Hi,

I thought that maybe the rasterization could be problematic (8-connexion, 
...) so I tried to apply a segmentation algorithm to the rasterized image. 
After the two commands that I used previously :
otbcli_ExtractROI -in 20120814_ZE_FUS_sudest_24bitscolor.img -out 
crop_0_0.tif -startx 0 -starty 0 -sizex 1701 -sizey 1782
otbcli_Rasterization -in MR_160sh5.shp -out test2.tif -im crop_0_0.tif 
-mode attribute -mode.attribute.field "ID"

I used the command :
otbcli_LSMSSegmentation -in test2.tif -out test_seg.tif uint32 -ranger 0.1 
-spatialr 1
in order to split regions that could be ill defined in the rasterized 
image. This actually works, I don't have the problem 
with otbLabelObjectToPolygonFunctor anymore.

Does anyone has an idea of what could be wrong with the rasterization step ?

Best regards, 

Bruno

Le vendredi 9 mai 2014 08:17:20 UTC+2, Bruno Belarte a écrit :
>
> Hi,
>
> Thank you anyway Rashad!
> I will do more tests to try to find what is wrong with my piece of code.
>
> Best regards,
>
> Bruno
>
> Le mercredi 7 mai 2014 11:40:43 UTC+2, Rashad a écrit :
>>
>> Hi,
>>
>> Compiling your code gives me the same error but I dont have any solution 
>> for now. Maybe someone in the list will have some information.
>>
>> On Wed, May 7, 2014 at 9:32 AM, Bruno Belarte <[email protected]>wrote:
>>
>>> I forgot to mention: I'm on ubuntu 12.04, with OTB 4.0, gdal 1.10
>>>
>>> Le mercredi 7 mai 2014 09:19:26 UTC+2, Bruno Belarte a écrit :
>>>
>>>> Hi,
>>>>
>>>> Here is a minimal code that reproduce the error. 
>>>>
>>>> The exact commands I typed for my testings are following:
>>>> otbcli_ExtractROI -in 20120814_ZE_FUS_sudest_24bitscolor.img -out 
>>>> crop_0_0.tif -startx 0 -starty 0 -sizex 1701 -sizey 1782
>>>> otbcli_Rasterization -in MR_160sh5.shp -out test2.tif -im crop_0_0.tif 
>>>> -mode attribute -mode.attribute.field "ID"
>>>>
>>>> Where 20120814_ZE_FUS_sudest_24bitscolor.img is a Pleiades image 
>>>> and MR_160sh5.shp is one of the two shapefiles. Both shapefiles and the 
>>>> image were produced by colleagues, I can ask them about how they did the 
>>>> segmentations.
>>>>
>>>> I also tried to do the same processes with ogr/gdal, the commands are:
>>>> ogr2ogr -clipsrc 1050016 6847483 1050116 6847583 extract.shp 
>>>> sertit_classification_2012_ZoneSE.shp
>>>> gdal_rasterize -a ID -te 1050016 6847483 1050116 6847583 -ts 200 200 
>>>> -ot Uint32 sertit_classification_2012_ZoneSE.shp crop.tif
>>>> Coordinates are given in the same projection as the input image, and 
>>>> the results are consistent in both Monteverdi and QGIS.
>>>>
>>>> Le mardi 6 mai 2014 20:29:43 UTC+2, Rashad a écrit :
>>>>>
>>>>>
>>>>>
>>>>>
>>>>> On Tue, May 6, 2014 at 3:03 AM, Bruno Belarte <[email protected]>wrote:
>>>>>
>>>>>> Hi,
>>>>>>
>>>>>> Here is a short extract of one of the shapefiles. I tested it again 
>>>>>> with my code and it still failed with the error itk::ERROR: 
>>>>>> LabelObjectToPolygonFunctor(0x7fff1701bc00): End point not with +/-1 
>>>>>> line from line
>>>>>>
>>>>>> I am not able to reproduce this error. Could you post a more complete 
>>>>> code.?
>>>>> and the exact command line you used for:
>>>>> rasterization
>>>>> segmentaion
>>>>>  
>>>>>
>>>>>> Le vendredi 2 mai 2014 19:12:01 UTC+2, Rashad a écrit :
>>>>>>>
>>>>>>>
>>>>>>> Hi,
>>>>>>>
>>>>>>> On Fri, May 2, 2014 at 8:51 AM, Bruno Belarte 
>>>>>>> <[email protected]>wrote:
>>>>>>>
>>>>>>>> Hi everyone,
>>>>>>>>
>>>>>>>> I have a piece of code that behaves differently with differents 
>>>>>>>> segmentations. The problematic piece of code can be found here: 
>>>>>>>> http://pastebin.com/Gfa9TdPb
>>>>>>>>
>>>>>>>> Segmentation 1: LabelImage generated  with the large scale 
>>>>>>>> meanshift framework from a Pleiades image (pan-ms fusion)
>>>>>>>> Segmentation 2 and 3: two LabelImage generated with the 
>>>>>>>> otbcli_Rasterization application from two different shapefiles
>>>>>>>>
>>>>>>>>
>>>>>>> Is it possible for you to provide your data to test?. shapefiles 
>>>>>>> would be fine.
>>>>>>>  
>>>>>>>
>>>>>>>>  With segmentation 1 everything works fine: the LabelMap and the 
>>>>>>>> corresponding polygons are computed and can be used further in the 
>>>>>>>> code.
>>>>>>>> With segmentation 2 and 3 the code failed at line 19 with the 
