[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Date: Thu, 23 Sep 2010 11:06:21 +0300 From: Maris Nartiss maris@gmail.com Subject: Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue To: nunosousa84 nuno_sousa_1...@hotmail.com Sorry for You. You do made backup of important data before messing around with QGIS, right? You can try to launch GRASS and rebuild vector topology with v.build. If it doesn't help - You'r SOL. And don't forget to infrorm QGIS developers how well their GRASS support works. Good luck, Maris. Hi all. Sorry I was unclear: it is a known bug for QGIS to crash in case of an unclean vector. There is no indication that QGIS is more inclined to spoil your vector than any other GRASS GUI. If you stop GRASS while it is writing a vector, problems are likely to occur, with any UI (even from the shell). You can clean your vector also from within QGIS, by simply running v.build.all. And yes, backupping data is always a good idea, regardless of what sw you are using. BTW: Maris, why are you trolling around? We do our best, like everybody, to improve the tools we are using, for the benefit of everybody, and it's no fun to read your polemic notes. All the best. -- Paolo Cavallini: http://www.faunalia.it/pc ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Micha wrote: We're used to perceiving shadows below and highlights above what is being lit. The same holds for a hill-shaded terrain. When we see shadows on a slope, it seems to be south facing, and highlights appear north facing. So if you place the sun at 130° (SE) then you are lighting the south facing slope, and making it appear north facing, thus causing the valleys to appear to be hills, and the peaks appear as valleys. We have a set of ortho-photos that were shot early in the morning (sun in the east) and a colleague always turns the maps upside down (N pointing down) so that the hills and valleys look right. The most natural looking hillshading is with azimuth between 270 - 315° (W to NW). note that in most places GRASS uses the cartesian convention of theta as degrees counter-clockwise from the positive x-axis, not navigational convention of degrees being measured clockwise from north. (convert between the two with '90-angle') By the way GRASS as any map composer like QGIS to be able to finish the work (put scale bar, north, legend, title,...)? Yes, QGIS is a good choice. GRASS does have a full set of map composer tools, but some graphic elements require a bit more fiddling to get them right. Mind you, the results can be quite rewarding. (see the help pages for the d.* modules and ps.map, and the cartography section of the grass web site) Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Sorry for You. You do made backup of important data before messing around with QGIS, right? You can try to launch GRASS and rebuild vector topology with v.build. If it doesn't help - You'r SOL. And don't forget to infrorm QGIS developers how well their GRASS support works. Good luck, Maris. 2010/9/22, nunosousa84 nuno_sousa_1...@hotmail.com: I can't believe this a nightmare just happened! This is ubelivable. Suddenly when i was working with the vector layer digitizing the values QGIS just crashed puff completely. Run out of the program. I tried to open QGIS it started again but opening the vector layers from the GRASS Plugin crashes it again, closing QGIS again, suddenly. And the biggest nightmare, this is really a nervous breakdown, since i have to deliver this work till suturday, the grass vector layers dessapeared from the mapset files!!! I have no words, im really afraid i've lost all the work. I still have the grass raster layers, well i think. So i think i can do the reverse process (from raster to vector) right? The problem is that i lost almost a day just doing vector adjustments and suddenly lost all that. Please if can anyone help me. Im afraid i wont be able to recover the layers. But how can they desapear!? Thank you -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5559644.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Thank you Maris. Well i was able to recover it yesterday. I will put here the error it gave me than the v.build solution. I have posted this in QGIS forum: This is what it gives back: terminate called after throwing an instance of 'QgsGrass::Exception' what(): Unable to open vector map cntouca_grassvec...@relevo on level 2. Try to rebuild vector topology by v.build. Aborted Then i runed v.rebuild: So i tried v.build: Coor files of vector map cntouca_grassvec...