Example_Deck.m

clear;
home;
close all;

%% The h5 file location

file1_folder='C:\Users\benja\OneDrive\Documents\GitHub\MTEX_Workshop\Data'; %location where the data is stored
file1_name='Mg_Pecs1 Specimen 1 Site 1 Map Data 1'; %should be a h5oina file, do not add in the .h5oina file extension

% file1_folder='/MATLAB Drive/MTEX_Workshop/Data/';
% file1_name='Mg_Pecs1 Specimen 1 Site 1 Map Data 1'; %should be a h5oina file, do not add in the .h5oina file extension

file_dset='1'; %data set number of interest in the h5 file

mb_length = 500; %micro bar length for plots - if you want to override, you can comment this out/clear this variable and the override will not happen

%% Load MTEX

%location of MTEX 5.10.2
%if MTEX loaded this doesn't matter
%if MTEX is not loaded, then this will be used to start MTEX up
mtex_location='C:\Users\benja\OneDrive\Documents\MATLAB\mtex-5.10.2';
% mtex_location='/MATLAB Drive/mtex-5.10.2';

%start mtex if needed
try EBSD;
catch
    run(fullfile(mtex_location,"startup.m"));
end

%create a results folder
file1_name_us=file1_name;
file1_name_us(strfind(file1_name_us,' '))='_';

resultsdir=fullfile(cd,'results',file1_name_us);
if isdir(resultsdir) == 0; mkdir(resultsdir); end

%% Read the h5oina data into Matlab

%Make this a full file name
file1_full=fullfile(file1_folder,[file1_name '.h5oina']);

%read the H5OINA EBSD Data - this will return two structures, with the data
%contained per layer/slice in the file, and also a contents file if you
%need to load anything else
[ebsd_data,header_data,h5oina_contents]=H5OINA_Read(file1_full);

%convert into a MTEX container
[ebsd] = H5OINA_Convert(ebsd_data,header_data,file_dset);

%% Set the plotting preferences - this has to be validated for your instrument

%this is set up for the pFIB and Oxford Insturments systems at UBC
setMTEXpref('xAxisDirection','east');           %aztec
setMTEXpref('zAxisDirection','outofPlane');     %aztec

%font size and overwrite the scale bar if you want
setMTEXpref('FontSize',12)

%% Now do some plots

%band contrast
figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');

if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read

%% Extract the phase to plot

%select the mineral name
mineral_name=ebsd.CS.mineral;
% mineral_name='Magnesium';

%% Plot the map of phases/indexed data
figure;
plot(ebsd');

%overlay with Band Contrast
figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');
hold on;
plot(ebsd,'FaceAlpha','0.3');
hold on;
if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read

%% IPF-x, y and z
% compute the colors
% ipfKey = ipfHSVKey(ebsd(mineral_name));
ipfKey = ipfTSLKey(ebsd(mineral_name)); %use the TSL

%plot the key
f1=figure;
plot(ipfKey)

% now generate the IPF colour maps
ipfKey.inversePoleFigureDirection = vector3d.X;
colors_X = ipfKey.orientation2color(ebsd(mineral_name).orientations);
ipfKey.inversePoleFigureDirection = vector3d.Y;
colors_Y = ipfKey.orientation2color(ebsd(mineral_name).orientations);
ipfKey.inversePoleFigureDirection = vector3d.Z;
colors_Z = ipfKey.orientation2color(ebsd(mineral_name).orientations);

%now plot the three IPF coloured maps
f2=figure;
plot(ebsd(mineral_name),colors_X,'micronbar','off');
nextAxis
plot(ebsd(mineral_name),colors_Y,'micronbar','off');
nextAxis
plot(ebsd(mineral_name),colors_Z,'micronbar','on');

if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read

%save the figures;
drawnow(); %updates the graphics engine to make sure it saves things properly
print(f1,fullfile(resultsdir,'Map_IPFkey.png'),'-dpng','-r600');
print(f2,fullfile(resultsdir,'Map_IPF.png'),'-dpng','-r600');
%% Plot the PF - scatter

