[GRASS-dev] [GRASS-SVN] r62845 - in grass/trunk/scripts: . d.shadedmap r.shadedmap

Sun Nov 30 22:32:04 PST 2014

Vaclav,

Please see below
____________________
C. Michael Barton
Director, Center for Social Dynamics & Complexity
Professor of Anthropology, School of Human Evolution & Social Change
Head, Graduate Faculty in Complex Adaptive Systems Science
Arizona State University

voice:  480-965-6262 (SHESC), 480-965-8130/727-9746 (CSDC)
fax: 480-965-7671 (SHESC),  480-727-0709 (CSDC)
www: http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu

On Nov 30, 2014, at 10:12 PM, Vaclav Petras <wenzeslaus at gmail.com<mailto:wenzeslaus at gmail.com>> wrote:

On Sun, Nov 30, 2014 at 11:37 PM, Michael Barton <Michael.Barton at asu.edu<mailto:Michael.Barton at asu.edu>> wrote:
My thinking was that a relief map simulates the appearance of topographic relief through the use of ‘hill shading’ (to use the ESRI term).

BTW, hill shading is not as bad term as other ESRI terms, the main problems is that it is really just ESRI term. It seems to me that nobody else is using it.

You can “shade” the relief map made with r.relief or other psuedo relief maps, like an aspect map (can you shade a slope map in this way? Not sure what it would tell you visually) with a color map of some kind (hence r.shaded.relief). Or you can “drape” the color map over the relief map—perhaps “r.color.drape”.

I'm not sure if I follow the first part. But I don't like the drape think as a name and as a parameter (d.shade has now drapemap, I would like to get rid of it). I know what drape would mean with 3D surface (e.g. in NVIZ) but I have no idea what this mean with elevation and its shade. It always seems to me that I'm draping the shade over the elevation (as it would draping do in 3D) but it is the other way around.

Yes. the color is draped over the relief map.

I think I understand what you are saying now. You are referring to the process that modifies the map to achieve the result. The grey shading of the relief map alters the colors of the colored map (elevation, land-cover, etc). In fact, d.shaded.relief is a wrapper for d.his. The colored map serves as the hue (H of d.his) and the relief map is the intensity (I of d.his).  In fact, something like this is what is happening in NVIZ too. In this sense, r.shaded.relief is accurate in that grey-scale shading creates the appearance of topographic relief.

On the other hand, I am referring to what *appears* to be happening visually. The grey-scale relief map *appears* to be a 3D terrain. A colored map *appears* to be draped over (or color shading) this 3D terrain. So the terminology I was suggesting referred to the visual appearance, not the underlying imaging processing. Note that in most interface menus, the description refers to what appears to be going on visually, not to the underlying processes that produce the visual effects.

Slope shading or whatever we want to call it is possible in some way. Higher slopes are darker.

r.colors -n -e map=slope color=grey
d.shade drapemap=elevation reliefmap=slope brighten=99

I tried this with a grey scale applied to slope. I didn’t find it very informative. But perhaps others would. Aspect works pretty well, but that is because aspect shading is similar to relief shading.

See also "red relief image maps" [1] but slope is there used as color, not shade. I'll try to add both examples to manual some time later.

[1] https://groups.google.com/forum/?hl=en#!searchin/lastools/grass/lastools/_Zv6ublzeSE/SRCa_WQ8gusJ

This use is more informative, though for an image fusion method, there is some duplication in the information fused (i.e., slope and topographic relief). In any case, the color is still being ‘applied’ to the 3D terrain.

There are other kinds of image fusion for sure. The goal is to use names that users can easily understand. If d.shade can shade something other than a relief map of some kind, then perhaps d.shade is best. If it mainly is for shading relief maps maps, d.shaded.relief might be better.

It's mainly for relief but it's general. If there is no strong opposition I would stick to d.shade but yes I consider this as important because it seemed to me that it is hard find relief shading functionality in GRASS (and QGIS too actually). Nothing was backported yet.

d.shade is OK and short. But I worry a bit that it is not informative enough. d.shademap or d.shaderelief seems to me like it better conveys what the script is doing. But maybe it is very clear to most people. If so, shorter is better.

And what about the following?

- d.shade drapemap=elevation reliefmap=relief  ## shading a slope map is not going to be useful for most people

This is OK to me. It tells the user what input is expected and implies what is going to happen.

