Plot reconstructed parent beta via fill and broken / bad edges

[accessT]

Hello together!

When i use MTEX 5.3, i get a nice plot:

With MTEX 5.3.1 and newer, i get bad edges:

Well, i found out, it has to do with the fill.m at line 17 untill 19:

I just changed only this lines back like in 5.3 and it works!

Is there another way to get it work like in 5.3?

My Example:
Probe 1 TEST Specimen 1 Site 1 Map Data 1-Subset1 - reconstruction_alpha2beta.txt

fullpath = 'Probe 1 TEST Specimen 1 Site 1 Map Data 1-Subset1 - reconstruction_alpha2beta.txt';
mineral_ = 'Ti (beta)';
angle_ = 5;
min_grain_Size_ = 10; % Pixels
ColumnNames.Names = { 'phi1' 'Phi' 'phi2' 'x' 'y'};

CS = {'notIndexed', crystalSymmetry('m-3m', 'mineral', mineral_ ), 'notIndexed'};
ebsd = EBSD.load(fullpath,CS,'interface','generic', 'ColumnNames', ColumnNames.Names, 'Bunge');
[grains_unfiltered,ebsd.grainId,ebsd.mis2mean] = calcGrains(ebsd(mineral_),'angle',angle_*degree);
grains_right=grains_unfiltered;
grains_right = grains_right(grains_right.grainSize >= min_grain_Size_); %filter
ebsd = ebsd(grains_right);
[grains_right,ebsd.grainId,ebsd.mis2mean] = calcGrains(ebsd(mineral_),'angle',angle_*degree); % ReCalc

ebsd_splined = smooth(ebsd, splineFilter, 'fill'); % HERE STARTS THE PROBLEM

[grains_splined,ebsd_splined.grainId,ebsd_splined.mis2mean] = calcGrains(ebsd_splined(mineral_),'angle',angle_*degree);
grains_right2 = grains_splined(grains_splined.grainSize >= min_grain_Size_);
ebsd_splined = ebsd_splined(grains_right2);
[grains_smoothed,ebsd_splined.grainId,ebsd_splined.mis2mean] = calcGrains(ebsd_splined(mineral_),'angle',angle_*degree);
grains_smoothed = smooth(grains_smoothed, 5);
ebsd_splined = ebsd_splined(grains_smoothed);

mineral_ = string(mineral_);
fig_grains_smoothed=figure;
plot(grains_smoothed(mineral_),grains_smoothed(mineral_).meanOrientation,'micronbar','on','figSize','large','Name', 'BETA: Grains_Smoothed'); %only grain(s)
fig_ebsd_smoothed=figure;
plot(ebsd_splined(mineral_), ebsd_splined(mineral_).orientations, 'figSize', 'large','Name', 'EBSD_Smoothed');

[ralfHielscher]

Hi,

for the moment you can replace the 'none' you marked by 'nearest' and you should get the result you are aiming for.
In the future, I the functions smooth / fill will accept the option 'extrapolate' which will do this.

The point is, that not everybody wants his data to be extrapolated to some rectangular boundaries.

Ralf.

[accessT]

great, thank you :)