%plot planes as spots, and use a subset
figure
plotPDF(ebsd(mineral_name).orientations,[Miller(1,1,0,ebsd(mineral_name).CS) Miller(0,1,0,ebsd(mineral_name).CS) Miller(0,0,1,ebsd(mineral_name).CS)])
%you can plot all - but this takes a while..
%plotPDF(ebsd(mineral_name).orientations,[Miller(1,1,0,ebsd(mineral_name).CS) Miller(0,1,0,ebsd(mineral_name).CS) Miller(0,0,1,ebsd(mineral_name).CS)],'Points','all')

%% Plot the IPF - scatter

figure
plotIPDF(ebsd(mineral_name).orientations,[xvector yvector zvector])

%% Construct the ODF

%suggested reading:
%https://mtex-toolbox.github.io/ODFTutorial.html
%https://mtex-toolbox.github.io/PoleFigure2ODF.html

% compute the ODF
odf = calcDensity(ebsd(mineral_name).orientations);
% You could also change/fix the half width - compare this different ODF
% odf = calcDensity(ebsd(mineral_name).orientations,'halfwidth',2*degree);

figure;
% plot the pole figure representation of the ODF
plotPDF(odf,[Miller(0,0,1,ebsd(mineral_name).CS) Miller(0,1,0,ebsd(mineral_name).CS)]);

mtexColorbar('location','southoutside')
mtexColorMap blue2red

%add contours
levels=[-7:7];
hold on;
plotPDF(odf,[Miller(0,0,1,ebsd(mineral_name).CS) Miller(0,1,0,ebsd(mineral_name).CS)],'contour',levels,'linecolor','k');
mtexColorbar('location','southoutside')


%you can also plot this ODF in the IPF
figure;
% plot the inverse pole figure representation of the ODF
plotIPDF(odf,[xvector yvector zvector])
mtexColorbar('location','southoutside')
mtexColorMap blue2red


%% Plot the uncorrelated misorientation distribution
%see https://mtex-toolbox.github.io/MisorientationDistributionFunction.html

mdf = calcMDF(odf);

%plot the axis
figure;
plotAxisDistribution(mdf)

%plot the angle
figure;
plotAngleDistribution(mdf)
hold all
plotAngleDistribution(ebsd(mineral_name).CS,ebsd(mineral_name).CS)
hold off
legend('data MDF','uniform ODF')

%% Calculate grains and grain boundaries
%see https://mtex-toolbox.github.io/GrainReconstruction.html

[grains,ebsd.grainId] = calcGrains(ebsd,'angle',10*degree); 
% [grains,ebsd.grainId] = calcGrains(ebsd,'angle',10*degree,'boundary','tight');

%care with this threshold value - you can look at the angle distribution for some guidance

%Now plot the grian boundary map on the band contrast map
figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');
if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read
hold on
plot(grains.boundary,'linewidth',0.5,'lineColor','r')
hold off

%contrast this with the grain boundaries for indexed only points 
% (look at the scratches)
grains_indexed = calcGrains(ebsd('indexed'));
% If the edge grains are being annoying - don't use the convex hull
% grains_indexed = calcGrains(ebsd('indexed'),'boundary','tight');

figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');
if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read
hold on
plot(grains_indexed.boundary,'linewidth',0.5,'lineColor','r')
hold off

%subset to only have the interior grains
grains_indexed_int=grains_indexed(~grains_indexed.isBoundary);
% hold on
% plot(grains_indexed_int.boundary,'linewidth',0.5,'lineColor','g')
% hold off

