 points. But you needs several GB of RAM for this.
(24.06.2019, 13:20)oniregoc Wrote: Is there a trick I missed ? Any help ?
Take a debugger and find the bottle necks ;-)
I see...380000 points...too bad for me. Will try with Qgis so. Thanks
Posts: 5
Threads: 1
Joined: Jun 2019
(04.07.2019, 07:31)nordlicht Wrote: (27.06.2019, 13:40)SaschaT Wrote: ....
https://data.opendataportal.at/dataset/dtm-germany ,
...
Sonny's data sets are scaled to SRTM1 resolution which is still the commonly available resolution for large areas, despite higher resolution data available locally.
So far the OP has not indicated which area he is looking at, so no chance to point him to higher resolution data.
My choice of tool would be Global Mapper as it has powerful and comfortable elevation data handling capabilities.
Hello, area is Hong-Kong. But using STRM or other data like these do not work for my project because probably I am not GIS expert: I absolutely need to be able to build my map with tiles that must join very precisely...and I don't know how to do this with Qgis and STRM maps...at the opposite I already found a process to do it differently. The only thing is, that I need to get elevation for each kms2 of my map with 1000x1000 points for each area...even it's not really accurate...my final goal is a simulation to be used in Unreal Engine. My actual map is only 100X100. I am sure that with 1002000 points I will get better results. About Global Mapper : any idea how to get elevation from opentopomap or openstreetmap for each point ?
|