[Qgis-user] Problem with raster calculator

George Silva georger.silva at gmail.com
Wed Apr 15 19:29:53 PDT 2015


Raster calculator has a nasty bug on 26 and 28 that its truncating values.
I dont have The correct number but I suggest you do a search on The
bugtracker and add dome weight to it.

it looks serious enough and its a blocker but without any comments.



On Wednesday, April 15, 2015, Michael Treglia <mtreglia at gmail.com> wrote:

> Hi All,
>
> Was encountering a similar problem with Raster Calculator [albeit a bit
> different] just yesterday - I had a layer that had values ranging ~0-1000
> (integers). I wanted the layer to represent probability values (ranging
> 0-1), so I used the raster calculator to divide it by 1000.
>
> I ended up with 'nan' values displayed as the max and min on the Layers
> pane and when I try to get into the properties I get a "Bad allocation"
> message.   Using the 'Identify Features' tool, the values look appropriate,
> but  I can't adjust the style or anything.
>
> I ended up doing what I needed in R for now, but it would be great to work
> through this issue - I'm open to any suggestions, or can send the data in
> case others want to test.
>
> Using QGIS 2.8.1 64-bit on Win7.
>
> Cheers,
> mike
>
> On Wed, Apr 15, 2015 at 4:30 AM, Ricardo Filipe Soares Garcia da <
> ricardo.garcia.silva at gmail.com
> <javascript:_e(%7B%7D,'cvml','ricardo.garcia.silva at gmail.com');>> wrote:
>
>> Hi Jesus
>>
>> Perhaps your input layer has no values that match the >= 73 condition?
>>
>> I can reproduce this behaviour if I use a layer in which pixel values go
>> up to 50 and I ask for pixels >= 60.
>>
>> Otherwise, if the input layer has values above the threshold, it seems
>> strange and I cannot seem to reproduce it here (QGIS 2.8.1)
>>
>>
>>
>>
>> On Wed, Apr 15, 2015 at 8:58 AM, Jesus Munoz <jesusmunozserrano at gmail.com
>> <javascript:_e(%7B%7D,'cvml','jesusmunozserrano at gmail.com');>> wrote:
>>
>>> Good morning list!!
>>>
>>> I am trying to do a cloud mask for a Landsat image based on the values
>>> of the pixels where the clouds are found. I am using the raster calculator
>>> for this purpose by introducing the expression:
>>>
>>>                 (“Band 1 layer name" >= 73) * "Band 1 layer name”
>>>
>>> So basically, I am trying to obtain a layer in which all pixels with a
>>> value lower than 73 are removed.
>>>
>>> However, after calculating this expression I only get an empty raster
>>> layer, only visible in the layers panel and populated only by “nan” values.
>>> Does anybody know what am I doing wrong???
>>>
>>> Jesus
>>> _______________________________________________
>>> Qgis-user mailing list
>>> Qgis-user at lists.osgeo.org
>>> <javascript:_e(%7B%7D,'cvml','Qgis-user at lists.osgeo.org');>
>>> http://lists.osgeo.org/mailman/listinfo/qgis-user
>>>
>>
>>
>>
>> --
>> ___________________________ ___ __
>> Ricardo Garcia Silva
>>
>> _______________________________________________
>> Qgis-user mailing list
>> Qgis-user at lists.osgeo.org
>> <javascript:_e(%7B%7D,'cvml','Qgis-user at lists.osgeo.org');>
>> http://lists.osgeo.org/mailman/listinfo/qgis-user
>>
>
>

-- 
George R. C. Silva
SIGMA Consultoria
----------------------------
http://www.consultoriasigma.com.br/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.osgeo.org/pipermail/qgis-user/attachments/20150415/3f48f386/attachment.html>


More information about the Qgis-user mailing list