>>>>>>>> following error:
>>>>>>>> /home/belarte/OTB/src/OTB/Utilities/ITK/Modules/Core/Common/
>>>>>>>> src/itkMultiThreader.cxx:359:
>>>>>>>> itk::ERROR: MultiThreader(0x1cdcf90): Exception occurred during 
>>>>>>>> SingleMethodExecute
>>>>>>>> /home/belarte/OTB/src/OTB/Code/OBIA/otbLabelObjectToPolygonFunctor
>>>>>>>> .txx:559:
>>>>>>>> itk::ERROR: LabelObjectToPolygonFunctor(0x7fff1701bc00): End point 
>>>>>>>> not with +/-1 line from line
>>>>>>>>
>>>>>>>> The program crashes while computing the polygons from the 
>>>>>>>> LabelImage. My guess was that some of the polygons in the shapefile 
>>>>>>>> were 
>>>>>>>> problematic, so I tried to simplify the polygons with the ogr2ogr 
>>>>>>>> application with different *simplify* parameters (0.001 0.01 0.1 
>>>>>>>> 0.5 and 0.75) and then use the rasterization application but it did 
>>>>>>>> not 
>>>>>>>> change the result. I also try to use the 
>>>>>>>> otbcli_LSMSSmallRegionsMerging 
>>>>>>>> with different mininmum region sizes (10 15 20 100 and 1000) after the 
>>>>>>>> rasterization, but again no changes.
>>>>>>>>
>>>>>>>> Does someone have an idea of what could be the problem here?
>>>>>>>>
>>>>>>>> Thanks for your help,
>>>>>>>>
>>>>>>>> Bruno
>>>>>>>>
>>>>>>>> -- 
>>>>>>>> -- 
>>>>>>>> Check the OTB FAQ at
>>>>>>>> http://www.orfeo-toolbox.org/FAQ.html
>>>>>>>>  
>>>>>>>> You received this message because you are subscribed to the Google
>>>>>>>> Groups "otb-users" group.
>>>>>>>> To post to this group, send email to [email protected]
>>>>>>>>
>>>>>>>> To unsubscribe from this group, send email to
>>>>>>>> [email protected]
>>>>>>>>
>>>>>>>> For more options, visit this group at
>>>>>>>> http://groups.google.com/group/otb-users?hl=en
>>>>>>>> --- 
>>>>>>>> You received this message because you are subscribed to the Google 
>>>>>>>> Groups "otb-users" group.
>>>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>>>> send an email to [email protected].
>>>>>>>>
>>>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>>
>>>>>>> -- 
>>>>>>> Regards,
>>>>>>>    Rashad
>>>>>>>  
>>>>>>  -- 
>>>>>> -- 
>>>>>> Check the OTB FAQ at
>>>>>> http://www.orfeo-toolbox.org/FAQ.html
>>>>>>  
>>>>>> You received this message because you are subscribed to the Google
>>>>>> Groups "otb-users" group.
>>>>>> To post to this group, send email to [email protected]
>>>>>> To unsubscribe from this group, send email to
>>>>>> [email protected]
>>>>>> For more options, visit this group at
>>>>>> http://groups.google.com/group/otb-users?hl=en
>>>>>> --- 
>>>>>> You received this message because you are subscribed to the Google 
>>>>>> Groups "otb-users" group.
>>>>>> To unsubscribe from this group and stop receiving emails from it, 
>>>>>> send an email to [email protected].
>>>>>> For more options, visit https://groups.google.com/d/optout.
>>>>>>
>>>>>
>>>>>
>>>>>
>>>>> -- 
>>>>> Regards,
>>>>>    Rashad
>>>>>  
>>>>  -- 
>>> -- 
>>> Check the OTB FAQ at
>>> http://www.orfeo-toolbox.org/FAQ.html
>>>  
>>> You received this message because you are subscribed to the Google
>>> Groups "otb-users" group.
>>> To post to this group, send email to [email protected]
>>> To unsubscribe from this group, send email to
>>> [email protected]
>>> For more options, visit this group at
>>> http://groups.google.com/group/otb-users?hl=en
>>> --- 
>>> You received this message because you are subscribed to the Google 
>>> Groups "otb-users" group.
>>> To unsubscribe from this group and stop receiving emails from it, send 
>>> an email to [email protected].
>>> For more options, visit https://groups.google.com/d/optout.
>>>
>>
>>
>>
>> -- 
>> Regards,
>>    Rashad
>>  
>

-- 
-- 
Check the OTB FAQ at
http://www.orfeo-toolbox.org/FAQ.html

You received this message because you are subscribed to the Google
Groups "otb-users" group.
To post to this group, send email to [email protected]
To unsubscribe from this group, send email to
[email protected]
For more options, visit this group at
http://groups.google.com/group/otb-users?hl=en
--- 
You received this message because you are subscribed to the Google Groups 
"otb-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to [email protected].
For more options, visit https://groups.google.com/d/optout.

Reply via email to