%% Plot the grain size distributions

%extract some different distributions
grain_radius=grains_indexed_int.equivalentRadius;
grain_area=grains_indexed_int.area;
grain_perimeter=grains_indexed_int.equivalentPerimeter;
% for a full list of shapes - https://mtex-toolbox.github.io/ShapeParameters.html

histogram_bins=100;

figure;
subplot(3,1,1);
histogram(grain_radius,histogram_bins,'Normalization','probability');
%meddle with the displays
ytix = get(gca, 'YTick');
set(gca, 'YTick',ytix, 'YTickLabel',ytix*100);
ylabel('Frequency %'); xlabel('Radius (\mum)')
xlims=xlim; xlim([0 xlims(2)]);

subplot(3,1,2);
histogram(grain_area,histogram_bins,'Normalization','probability');
%meddle with the displays
ytix = get(gca, 'YTick');
set(gca, 'YTick',ytix, 'YTickLabel',ytix*100);
ylabel('Frequency %'); xlabel('Area (\mum^2)')
xlims=xlim; xlim([0 xlims(2)]);

subplot(3,1,3);
histogram(grain_perimeter,histogram_bins,'Normalization','probability');
%meddle with the displays
ytix = get(gca, 'YTick');
set(gca, 'YTick',ytix, 'YTickLabel',ytix*100);
ylabel('Frequency %'); xlabel('Perimeter (\mum)')
xlims=xlim; xlim([0 xlims(2)]);

%are any of these Normally distributed ? If so, have a think about what you
%call an average 'grain size'

%% Filter the grain size data
%mostly to demo that this can be done

min_grain_radius=10;
grains_reduced=grains_indexed_int(grains_indexed_int.equivalentRadius>min_grain_radius);
grains_excluded=grains_indexed_int(grains_indexed_int.equivalentRadius<=min_grain_radius);
figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');
hold on;
plot(ebsd,'FaceAlpha','0.3');
hold on;
if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read

hold on
plot(grains_reduced.boundary,'linewidth',0.5,'lineColor','g')
plot(grains_excluded.boundary,'linewidth',0.5,'lineColor','r')

hold off

%% Demonstrate that we can extract a point from this map

%plot the phase data + grain boundaries + bc
figure;
plot(ebsd');

%overlay with Band Contrast
figure;
plot(ebsd,ebsd.prop.Band_Contrast); colormap('gray');
hold on;
plot(ebsd,'FaceAlpha','0.3');
hold on;
if exist('mb_length','var'); mp = MTEX_mb_fix(mb_length); end %change the micronbar length to make it prettier/easier to read

hold on
plot(grains_indexed_int.boundary,'linewidth',0.5,'lineColor','g')
hold off

%take a mouse input
disp('Click a grain to select it')
[x_g,y_g]=ginput(1);

%find the nearest point in the data to this point
p_map=(ebsd.prop.x-x_g).^2+(ebsd.prop.y-y_g).^2;
[mv,pattern_number]=min(p_map);
ebsd_point=ebsd(pattern_number);

hold on;
% % scatter(x_g,y_g,'r');
%mark this point
scatter(ebsd_point.x,ebsd_point.y,'w');

%plot the grain for this
hold on
plot(grains(ebsd_point.x,ebsd_point.y).boundary,'linewidth',4,'linecolor','y')
hold off

%plot the crystal shape for this point
hold on
scaling = 100; % scale the crystal shape to have a nice size
%here we use a hexagon, but you could use a cube for cubic
%crystalShape.cube
cS = crystalShape.hex(ebsd(mineral_name).CS);
plot(ebsd(pattern_number).prop.x,ebsd(pattern_number).prop.y,50, ebsd(pattern_number).orientations * cS * scaling)
hold off

%% Extract a subset of the map
figure;
plot(ebsd,ebsd.prop.Band_Contrast)

%here we will use a mouse click pair to cut out the box - pick two opposite
%diagonals
disp('Click two diagonal points to select a box to subset the data')
[x_i,y_i]=ginput(2);

% x_i=[-918;-205];
% y_i=[572;1010];

hold on;
scatter(x_i,y_i);

x_i=sort(x_i); y_i=sort(y_i);
box_bounds=[x_i(1),x_i(2),x_i(2),x_i(1),x_i(1);y_i(1),y_i(1),y_i(2),y_i(2),y_i(1)];
plot(box_bounds(1,:),box_bounds(2,:));
points_box=inpolygon(ebsd,box_bounds');
ebsd_small=ebsd(points_box);

nextAxis;
plot(ebsd_small,ebsd_small.prop.Band_Contrast);