@relevo is larger than it should be (551 bytes excess) Building topology for vector map cntouca_grassvector... Registering primitives... 540 primitives registered Number of boundaries: 1 Number of centroids: 0 Building topology for vector map cntouca_grassvectorrebuild2... Number of centroids: 0 Number of areas: 0 Number of isles: 0 Number of incorrect boundaries: 1 Number of primitives: 1 Number of points: 0 Number of lines: 0 Number of boundaries: 1 0 areas built 0 isles built Attaching islands... Attaching centroids... Number of nodes: 2 Registering primitives... 1 primitives registered 2 vertices registered Building areas... Number of areas: 0 Number of isles: 0 Number of incorrect boundaries: 1 23243 vertices registered Building areas... 0 areas built 0 isles built Attaching islands... Attaching centroids... Number of nodes: 692 Number of primitives: 540 Number of points: 19 Number of lines: 520 So apparently the layer is back again and working, for now :-). It seems the file (coor) has a storing limit capacity that was surpassed (makes some sense cause it happened just when i was adding a vector point)? By the way i notice one think. When trying to open the recovered grass vector layer it seems he automatically devided it in two (one for the lines and one for the points). Well the problem is how am i gona edit a line/point layer (to solve those flat top surfaces)? I could use lines too but the effects would be similar (flat surfaces but smaller). Im afraid i will have to try your suggestion Sab (v.digit). I will see it tomorrow morning. For now thank you. Regards Nuno So aparently im stuck to how iam i gona be able to make a point/line layer. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5562140.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 09/23/2010 10:18 AM, nunosousa84 wrote: Thank you Maris. Well i was able to recover it yesterday. I will put here the error it gave me than the v.build solution. I have posted this in QGIS forum: This is what it gives back: terminate called after throwing an instance of 'QgsGrass::Exception' what(): Unable to open vector mapcntouca_grassvec...@relevo on level 2. Try to rebuild vector topology by v.build. Aborted Then i runed v.rebuild: So i tried v.build: Coor files of vector mapcntouca_grassvec...@relevo is larger than it should be (551 bytes excess) Building topology for vector mapcntouca_grassvector... Registering primitives... 540 primitives registered Number of boundaries: 1 Number of centroids: 0 Building topology for vector mapcntouca_grassvectorrebuild2... Number of centroids: 0 Number of areas: 0 Number of isles: 0 Number of incorrect boundaries: 1 Number of primitives: 1 Number of points: 0 Number of lines: 0 Number of boundaries: 1 0 areas built 0 isles built Attaching islands... Attaching centroids... Number of nodes: 2 Registering primitives... 1 primitives registered 2 vertices registered Building areas... Number of areas: 0 Number of isles: 0 Number of incorrect boundaries: 1 23243 vertices registered Building areas... 0 areas built 0 isles built Attaching islands... Attaching centroids... Number of nodes: 692 Number of primitives: 540 Number of points: 19 Number of lines: 520 So apparently the layer is back again and working, for now :-). It seems the file (coor) has a storing limit capacity that was surpassed (makes some sense cause it happened just when i was adding a vector point)? By the way i notice one think. When trying to open the recovered grass vector layer it seems he automatically devided it in two (one for the lines and one for the points). Well the problem is how am i gona edit a line/point layer (to solve those flat top surfaces)? I could use lines too but the effects would be similar (flat surfaces but smaller). Im afraid i will have to try your suggestion Sab (v.digit). I will see it tomorrow morning. For now thank you. I think Hamish mentioned this procedure: First digitize a new point vector map of the peaks, with elevation in an attrib. column. Then convert the points to raster, using the elevation attribute as raster value. Now merge this new point raster (it should have values of zero everywhere except the cells at mountain peaks) with your rasterized contours map. And finally recreate the DEM from this combined contour+peaks raster. Regards Nuno So aparently im stuck to how iam i gona be able to make a point/line layer. -- Micha Silver Arava Development Co. +972-52-3665918 http://surfaces.co.il ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
