How to visualise results of regionprops3 function?

6 views (last 30 days)
Dear all,
I am currently working with some CT data, which I have created by image stacking.
I would like to analyse properties using regionprops3 function. I would like to plot/visualise each found volume by this function so I can check, if the results are correct.
Here is my code:
% V_initial -- size 1000x1100x71
Volsize = size(V_initial);
for i=1:Volsize(3) % could be speed up by using parfor, but the volume is small
XY_vol{i} = V_initial(:,:,i); % slicing volume into each plane
XY_vol_adjst{i} = imadjust(XY_vol{i}); % adjusting contrast
threshold_multi(i,:) = multithresh(XY_vol_adjst{i},3); % Otsu multhreshold method to identify phases
XY_vol_bin_pores{i} = imbinarize(XY_vol_adjst{i},threshold_multi(i,1)); %pores
XY_vol_bin_solid{i} = imbinarize(XY_vol_adjst{i},threshold_multi(i,2)); %paste
XY_vol_bin_agg{i} = imbinarize(XY_vol_adjst{i},threshold_multi(i,3)); %aggregates
volstack_pores(:,:,i) = XY_vol_bin_pores{i}; %stacking volume of pores
volstack_solid(:,:,i) = XY_vol_bin_solid{i}; %stacking volume of paste
volstack_agg(:,:,i) = XY_vol_bin_agg{i}; %stacking volume of aggregates
vol_solid_flipped_BW{i} = imcomplement(XY_vol_bin_solid{i}); %inversion of binarized paste
volstack_solid_1(:,:,i) = vol_solid_flipped_BW{i}; %stacking volume of inverted paste
%region analysis
stats = regionprops3(volstack_solid,'all'); %takes some time
voxelID = cat(1,stats.VoxelIdxList);
%selecting the first and the biggest volume identified by regionprops3
volume = volstack_solid(voxelID{1}); %using volume indices should allow to new volume out
% of volstack_solid this time only using identified volume
volshow(volume) %visualise volume
%Error using volshowvolumeViewer requires 3-D data with no singleton dimensions.
I am stuck using identified indices. Maybe I am doing something wrong using identified indices.
Could you help me to fix this problem, please ?
Thank you in advance,
  1 Comment
Alvaro on 24 Jan 2023
From the documentation of volshow, this function takes a 3D array.
What are the dimensions of your variable volume?

Sign in to comment.

Answers (1)

Constantino Carlos Reyes-Aldasoro
What regionprops will do is to calculate the metrics for each of the elements in your data, and you can then use these to select a subset of the elements to visualise, for example:
a=randn(256); % random values
b=bwlabel(a>1.5); % threshold to select a few of these, label to have unique ids
c=regionprops(b); % find properties, for this example we will use the area, you can open with openvar('c')
d=ismember(b,find([c.Area]==2)); % select only those elements with area 2
imagesc((b>0)+d) % display original elements + those selected
elements with area ==2 appear in yellow, all others appear in light blue.
Hope this helps answer your question.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!