%% For this smaller region, plot all the IPFs
ipfKey2 = ipfHSVKey(ebsd_small(mineral_name));

ipfKey2.inversePoleFigureDirection = vector3d.X;
colors_X2 = ipfKey2.orientation2color(ebsd_small(mineral_name).orientations);
ipfKey2.inversePoleFigureDirection = vector3d.Y;
colors_Y2 = ipfKey2.orientation2color(ebsd_small(mineral_name).orientations);
ipfKey2.inversePoleFigureDirection = vector3d.Z;
colors_Z2 = ipfKey2.orientation2color(ebsd_small(mineral_name).orientations);

%now plot the three IPF coloured maps
figure;
plot(ebsd_small(mineral_name),colors_X2,'micronbar','off');
nextAxis
plot(ebsd_small(mineral_name),colors_Y2,'micronbar','off');
nextAxis
plot(ebsd_small(mineral_name),colors_Z2,'micronbar','on');

%% extract the grain data from the big map that fits within this smaller map

grains_small_ID=inpolygon(grains,box_bounds');
grains_small=grains(grains_small_ID);
grains_small=grains_small(grains_small.grainSize>10); %at least 10 pts per grain

figure;
plot(ebsd_small(mineral_name),colors_Z2,'micronbar','on');

hold on
plot(grains_small.boundary,'linewidth',0.5,'lineColor','k')
hold off

%smooth these boundaries - note smoothing changes the data and its statistics
% see -  https://mtex-toolbox.github.io/GrainSmoothing.html
% grains_smooth = smooth(grains_small); %single pass
grains_smooth = smooth(grains_small,5); %5 iterations
% see https://mtex-toolbox.github.io/GrainSmoothing.html

hold on
plot(grains_smooth.boundary,'linewidth',2,'lineColor','b')
hold off

%fix the bounding edges
xlim(x_i); ylim(y_i);

%% Plot a subset map with the unit cells over the top
figure;
plot(ebsd_small,ebsd_small.prop.Band_Contrast,'micronbar','off'); %the scale bar is turned off because we are going to make it hard to see
colormap('gray');
hold on
plot(grains_smooth.boundary,'linewidth',2,'lineColor','r')

% plot the unit cells
%plot the crystal shape for this point
scaling = 25; % scale the crystal shape to have a nice size
%here we use a hexagon, but you could use a cube for cubic
%crystalShape.cube
cS = crystalShape.hex(grains_small(mineral_name).CS);
grain_centroids=grains_small.centroid;
plot(grain_centroids(:,1),grain_centroids(:,2),50, grains_small.meanOrientation * cS * scaling)

%fix the bounding edges
xlim(x_i); ylim(y_i);

%% Select an individual grain for texture based segmentation
% use grains(x,y) to select a grain from a map based on x and y
% coordinates
% grain_selected = grains(200,500);
%3 check you have the right grain by plotting, then calculate the mean
%orientation

%create a container for EBSD phase 1 (alpha - HCP)
ebsd_2=ebsd(mineral_name);

% figure('Position',PlotData.ssize);
figure;
subplot(1,3,1);
ipfKey.inversePoleFigureDirection = vector3d.Z;
plot(grains(mineral_name),ipfKey.orientation2color(grains(mineral_name).meanOrientation));
title('Crystal Orientation - Z')


% take a mouse input
disp('Select an individual grain as your seed grain orientation');
 [x,y]=ginput(1);

% % hard code the mouse input to run this - swap the commenting around
% % if you want a ginput
% % hard coded position, in um
% x=1379; y=1331;

%add the point to the figure as a black spot
hold on;
scatter(x,y,100,'k','filled');

