Re: [MORPHMET] Method to study outline data suffered from the 'Pinocchio effect'?

[Vincent Bonhomme]

Hi there,

As concerns outline analysis w/ EFT, my package Momocs does the job (last
version , CRAN
, vignette
,
doc ) and I'll be happy to help.

As concerns the Pinocchio effect, it once happened to me on exotic shapes,
but the 'nose' wasn't so Pinocchio so that was fine. You should have a try.

All the best,

vincent


2016-07-24 7:13 GMT+02:00 wongjinyung :

> Dear community,
>
> I'm studying outlines that have spines on them and the spines (especially
> some extreme cases with very long spines relative to the rest) affect the
> result of superimposition of the semi-landmarks quite a lot (a.k.a. the
> 'Pinocchio effect'), I hope I can get some suggestion on the method to use,
> as well as the software implementation. (I work on 2D outlines, currently
> using sliding semi-landmark method / elliptic Fourier analysis in R).
>
> I'm wondering whether there is an implementation of resistant-fit
> superimposition (with sliding) for semi-landmarks to deal with data like
> this? Or is there any other method that will not be affected by the
> 'Pinocchio' effect (as much)? Literature or experience on similar issue?
>
> Thanks,
> Wong Jin Yung
>
> --
> MORPHMET may be accessed via its webpage at http://www.morphometrics.org
> ---
> You received this message because you are subscribed to the Google Groups
> "MORPHMET" group.
> To unsubscribe from this group and stop receiving emails from it, send an
> email to morphmet+unsubscr...@morphometrics.org.
>



-- 
vincentbonhomme.fr 

-- 
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- 
You received this message because you are subscribed to the Google Groups 
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.

[MORPHMET] Method to study outline data suffered from the 'Pinocchio effect'?

[wongjinyung]

Dear community, 

I'm studying outlines that have spines on them and the spines (especially 
some extreme cases with very long spines relative to the rest) affect the 
result of superimposition of the semi-landmarks quite a lot (a.k.a. the 
'Pinocchio effect'), I hope I can get some suggestion on the method to use, 
as well as the software implementation. (I work on 2D outlines, currently 
using sliding semi-landmark method / elliptic Fourier analysis in R).

I'm wondering whether there is an implementation of resistant-fit 
superimposition (with sliding) for semi-landmarks to deal with data like 
this? Or is there any other method that will not be affected by the 
'Pinocchio' effect (as much)? Literature or experience on similar issue?

Thanks, 
Wong Jin Yung

-- 
MORPHMET may be accessed via its webpage at http://www.morphometrics.org
--- 
You received this message because you are subscribed to the Google Groups 
"MORPHMET" group.
To unsubscribe from this group and stop receiving emails from it, send an email 
to morphmet+unsubscr...@morphometrics.org.