[Qgis-user] Imagine raster details in QGIS v ArcGIS

john polo jpolo at mail.usf.edu
Tue Aug 30 11:12:32 PDT 2016


Mike and Nicolas,

Thanks for taking the time to explain all of this to me, it did clear up 
some misunderstandings I had.

best regards,
John

On 8/30/2016 12:54 PM, Michael Treglia wrote:
> Hey John,
>
> I've dealt with this layer a bit too - one thing for QGIS is that when 
> you import a layer, I think the default to calculate the min/max based 
> on the 2-98% range of the data. If you're in the Layer Properties -> 
> Style tab, expand the 'Load min/max values' area, and then select 'Min 
> / max' (and change Accuracy to 'Actual'), and then hit the load 
> button. That yields the range as 207-15516. (see screen-shot below). I 
> know this is a categorical raster, so displaying based on the color 
> ramp is meaningless, but that at least shows you're getting the right 
> values.
>
> Re the projection - I thought this used to display correctly based on 
> the projection parameters that get read in, but it's displaying way 
> far north. Toying around, it looks like the equivalent EPSG code is 
> 5070 (+proj=aea +lat_1=29.5 +lat_2=45.5 +lat_0=23 +lon_0=-96 +x_0=0 
> +y_0=0 +ellps=GRS80 +towgs84=0,0,0,0,0,0,0 +units=m +no_defs)
>
> FWIW, I did this on QGIS 2.16.1
>
> Hope that helps,
> mike
>
>
> Inline image 1
>
> On Tue, Aug 30, 2016 at 12:23 PM, Nicolas Cadieux 
> <nicolas.cadieux at archeotec.ca <mailto:nicolas.cadieux at archeotec.ca>> 
> wrote:
>
>     Hi,
>     The easy answer is no:(
>
>     That could happen If the nature of the file was different (like a
>     simple picture .jpg) but in this case, as in most, you have
>     choices to make and assumptions are made by the programs based
>     partly on the user settings. The problem is that most rasters used
>     in GIS have more than the usual 256 values you find in a picture.
>     Therefore, you need to select what part of the image you want to
>     display or render on the screen which is limited to 256 values only.
>
>     As an example, you can ask QGIS to automatically open a raster
>     using the min-max pixel or you can ask Qgis to cut off the higher
>     and lower 1%. (Also, min max values can be real (slow) or
>     estimated (fast)). The second option may be better if you know you
>     have noise in the data.  To have a true idea of what the real min
>     max values are, you need to run the statistics. You can't rely on
>     rendering values shown under the layers.
>
>     The problem you are having with the CRS is probably that ArcGis or
>     ERDAS, is coding the information differently than in Qgis. Some
>     software companies do it differently either because they want to
>     lock you in a proprietary format or because they think their
>     format is better.    If I recall, ERDAS does not put the
>     information in the .tiff but rather, in a text file.  (I may be
>     wrong here.)  In any case, look an compare the CRS definitions.
>     They maybe identical.  You can also click on the layer and specify
>     the CRS. If you want to change the CRS, you will need to use the
>     save as option.
>
>     ArcGIS also has a concepts of dealing with rasters that have
>     unique values.  The idea, is that you can have a raster with a
>     data base.  After that, pixels with the value of 1, for example,
>     could be identified as "Roads". That concept does not exist in
>     Qgis (as of 2.12).  You can however, make a colour palette for the
>     unique values but Qgis will not id the values as objects like
>     roads or building.  I imagine this will come soon if it's not
>     already in 2.16.
>
>     Unfortunately, all those choices used for rendering get saved in
>     the project files and while most GIS will open a great number of
>     file formats correctly, most are completely incapable of reading
>     each other's project files:(. If they can, they have very limited
>     options.
>
>     Hope this clarifies things a bit.
>     Nicolas
>
>
>     Envoyé de mon iPad
>
>     Le 30 août 2016 à 11:24, john.polo [via OSGeo.org
