Hello,

You are right, there is a mistake in this functor. From looking at the code and the reference paper (here <http://www.lmars.whu.edu.cn/prof_web/zhangliangpei/rs/publication/Classification%20and%20Extraction%20of%20Spatial%20Features%20in.pdf>), the *m_Alpha* parameter is supposed to be the constant applied to the weighted mean. The parameter *m_DirectionStep* should be used instead.

I have also noticed something that bothers me : in otbSFSTexturesFunctor.h, line 134, the distance of the segment is computed between the _centre_ and the _last offset _in a particular direction. I don't see why the centre is be used here. I would rather use the _last offset in the opposite direction_. Otherwise, the length and width measures are meaningless.
Here is the approach I propose :
- let choose a set of 5 directions.
- the last offsets will be computed for a set of 5*2 angles, giving an angle step of : CONST_PI / 5 (as proposed by Julien Radoux) - the distance should be computed between the last offsets measured at two ends of the same direction

I have filled a bug report (http://bugs.orfeo-toolbox.org/view.php?id=955)

Any opinions on this ?

Best Regards,
Guillaume

Le 08/07/2014 16:47, Julien Radoux a écrit :
Hello,

I think that there is something wrong with otbSFSTexturesFunctor.h, but maybe I did not understand something. Shouldn't the angle be distributed between 0 and 2*PI instead of using the m_Alpha parameter? Can anyone double check ?

 //angle = m_Alpha * static_cast<double>(d);
  angle = CONST_PI * static_cast<double>(d) / m_NumberOfDirections;

Thanks,

Julien
--
--
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] <mailto:[email protected]>.
For more options, visit https://groups.google.com/d/optout.


--
<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]

--
--
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