Now its clear - and I understand ...!

Thanks alot

Ofer

On Tuesday, February 18, 2014 7:03:50 PM UTC+2, Guillaume Pasero wrote:
>
>  Hi,
>
> - 'band3' is not a typo. As said by Jonathan, band3 contains *NCC 
> metric*values. By creating a mask of pixel with a correlation score higher 
> than 
> 0.9 you can select only good matches (NCC optimum is 1).
>
> - Band1 contains *horizontal **disparities*. Checking pixel values means 
> that you should check that the disparity values obtained on your scene are 
> coherent with the search interval specified in the BlockMatching 
> application. For instance:  not setting the minimum horizontal disparity 
> low enough can prevent the matches on features with a high elevation. 
>
> - Band2 *does *contain *vertical disparities* (which is obviously 0 
> everywhere). When doing the BlockMatching step, you have to specify a 
> minimum and maximum vertical disparity for the exploration if you want to 
> see vertical residues. By default, min and max vertical disparities are 
> both set to 0 which imply no vertical exploration.
>
> Regards,
> Guillaume
>
> Le 18/02/2014 17:37, Ofer Zilberstein a écrit :
>  
> Hi Jonathan, 
>
>  Your answer confused me ... 
>
>   First - in the cookbook example  --  otbcli_BandMath -il 
> disparity_map_ncc.tif  -out thres_hdisparity.tif uint8  -exp 
> "if(im1b3>0.9,255,0)"  which mean that the h' disparity is in band 3 ??? is 
> it a typo error ? 
>
>
>            Second - checking the pixel values the disparity image the 
> values should not be between 0 -1 ?
>           
>           and finally - In my disparity image band 2 is empty ...
>
>  can you help me resolve this puzzle ?
>
>  Thanks in advanced
>
>  Ofer  
>
>  
>
>  
>  
>  
>  
>  
> On Monday, February 17, 2014 11:03:56 PM UTC+2, Jonathan Guinet wrote: 
>>
>>  Hi below my response about  Q1
>>
>>  Band1 : Horizontal Disparity map
>>  Band2 : Vertical Disparity Map
>>  Band3: Metric values
>>
>>  The BlockMatch output has 3 bands, the fourth is computed by Monteverdi 
>> and does not exist in input imagee
>>
>>  cheers
>>  
>>  
>>
>> 2014-02-17 17:33 GMT+01:00 Ofer Zilberstein <[email protected]>:
>>
>>>
>>> Hi  OTB uses,
>>>
>>>  I would like to share my experience and raise some Q' regard the 
>>> process of extraction DSM from Pleiades Stereo Model 
>>>
>>>  This is my first step toward change detection between Pleiades fresh 
>>> imagery & existing geo-data-base   
>>>
>>>  0) OS - Win 7 - 64bit
>>>
>>>  1) I have Pleiades Regular Stereo Model
>>>
>>>  2) I extracted ROI of 4000 x 4000 pixels for the left and then I 
>>> created ROI with -mode.fit option for the right (according to best 
>>> practice).
>>>
>>>  3) I started with StereoFramework, and after it crashed I followed 
>>> section 3.6  step by step and I successfully generated DSM file.
>>>
>>>  4) Although there are some typo errors (e.g - 
>>> in otbcli_StereoRectificationGridGenerator the -io.outimage1 should 
>>> be -io.outleft) the process went fine. 
>>>
>>>  *Q-1* - it is stated that otbcli_BlockMatching creates by default a 
>>> two bands image : the horizontal and vertical disparities. When I look at 
>>>  the image (e.g. disparity_map_ncc.tif) with Monteverdi I see 3 bends and 
>>> amplitude. 
>>>
>>>             bend 1 has values between -43 to 5
>>>             bend 2 is empty
>>>            bend 3 looks like disparity map with proper NCC values 0.5 to 
>>> 0.98 (I think that this is the horizontal)
>>>             bend 4 (named amplitude) has values between 0 to 40
>>>
>>>  In the step of filtering the NCC values (p. 81) , the example only 
>>> shows how to filter horizontal (bend 3) then concatenate thresholded 
>>> horizontal & vertical disparities...
>>>
>>>  However - I cannot figure out where the vertical disparity resides ?
>>>
>>>  ==================================
>>>
>>>  *Q-2* - Is it correct' or essential for accurate DSM  and ortho' to 
>>> follow chapter 3.2.5 Residual registration (between the two stereo images) 
>>> before the DSM extraction ?
>>>
>>>  Thanks in advanced ...
>>>
>>>  Ofer
>>>
>>>  -- 
>>> -- 
>>> 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/groups/opt_out.
>>>
>>  
>>   -- 
> -- 
> 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]<javascript:>
> To unsubscribe from this group, send email to
> [email protected] <javascript:>
> 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] <javascript:>.
> For more options, visit https://groups.google.com/groups/opt_out.
>
>
>
> -- 
>   <http://www.c-s.fr>  *Guillaume PASERO*
> Ingénieur d'études et développement
>  *Business Unit E-SPACE & Geo 
> Information*<https://thor.si.c-s.fr/blogs/cs-blogs-business/>* 
> - Département APPLICATIONS*
>
> *CS Systèmes d'Information*
> Parc de la Grande Plaine - 5, Rue Brindejonc des Moulinais - BP 15872
> 31506 Toulouse Cedex 05 - FRANCE
> +33 561 17 64 21 - [email protected] <javascript:>   
>  

-- 
-- 
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/groups/opt_out.

Reply via email to