yes you're right, my distraction. Well i was able to work it out much more relaxed now :-) I still think the results from the shaded relief are a bit odd. For example if use 130º azimuth the river valleys appears above the near mountains, but if i use degrees between 270 and 0º or between 0 and 90º the river valleys appear below the rest like it should be. Is it normal? I will try to post a picture here. Im a bit concerned about this cause i will create the hidrographic system also (basins, ...). By the way GRASS as any map composer like QGIS to be able to finish the work (put scale bar, north, legend, title,...)? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5563815.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On Thu, Sep 23, 2010 at 3:35 PM, Micha Silver mi...@arava.co.il wrote: On 09/23/2010 10:18 AM, nunosousa84 wrote: ... By the way i notice one think. When trying to open the recovered grass vector layer it seems he automatically devided it in two (one for the lines and one for the points). Well the problem is how am i gona edit a line/point layer (to solve those flat top surfaces)? I could use lines too but the effects would be similar (flat surfaces but smaller). Im afraid i will have to try your suggestion Sab (v.digit). I will see it tomorrow morning. For now thank you. I think Hamish mentioned this procedure: First digitize a new point vector map of the peaks, with elevation in an attrib. column. Then convert the points to raster, using the elevation attribute as raster value. Now merge this new point raster (it should have values of zero everywhere except the cells at mountain peaks) with your rasterized contours map. And finally recreate the DEM from this combined contour+peaks raster. ... So aparently im stuck to how iam i gona be able to make a point/line layer. Please consider to add suggestions to a Wiki page. We have http://grass.osgeo.org/wiki/Digitizing_Area_Features maybe we need http://grass.osgeo.org/wiki/Digitizing_Line_Features ? Would be good to create a best practice collection. Markus ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 09/23/2010 06:46 PM, nunosousa84 wrote: yes you're right, my distraction. Well i was able to work it out much more relaxed now :-) I still think the results from the shaded relief are a bit odd. For example if use 130º azimuth the river valleys appears above the near mountains, but if i use degrees between 270 and 0º or between 0 and 90º the river valleys appear below the rest like it should be. Is it normal? I We're used to perceiving shadows below and highlights above what is being lit. The same holds for a hill-shaded terrain. When we see shadows on a slope, it seems to be south facing, and highlights appear north facing. So if you place the sun at 130° (SE) then you are lighting the south facing slope, and making it appear north facing, thus causing the valleys to appear to be hills, and the peaks appear as valleys. We have a set of ortho-photos that were shot early in the morning (sun in the east) and a colleague always turns the maps upside down (N pointing down) so that the hills and valleys look right. The most natural looking hillshading is with azimuth between 270 - 315° (W to NW). will try to post a picture here. Im a bit concerned about this cause i will create the hidrographic system also (basins, ...). By the way GRASS as any map composer like QGIS to be able to finish the work (put scale bar, north, legend, title,...)? Yes, QGIS is a good choice. GRASS does have a full set of map composer tools, but some graphic elements require a bit more fiddling to get them right. Mind you, the results can be quite rewarding. -- Micha Silver Arava Development Co. +972-52-3665918 http://surfaces.co.il ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