+ d.shade color=elevation shade=relief

This is confusing. While the relief map is altering the intensity in the resulting HIS image, it visually appears that the color is doing the shading of the relief map. I would express the command and its arguments to make sense in this second way. If you want to change the arguments, I’d suggest

d.shade color[or colormap or drapemap]=elevation relief[or reliefmap]=relief

Cheers
Michael

Michael
____________________
C. Michael Barton
Director, Center for Social Dynamics & Complexity
Professor of Anthropology, School of Human Evolution & Social Change
Head, Graduate Faculty in Complex Adaptive Systems Science
Arizona State University

voice:  480-965-6262<tel:480-965-6262> (SHESC), 480-965-8130<tel:480-965-8130>/727-9746 (CSDC)
fax: 480-965-7671<tel:480-965-7671> (SHESC),  480-727-0709<tel:480-727-0709> (CSDC)
www: http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu<http://csdc.asu.edu/>

On Nov 30, 2014, at 7:54 PM, Vaclav Petras <wenzeslaus at gmail.com<mailto:wenzeslaus at gmail.com>> wrote:

On Sun, Nov 30, 2014 at 8:12 PM, Michael Barton <Michael.Barton at asu.edu<mailto:Michael.Barton at asu.edu>> wrote:
My idea is that changing r.shaded.relief -> r.relief allows us to use r/d.shaded.relief for relief maps that are colored by another variable (elevation, land-cover, etc).

I don't think this would be a good idea. r.relief (former r.shaded.relief) creates "shade from the (terrain) relief". r.shade (former r.shadedmap, and d.shade) allows to "put the shade on a colored/colorful raster" where both shade and raster can be anything. Shade can be e.g. aspect, slope or shade derived from the relief. I know this is probably clear but his was my thinking.

I didn't though about r.shade and r.relief names before you suggested them and I'm not 100% satisfied with r.relief because it is in fact not relief but shade from relief. However, I like the new names because they are short, there is no problem with using or not using the dot between words and they are distinct from each other.

Michael
____________________
C. Michael Barton
Director, Center for Social Dynamics & Complexity
Professor of Anthropology, School of Human Evolution & Social Change
Head, Graduate Faculty in Complex Adaptive Systems Science
Arizona State University

voice:  480-965-6262<tel:480-965-6262> (SHESC), 480-965-8130<tel:480-965-8130>/727-9746 (CSDC)
fax: 480-965-7671<tel:480-965-7671> (SHESC),  480-727-0709<tel:480-727-0709> (CSDC)
www: http://www.public.asu.edu/~cmbarton, http://csdc.asu.edu<http://csdc.asu.edu/>

On Nov 30, 2014, at 5:50 PM, grass-dev-request at lists.osgeo.org<mailto:grass-dev-request at lists.osgeo.org> wrote:

From: Vaclav Petras <wenzeslaus at gmail.com<mailto:wenzeslaus at gmail.com>>
Cc: GRASS developers list <grass-dev at lists.osgeo.org<mailto:grass-dev at lists.osgeo.org>>
To: Markus Neteler <neteler at osgeo.org<mailto:neteler at osgeo.org>>
Date: November 30, 2014 at 5:31:35 PM MST
Subject: Re: [GRASS-dev] [GRASS-SVN] r62845 - in grass/trunk/scripts: . d.shadedmap r.shadedmap

On Sat, Nov 29, 2014 at 9:43 AM, Vaclav Petras <wenzeslaus at gmail.com<mailto:wenzeslaus at gmail.com>> wrote:

On Sat, Nov 29, 2014 at 9:30 AM, Markus Neteler <neteler at osgeo.org<mailto:neteler at osgeo.org>> wrote:
On Sat, Nov 29, 2014 at 12:37 AM, Vaclav Petras <wenzeslaus at gmail.com<mailto:wenzeslaus at gmail.com>> wrote:
>
> On Fri, Nov 28, 2014 at 5:40 PM, Michael Barton <Michael.Barton at asu.edu<mailto:Michael.Barton at asu.edu>>
> wrote:
>>
>> Probably nothing to be done about this since r.shaded.relief goes back a
>> long time. But a more sensible renaming would have the script now called
>> r.chaded.relief renamed to simply r.relief. Then the modules that drape a
>> colored map onto a relief map could be called x.shaded.relief,
>> x.shadedrelief, or something like that.
>>
> That's an interesting option. I'm not sure if I like it but it's worth
> exploring. If we go even further we get:
>
> r.shaded.relief -> r.relief
> r.shadedmap -> r.shade
> d.shadedmap -> d.shade

... along with:
- r.blend

Yes, I like this list better than the current names of discussed modules.

So, if you like it, I can like it too and I will change it hopefully soon. Just to be sure: including r.shaded.relief -> r.relief?

r.shaded.relief -> r.relief
r.shadedmap -> r.shade
d.shadedmap -> d.shade

Done in r63305 with few other improvements.

Now I'm not sure with [rd].shade parameters. reliefmap and drapemap are not particularly explanatory. My idea was shade and color which is better when shade=slope compared to relief=slope. color=elevation is also quite understandable but of course it is quite unexpected that the color option is a raster map.

There is also one detail I'm not sure about. r.relief might be now missing one keyword and that is "shaded relief". "relief" should be there too but having also "shaded relief" distinguishes it from things like "local relief (model)".

http://trac.osgeo.org/grass/changeset/63305

Vaclav

Markus

_______________________________________________
grass-dev mailing list
grass-dev at lists.osgeo.org<mailto:grass-dev at lists.osgeo.org>
http://lists.osgeo.org/mailman/listinfo/grass-dev

<slope_shading.png>

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/grass-dev/attachments/20141201/73065e1e/attachment-0001.html>