% find the corresponding grain
grain_sel = grains(x,y);

plot(grain_sel.boundary,'linecolor','g','LineWidth',3,'parent',gca)
plot(grain_sel.boundary,'linecolor','r','LineWidth',1,'parent',gca)
hold off

orientation_threshold=15; %in degrees
% Segment EBSD Cubes based on grain orientation 
%1 Define a fibre
% fibre_test = fibre(Miller(1,1,0,ebsd(phase1).CS),yvector);
%2 Extract Orientations using angle function to within 20 degrees of the fibre axis (generates a logical array) 
test_orientations_2 = angle(ebsd_2.orientations,grain_sel.meanOrientation,'antipodal')<orientation_threshold*degree; % 20 degrees is arbitary
% test_orientations_2 = angle(ebsd_2.orientations,fibre_test,'antipodal')<20*degree; 

%3 Segment original cube based on Phase (can only do this with one phase)
test_orientations_in = ebsd_2(test_orientations_2 == 1);
test_orientations_out = ebsd_2(test_orientations_2 == 0);

%5 Plot to check
nextAxis;
plot(test_orientations_in, ipfKey.orientation2color(test_orientations_in.orientations));
title(['Data close to selected grain (<' int2str(orientation_threshold) '^o)'])
nextAxis;
plot(test_orientations_out, ipfKey.orientation2color(test_orientations_out.orientations));
title(['Data far from selected grain (>' int2str(orientation_threshold) '^o)'])

ipfKey.inversePoleFigureDirection = vector3d.X;
nextAxis;
plot(grains(mineral_name),ipfKey.orientation2color(grains(mineral_name).meanOrientation));
title('Crystal Orientation - X')
nextAxis;
plot(test_orientations_in, ipfKey.orientation2color(test_orientations_in.orientations));
title(['Data close to selected grain (<' int2str(orientation_threshold) '^o)'])
nextAxis;
plot(test_orientations_out, ipfKey.orientation2color(test_orientations_out.orientations));
title(['Data far from selected grain (>' int2str(orientation_threshold) '^o)'])

ipfKey.inversePoleFigureDirection = vector3d.Y;
nextAxis;
plot(grains(mineral_name),ipfKey.orientation2color(grains(mineral_name).meanOrientation));
title('Crystal Orientation - Y')
nextAxis;
plot(test_orientations_in, ipfKey.orientation2color(test_orientations_in.orientations));
title(['Data close to selected grain (<' int2str(orientation_threshold) '^o)'])
nextAxis;
plot(test_orientations_out, ipfKey.orientation2color(test_orientations_out.orientations));
title(['Data far from selected grain (>' int2str(orientation_threshold) '^o)'])

%% plot some unit cells for this

grain_IDs_close=unique(test_orientations_in.grainId);
grains_close=grains(grain_IDs_close);
grains_close=grains_close(grains_close.grainSize>3);
grains_close=grains_close(~grains_close.isBoundary);

figure;
plot(test_orientations_in, test_orientations_in.prop.Band_Contrast);
% title(['Data far from selected grain (>' int2str(orientation_threshold) '^o)'])

hold on
plot(grains_close.boundary,'linewidth',1,'lineColor','r')

% plot the unit cells
%plot the crystal shape for this point
scaling = 125; % scale the crystal shape to have a nice size
%here we use a hexagon, but you could use a cube for cubic
%crystalShape.cube
cS = crystalShape.hex(grains_close(mineral_name).CS);
grain_centroids=grains_close.centroid;
plot(grain_centroids(:,1),grain_centroids(:,2),50, grains_close.meanOrientation * cS * scaling,'FaceAlpha',0.2); %make the hexagons also transparent so you can see the grains behind

%% copy this m file over, for archival purposes

mf_long=mfilename('fullpath');
[f1,f2,f3]=fileparts(mf_long);
mf_start=[mf_long '.m'];
mf_end=fullfile(resultsdir,[f2 '.m']);
try
copyfile(mf_start,mf_end)
catch
    warning('m file not saved, likely due to spaces in the file name');
end