I can't believe this a nightmare just happened! This is ubelivable. Suddenly when i was working with the vector layer digitizing the values QGIS just crashed puff completely. Run out of the program. I tried to open QGIS it started again but opening the vector layers from the GRASS Plugin crashes it again, closing QGIS again, suddenly. And the biggest nightmare, this is really a nervous breakdown, since i have to deliver this work till suturday, the grass vector layers dessapeared from the mapset files!!! I have no words, im really afraid i've lost all the work. I still have the grass raster layers, well i think. So i think i can do the reverse process (from raster to vector) right? The problem is that i lost almost a day just doing vector adjustments and suddenly lost all that. Please if can anyone help me. Im afraid i wont be able to recover the layers. But how can they desapear!? Thank you -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5559644.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 20/09/2010 23:39, nunosousa84 wrote: Thank you Micha. Just to confirm that the output of slope module is: total null and non-null cells: 2256660 total null cells: 6164 These null cells should be *only* around the edges, where there is not a 3X3 window to calculate the slope. By default they'll show in an orange color in the display. Of the non-null cells: -- n: 2250496 minimum: 0 maximum: 78.8343 range: 78.8343 That's a pretty steep maximum slope. 78% slope is almost vertical. I wonder if the black spots are just very dark- that is points with a very steep slope, i.e. a cliff, that correctly show in a dark shade? Again these might be some misses in the digitizing process. mean: 29.4765 mean of absolute values: 29.4765 standard deviation: 17.1799 variance: 295.148 variation coefficient: 58.2834 % sum: 66336658.9338117689 I dont know maybe there is something wrong with my digitizing. Besides this problem do you think the result of the shaded relief command is acceptable? Well, that you'll have to say. You can improve the mesa tops effect by adding point altitudes to the contour vector at the peaks. See the manual page, and Hamish's explanation earlier. I dont know but it seems some parts of the map reveals some geometric artifacts, sometimes cutting through. -- Micha Silver http://www.surfaces.co.il/ Arava Development Co. +972-52-3665918 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Nuno wrote:
For all this steps i worked directly through
GRASS 6.4 0RC6 and not QGIS. I have opened the grass raster
layer with the contours (lines) and applyed the commands
suggested:
http://grass.osgeo.org/wiki/Contour_lines_to_DEM
The result was this:
http://osgeo-org.1803224.n2.nabble.com/file/n5549840/shadedtest1.jpg
that looks much better!
Well it seems the terrace effect desappeared but there are
some glintches, do you think its normal?
which glitches? the slight thatched-roof look? perhaps a quick
3x3 r.neighbors smoothing filer could fix that? maybe there is
an off-by-one bug somewhere that's causing that effect at the
contour lines; something for us to look at after floating point
and NULL support I guess.
Concerning the flat mesa tops they look really ugly, any
suggestion how to remake/remodel them?
I suppose i have to go to the original geotiff topography chart
and remake some vector lines to overcome that problem? What
should i do?
As for the mesa effect, that is because there is no input data
above it, and unlike a spine interpolation it has no memory about
the slope, so above the highest elevation contour given it does
not know to go any higher. As suggested by the r.surf.contour help
page, it really helps to digitize some trig station / mountain
peak elevation data and if possible spot elevations along ridge
lines to help with that. hopefully your scanned topo map has
some of that information.
individual raster cells are ok, they don't have to be lines.
('v.to.rast type=points' + r.patch to join the points and vector
lines rasters)
About the blackspots and the color table im not sure but
changing it doesnt overcome the problem, the spots remain the
same.
I'm not seeing them in the new version.. ?
About the Region Settings they were set in QGIS. Since the
region im working is fairly small i used this parameters:
N-S Res = 1.00014 and E-W Res = 1.00024 (I think this is the
minimum resolution supported).
grass doesn't really have a minimum or maximum resolution, just
that a huge number of cells ( 40,000 x 40,000 ) can be very
slow and take lots of RAM and disk space; and at the other end
the minimum you can have is 1x1 cell.
In grass you can try 'g.region -a res=1' to round outward to
whole numbers.
One problem i noticed and it might be a QGIS bug is that
everytime i open QGIS or add a layer it resets the
projection/coordination values (For example im using
Datum ED50 UTM zone 29 international ellipsoid but when i
open a layer or QGIS it always resets this values to others).
no idea, you'd have to ask a qgis expert about that.
grass Locations are locked to all maps using a single map
projection and you need to use r.proj, v.proj to move them
between different ones.