>     <http://osgeo.org>] <[hidden email]
>     <http:///user/SendEmail.jtp?type=node&node=5283092&i=0>> a écrit :
>
>>     Hi,
>>     I downloaded the raster at this site:
>>     http://www.wildlifedepartment.com/facts_maps/ecoregions.htm
>>     <http://www.wildlifedepartment.com/facts_maps/ecoregions.htm>
>>     (scroll down about halfway to find the raster link. The file is
>>     kind of
>>     big, ArcGIS says the uncompressed size is 5.69 GB, 225 MB as a zip)
>>     and loaded it into QGIS as a standalone raster file. It is a
>>     raster of
>>     Oklahoma, USA classified into vegetation types at 10mx10m
>>     resolution.
>>
>>     It is saved as ERDAS Imagine, if I understand the extension. I don't
>>     know if the file type is relevant to the issues I am running into.
>>
>>     When the file is open in QGIS, the min and max pixel values have a
>>     smaller range, 314 - 14797, than when the file is open in ArcGIS,
>>     207 -
>>     15516. I can redraw the raster in ArcGIS to "Unique" instead of
>>     "Stretched" pixel values in the Symbology, which will then give me
>>     classifications for the pixels. Additionally, the CRS is
>>     USER:100002 in
>>     QGIS, but Albers Conical Equal Area in QGIS. Why does this
>>     difference in
>>     rendering happen? How can I get QGIS to render the file with the
>>     same
>>     characteristics as ArcGIS automatically? I know I can save with a
>>     new
>>     CRS to fix that issue, but if the file can load in ArcGIS with
>>     the CRS
>>     from the start, shouldn't QGIS render it the same? I don't know
>>     how to
>>     get the pixel values to the same range or how to get the
>>     classifications
>>     assigned.
>>
>>     John
>>
>>     _______________________________________________
>>     Qgis-user mailing list
>>     [hidden email]
>>     <http:///user/SendEmail.jtp?type=node&node=5283073&i=0>
>>     List info: http://lists.osgeo.org/mailman/listinfo/qgis-user
>>     <http://lists.osgeo.org/mailman/listinfo/qgis-user>
>>     Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-user
>>     <http://lists.osgeo.org/mailman/listinfo/qgis-user>
>>
>>     ------------------------------------------------------------------------
>>     If you reply to this email, your message will be added to the
>>     discussion below:
>>     http://osgeo-org.1560.x6.nabble.com/Imagine-raster-details-in-QGIS-v-ArcGIS-tp5283073.html
>>     <http://osgeo-org.1560.x6.nabble.com/Imagine-raster-details-in-QGIS-v-ArcGIS-tp5283073.html>
>>
>>     To start a new topic under Quantum GIS - User, email [hidden
>>     email] <http:///user/SendEmail.jtp?type=node&node=5283092&i=1>
>>     To unsubscribe from Quantum GIS - User, click here.
>>     NAML
>>     <http://osgeo-org.1560.x6.nabble.com/template/NamlServlet.jtp?macro=macro_viewer&id=instant_html%21nabble%3Aemail.naml&base=nabble.naml.namespaces.BasicNamespace-nabble.view.web.template.NabbleNamespace-nabble.view.web.template.NodeNamespace&breadcrumbs=notify_subscribers%21nabble%3Aemail.naml-instant_emails%21nabble%3Aemail.naml-send_instant_email%21nabble%3Aemail.naml>
>>
>
>     ------------------------------------------------------------------------
>     View this message in context: Re: Imagine raster details in QGIS v
>     ArcGIS
>     <http://osgeo-org.1560.x6.nabble.com/Imagine-raster-details-in-QGIS-v-ArcGIS-tp5283073p5283092.html>
>     Sent from the Quantum GIS - User mailing list archive
>     <http://osgeo-org.1560.x6.nabble.com/Quantum-GIS-User-f4125267.html>
>     at Nabble.com.
>
>     _______________________________________________
>     Qgis-user mailing list
>     Qgis-user at lists.osgeo.org <mailto:Qgis-user at lists.osgeo.org>
>     List info: http://lists.osgeo.org/mailman/listinfo/qgis-user
>     <http://lists.osgeo.org/mailman/listinfo/qgis-user>
>     Unsubscribe: http://lists.osgeo.org/mailman/listinfo/qgis-user
>     <http://lists.osgeo.org/mailman/listinfo/qgis-user>
>
>


-- 
The Tao never does anything,
yet through it all things are done.
-Lao Tzu, from a translation by S. Mitchell




More information about the Qgis-user mailing list