IsOuterBoundaries do not give the expected outcome

[Remsdel]

Function isOuterBoundary gives me a 2D logical array unlike the isBoundary which gives the expected logical column array of the same length at the number of grains. The 2D array has a number of columns equals to the number of grains but numbers of lines (which is quite large) seem unrelated to any other dimensions. It is unclear how to extract relevant data from it.

[kilir]

Hi,

this (6baac7a) should fix it.

Cheers,
Rüdiger

[Remsdel]

It does indeed put it back to the right format. However, identification of grains touching outer boundary appears to fail sometimes for some of my maps. See illustrations attached. I could not yet figure out why.

[kilir]

In all cases, you computed grains from EBSDsquare/gfridified EBSD data, correct?

[Remsdel]

Indeed, they were computed from the gridified EBSD data

[kilir]

Ok, does it make a difference if you increase 'delta' to 1 i.e. 'grains.isOuterBoundary('delta',1)' ?

If this doesn't do the trick, any chance you could share the data?

[Remsdel]

I did try it but no difference. Which data do you need? Only the grains variable? From: Rüdiger Kilian ***@***.***> Sent: Monday, March 16, 2026 3:10 PM To: mtex-toolbox/mtex ***@***.***> Cc: Delville Rémi ***@***.***>; Author ***@***.***> Subject: Re: [mtex-toolbox/mtex] IsOuterBoundaries do not give the expected outcome (Discussion #2525) Ok, does it make a difference if you increase 'delta' to 1 i.e. 'grains.isOuterBoundary('delta',1)' ? If this doesn't do the trick, any chance you could share the data? — Reply to this email directly, view it on GitHub<#2525 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASW2MWJ5RTAIFRNXUY4OVE34RADKZAVCNFSM6AAAAACWTK6ZLCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMJWGA3DEMQ>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[kilir]

Yes, only grains should be fine.

[Remsdel]

Attached From: Rüdiger Kilian ***@***.***> Sent: Monday, March 16, 2026 3:19 PM To: mtex-toolbox/mtex ***@***.***> Cc: Delville Rémi ***@***.***>; Author ***@***.***> Subject: Re: [mtex-toolbox/mtex] IsOuterBoundaries do not give the expected outcome (Discussion #2525) Yes, only grain should be fine. — Reply to this email directly, view it on GitHub<#2525 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASW2MWLQELMMT5XA7KKXX434RAEOFAVCNFSM6AAAAACWTK6ZLCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMJWGA3TCOI>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[kilir]

You may have to zip the .mat in order to be able to upload it.

[Remsdel]

Sorry, Here attached
grains.zip

[kilir]

Ok,

while at first sight this looks like a bug in extern/insidepoly, for me it works properly with matlab/inpolygon. The choice is made in mtex_settings.

So if you set in mtex_setting

mtex/mtex_settings.m

Line 173 in 6baac7a

setMTEXpref('insidepoly',true)

to false it should do what it is supposed to do.

You can change the setting also by running just

`setMTEXpref('insidepoly',false)`

in your script.

Could you confirm this?

[Remsdel]

Yes, this solves the issue indeed. From: Rüdiger Kilian ***@***.***> Sent: Monday, March 16, 2026 4:19 PM To: mtex-toolbox/mtex ***@***.***> Cc: Delville Rémi ***@***.***>; Author ***@***.***> Subject: Re: [mtex-toolbox/mtex] IsOuterBoundaries do not give the expected outcome (Discussion #2525) Ok, while at first sight this looks like a bug in extern/insidepoly, for me it works properly with matlab/inpolygon. The choice is made in mtex_settings. So if you set in mtex_setting https://github.com/mtex-toolbox/mtex/blob/6baac7a7575bf4673c357b37648e57cc370230a5/mtex_settings.m#L173 to false it should do what it is supposed to do. You can change the setting also by running just `setMTEXpref('insidepoly',false)` in your script. Could you confirm this? — Reply to this email directly, view it on GitHub<#2525 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASW2MWL6XMEEMLSXZJVWDJT4RALNFAVCNFSM6AAAAACWTK6ZLCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMJWGE3DAMA>. You are receiving this because you authored the thread.Message ID: ***@***.******@***.***>>

[kilir]

I think I found a reason for this behavior. Would you mind sharing a part of the EBSD data (entire map or simply the rightmost part of the map) so that we might better understand what causes those issues?

[Remsdel]

ebsd.zip

[kilir]

Thanks a lot. It seems the reason for the the observed behaviour related to (all so common) rounding errors in the coordinates of the EBSD input. I guess the plan was to do 0.25 µm steps.

See here a workaround when you would want to use the faster 'insidepoly'.
#2527 (comment)

[Remsdel]

I shall send it tomorrow, thanks

…

________________________________ From: Rüdiger Kilian ***@***.***> Sent: Tuesday, March 17, 2026 12:08:55 PM To: mtex-toolbox/mtex Cc: Delville Rémi; Author Subject: Re: [mtex-toolbox/mtex] IsOuterBoundaries do not give the expected outcome (Discussion #2525) I think I found a reason for this behavior. Would you mind sharing a part of the EBSD data (entire map or simply the rightmost part of the map) so that we might better understand what causes those issues? — Reply to this email directly, view it on GitHub<#2525 (reply in thread)>, or unsubscribe<https://github.com/notifications/unsubscribe-auth/ASW2MWOQ223QZYNQHBOEZRD4REW4PAVCNFSM6AAAAACWTK6ZLCVHI2DSMVQWIX3LMV43URDJONRXK43TNFXW4Q3PNVWWK3TUHMYTMMJXGY2DSMI>. You are receiving this because you authored the thread.Message ID: ***@***.***>