Hamish
___
grass-user mailing list
grass-user@lists.osgeo.org
http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
One more thank you for your patient :-). I just noticed (believe me i was sleeping, hehehe) that all the vector lines don't snap to the region boundaries. Well let me explain with an image: http://osgeo-org.1803224.n2.nabble.com/file/n299/Vectorlinesnotouching.jpeg You can see the vectorized lines not touching the region margins/boundaries. Could this be part of the problem? can i readapt the region settings to fit the lines or have to extend the lines? Regards Nuno -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p299.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Ok solved the problem of the null-cells. Did g.region.zoom to the shadedrelief layer and shrink the region boundaries, so now r.univar gives 0 null-cells :-). But i think i still have to solve the vector/raster lines problem that might be causing some artifacts on the boundaries of the map. Maybe extend them till they reach the region line?. Otherwise it seems to drag the values till they reach the region boundary. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p565.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
nuno wrote: One more thank you for your patient :-). I just noticed (believe me i was sleeping, hehehe) that all the vector lines don't snap to the region boundaries. Well let me explain with an image: http://osgeo-org.1803224.n2.nabble.com/file/n299/Vectorlinesnotouching.jpeg no, I don't think the vector line snapping is a problem. it seems that the black areas are slope independent and all north- facing aspect. I'm strongly betting on the just-out-of-range color map as the cause of the holes. e.g. range is floating point but max used in the color table is truncated to the integer just below the max. and anything out of range looks like a whole. I'd have expect those to be white though oh well. perhaps try tweaking the $MAPSET/colr/ file by hand and see if it helps. Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
nuno: Ok solved the problem of the null-cells. Did g.region.zoom to the shadedrelief layer and shrink the region boundaries, so now r.univar gives 0 null-cells :-). great But i think i still have to solve the vector/raster lines problem that might be causing some artifacts on the boundaries of the map. Maybe extend them till they reach the region line?. Otherwise it seems to drag the values till they reach the region boundary. perhaps just crop off a few edge cells and forget about them..? Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 19/09/2010 22:26, Hamish wrote: I wonder if the black-spots are a color table problem? what does r.univar say about the shaded relief map? That's what just occurred to me. If the spots just magically disappear in GRASS, then try setting a different colormap in QGIS. Micha Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user This mail was received via Mail-SeCure System. -- Micha Silver http://www.surfaces.co.il/ Arava Development Co. +972-52-3665918 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Thank you all. Well i have some news. Hamish i have followed your suggestion of using the mapcalc commands. For all this steps i worked directly through GRASS 6.4 0RC6 and not QGIS. I have opened the grass raster layer with the contours (lines) and applyed the commands suggested: http://grass.osgeo.org/wiki/Contour_lines_to_DEM The result was this: http://osgeo-org.1803224.n2.nabble.com/file/n5549840/shadedtest1.jpg Well it seems the terrace effect desappeared but there are some glintches, do you think its normal? Concerning the flat mesa tops they look really ugly, any suggestion how to remake/remodel them? I suppose i have to go to the original geotiff topography chart and remake some vector lines to overcome that problem? What should i do? About the blackspots and the color table im not sure but changing it doesnt overcome the problem, the spots remain the same. Here it is the r.univar report of the shadedrelief layer: total null and non-null cells: 2256660 total null cells: 6164 Of the non-null cells: -- n: 2250496 minimum: -75.1119 maximum: 100 range: 175.112 mean: 42.1701 mean of absolute values: 48.816 standard deviation: 35.304 variance: 1246.37 variation coefficient: 83.718 % sum: 94903752.6220222563 Now to the strange issue (QGIS Shaded Relief Plugin). Remember the error i posted earlier? Well it seems i was able to work with that plugin but only after i run the surf contour layer in GRASS. Let me explain i redone the steps decribed before as suggested by Hamish (mapcalc, surfcontour,etc...), all in GRASS 6.4 0RC6. Then changed to QGIS, opened the layer just made in GRASS and run the QGIS Shaded Relief Plugin. Now it worked. About the Region Settings they were set in QGIS. Since the region im working is fairly small i used this parameters: N-S Res = 1.00014 and E-W Res = 1.00024 (I think this is the minimum resolution supported). One problem i noticed and it might be a QGIS bug is that everytime i open QGIS or add a layer it resets the projection/coordination values (For example im using Datum ED50 UTM zone 29 international ellipsoid but when i open a layer or QGIS it always resets this values to others). Maybe i should save this parameters directly in GRASS? My dificulty is how, there are so many commands for a begginer :-) Regards and thank you Nuno -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5549840.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 20/09/2010 12:32, nunosousa84 wrote: Thank you all. Well i have some news. Hamish i have followed your suggestion of using the mapcalc commands. For all this steps i worked directly through GRASS 6.4 0RC6 and not QGIS. I have opened the grass raster layer with the contours (lines) and applyed the commands suggested: http://grass.osgeo.org/wiki/Contour_lines_to_DEM The result was this: http://osgeo-org.1803224.n2.nabble.com/file/n5549840/shadedtest1.jpg Well it seems the terrace effect desappeared but there are some glintches, do you think its normal? Concerning the flat mesa tops they look really ugly, any suggestion how to remake/remodel them? I suppose i have to go to the original geotiff topography chart and remake some vector lines to overcome that problem? What should i do? About the blackspots and the color table im not sure but changing it doesnt overcome the problem, the spots remain the same. Here it is the r.univar report of the shadedrelief layer: total null and non-null cells: 2256660 total null cells: 6164 There must be your black spots. But why you are getting null values, I can't say. Is it possible that you have accidentally made some contour lines that cross or touch? Of the non-null cells: -- n: 2250496 minimum: -75.1119 maximum: 100 range: 175.112 mean: 42.1701 mean of absolute values: 48.816 standard deviation: 35.304 variance: 1246.37 variation coefficient: 83.718 % sum: 94903752.6220222563 Now to the strange issue (QGIS Shaded Relief Plugin). Remember the error i posted earlier? Well it seems i was able to work with that plugin but only after i run the surf contour layer in GRASS. Let me explain i redone the steps decribed before as suggested by Hamish (mapcalc, surfcontour,etc...), all in GRASS 6.4 0RC6. Then changed to QGIS, opened the layer just made in GRASS and run the QGIS Shaded Relief Plugin. Now it worked. About the Region Settings they were set in QGIS. Since the region im working is fairly small i used this parameters: N-S Res = 1.00014 and E-W Res = 1.00024 (I think this is the minimum resolution supported). One problem i noticed and it might be a QGIS bug is that everytime i open QGIS or add a layer it resets the projection/coordination values (For example im using Datum ED50 UTM zone 29 international ellipsoid but when i open a layer or QGIS it always resets this values to others). Maybe i should save this parameters directly in GRASS? My dificulty is how, there are so many commands for a begginer :-) Regards and thank you Nuno -- Micha Silver http://www.surfaces.co.il/ Arava Development Co. +972-52-3665918 ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Good question Micha. Well i remember i checked the vector lines after completing them (things like not crossing, not touching and confirming that all of the lines has attributes to them, in this case altitude values). Then i checked overlapping raster contours in relation to the vector lines to see if there was any deviance from the original vector lines. I cant find any problem with this. mmm... what could it be? Somekind of overlapping cells? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5550439.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
nunosousa84 wrote: Good question Micha. Well i remember i checked the vector lines after completing them (things like not crossing, not touching and confirming that all of the lines has attributes to them, in this case altitude values). Then i checked overlapping raster contours in relation to the vector lines to see if there was any deviance from the original vector lines. I cant find any problem with this. mmm... what could it be? Somekind of overlapping cells? The way you can check is with v.clean. It has an option to break line vector at all line crossings and create a point vector of the crossings: v.clean your contour line vector out=contour_clean tool=break err=crossings If the resulting crossings vector has any points, compare them with the black spots. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Here are the results: http://osgeo-org.1803224.n2.nabble.com/file/n5551158/cleanresult.jpeg Well im not sure but at first glance doesnt seem to have much relation. This vector points layer that appear after runing the v.clean command shows also possible duplication of lines or just crossing lines? I have made a zoom to the vector lines where those 11 vector points are located and dont seem to recognize the problem. Here goes the comparation results for both r.surf.contour layer and shaded relief layer, using the r.univar query: (surfcontour layer) total null and non-null cells: 2256660 total null cells: 0 Of the non-null cells: -- n: 2256660 minimum: 430 maximum: 760 range: 330 mean: 541.46 mean of absolute values: 541.46 standard deviation: 66.726 variance: 4452.36 variation coefficient: 12.3234 % sum: 1221890343.312690258 ---// (shadedrelief layer) total null and non-null cells: 2256660 total null cells: 6164 Of the non-null cells: -- n: 2250496 minimum: -75.1119 maximum: 100 range: 175.112 mean: 42.1701 mean of absolute values: 48.816 standard deviation: 35.304 variance: 1246.37 variation coefficient: 83.718 % sum: 94903752.6220222563 It seems only when i run the command shaded relief that null cells appears. Is it normal? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5551158.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Thank you Micha. Just to confirm that the output of slope module is: total null and non-null cells: 2256660 total null cells: 6164 Of the non-null cells: -- n: 2250496 minimum: 0 maximum: 78.8343 range: 78.8343 mean: 29.4765 mean of absolute values: 29.4765 standard deviation: 17.1799 variance: 295.148 variation coefficient: 58.2834 % sum: 66336658.9338117689 I dont know maybe there is something wrong with my digitizing. Besides this problem do you think the result of the shaded relief command is acceptable? I dont know but it seems some parts of the map reveals some geometric artifacts, sometimes cutting through. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5552430.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Sorry didnt understand. You mean the GRASS plugin? available in QGIS. Plugins - Manage Plugins - Grass. Thats what im using. Working inside QGIS but through GRASS plugin. Sorry for any misunderstanding. Nuno -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5547491.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
There is a native QGIS plugins: http://spatialserver.net/pyqgis_1.0/contributed/shadedrelief.zip To activate it in QGIS: Fetch Plugins, a new window will pop up: Click on Repositories tab Add 3rd Party repository Click on Plugins Install DEM relief Shader You should be able to create relief shade as TIFF format from your dem. Cheers Sab On Sun, 2010-09-19 at 04:37 -0700, nunosousa84 wrote: Sorry didnt understand. You mean the GRASS plugin? available in QGIS. Plugins - Manage Plugins - Grass. Thats what im using. Working inside QGIS but through GRASS plugin. Sorry for any misunderstanding. Nuno ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Thank you Sab. I have tried to use the shaded relief plugin but it gives the following error: An error has occured while executing Python code: Traceback (most recent call last): File /home/nmc/.qgis//python/plugins/shadedrelief/ShadedReliefDialog.py, line 82, in accept myEngine.run() File /home/nmc/.qgis//python/plugins/shadedrelief/ShadedReliefEngine.py, line 118, in run self.outFileParam=os.path.normpath(str(self.outFileParam)) UnicodeEncodeError: 'ascii' codec can't encode character u'\xc1' in position 10: ordinal not in range(128) Python version: 2.6.5 (r265:79063, Apr 16 2010, 13:28:26) [GCC 4.4.3] Python path: ['/usr/share/qgis/python', '/home/nmc/.qgis//python', '/home/nmc/.qgis//python/plugins', '/usr/share/qgis/python/plugins', '/usr/lib/python2.6', '/usr/lib/python2.6/plat-linux2', '/usr/lib/python2.6/lib-tk', '/usr/lib/python2.6/lib-old', '/usr/lib/python2.6/lib-dynload', '/usr/lib/python2.6/dist-packages', '/usr/lib/python2.6/dist-packages/PIL', '/usr/lib/python2.6/dist-packages/gst-0.10', '/usr/lib/pymodules/python2.6', '/usr/lib/python2.6/dist-packages/gtk-2.0', '/usr/lib/pymodules/python2.6/gtk-2.0', '/usr/local/lib/python2.6/dist-packages', '/usr/share/qgis/python/plugins/fTools/tools'] Any help interpret this? -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5547792.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
i created a surface from rasterized contours (r.surf.contour). Yes i have acess to them i made them myself. First i created vector lines on top of a topography chart (digital geotiff) in QGIS. From then on i worked with GRASS plugin. Imported vector layer to GRASS, converted to Raster, used r.surf.contour and then r.shaded.relief. Following your suggestion I have tried v.surf.rst using my vector lines layer and the result seems more smooth, but those black spots still appear, more or less, depending on the value (degrees) of the altitude in relation to the horizon. By the way Sab im still having dificulties running shaded relief plugin even after installing python-numpy. Thank you Nuno -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5547899.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Well i dont know what to say but this is what i got working directly through GRASS 6.4.0RC6: http://osgeo-org.1803224.n2.nabble.com/file/n5548006/Grasstest.jpg Opening the same layer through QGIS gives those black holes/spots but through GRASS not. -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5548006.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
On 09/19/2010 06:47 PM, nunosousa84 wrote: Well i dont know what to say but this is what i got working directly through GRASS 6.4.0RC6: http://osgeo-org.1803224.n2.nabble.com/file/n5548006/Grasstest.jpg Opening the same layer through QGIS gives those black holes/spots but through GRASS not. Interesting... So you're saying that running r.shaded.relief with the QGIS GRASS plugin gives different results than running the module directly in GRASS? -- Micha Silver Arava Development Co. +972-52-3665918 http://www.surfaces.co.il ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
[GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Thats totally correct in what concerns those black holes/spots disapearing, besides that the results seem similar using r.shaded.relief or v.surf.rst as you suggested, in both GRASS plugin and GRASS. I normally dont work directly with grass, i find for now, since im a begginer, more easy to work with the grass plugin in QGIS. But i said to me well lets try and run these layers in GRASS. What i've found is that not only runing modules in GRASS makes those spots disappear instantanly but also just opening a layer that was previously manipulated in QGIS GRASS plugin, gives the same result. By the way this last image was using v.surf.rst through GRASS 6.4.0RC6 -- View this message in context: http://osgeo-org.1803224.n2.nabble.com/QGIS-GRASS-plugin-shadedrelief-issue-tp5547443p5548253.html Sent from the Grass - Users mailing list archive at Nabble.com. ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
Re: [GRASS-user] Re: QGIS GRASS plugin shadedrelief issue
Sab wrote: There is a native QGIS plugins: http://spatialserver.net/pyqgis_1.0/contributed/shadedrelief.zip To activate it in QGIS: Fetch Plugins, a new window will pop up: Click on Repositories tab Add 3rd Party repository Click on Plugins Install DEM relief Shader You should be able to create relief shade as TIFF format from your dem. note that (gdaldem/hillshade) is just based on repackaged GRASS algorithms, so a problem in one might also be seen in the other. - see http://grass.osgeo.org/wiki/Contour_lines_to_DEM r.surf.contour is the right tool to use here, but the user should pull off peak heights from the original topo map scan so that above the top contour is not a flat mesa. also the trick shown on that page should be used to multiply by a large number before running it, and then divide afterwards, to avoid the terracing effect which is caused by r.surf.contour only working with integer heights so far. I wonder if the black-spots are a color table problem? what does r.univar say about the shaded relief map? before running v.surf.rst (or any raster op.) be sure to run g.region to set up the computational region as desired. maybe your original topo map scan gives good values to match, although you'll probably want the resolution to be a nice round number (use the g.region res=.. -a flag). Hamish ___ grass-user mailing list grass-user@lists.osgeo.org http://lists.osgeo.org/mailman/listinfo/grass-user
