MATLAB Answers

Results from a loop in one table

10 views (last 30 days)
Paul Costache
Paul Costache on 14 Aug 2021
Edited: Adam Danz on 16 Aug 2021
I am using a loop to calculate Area, Perimeter, etc. of some particles from a set of pictures and I would like to save the results from all the iterations in one table. I only manage to get the results from the last one. When i tried indexing I get an error:
Subscripting into a table using one subscript (as in t(i)) is not supported. Specify a row subscript and a variable subscript, as in
t(rows,vars). To select variables, use t(:,i) or for one variable t.(i). To select rows, use t(i,:).
for i=1:5
% rest of code
results = regionprops('table',lm2, 'Area','Perimeter','MinFeretProperties','MaxFeretProperties');
results_total = table;
results_total = [results_total;results];
  1 Comment
Paul Costache
Paul Costache on 14 Aug 2021
results_total = table; I think I found the problem as this row of the code had to be outside of the loop.

Sign in to comment.

Accepted Answer

Adam Danz
Adam Danz on 14 Aug 2021
Edited: Adam Danz on 16 Aug 2021
As you noted, the second line of code in your for-loop should be moved outside and prior to the for-loop. However, there are two details that could make this solution inefficient.
First, the number of rows in each table is equal to the number of detected objects in the image so if you vertically concatenate all tables you will no longer be able to associate the rows of the table to the images you're looping through. If the images are very clean and you are certain of the number of expected objects in each image, you could compute the image index value for each row of the final table but I wouldn't trust that code.
Second, pre-allocation is almost always more efficient and eliminates indexing problems that occur when overwriting a variable.
This modification of your solution addresses those issues. It adds a column, imageIdx, that specifies the index value of the image. It's also a bit faster than without pre-allocation.
n = 5;
stats = cell(n,1);
for i = 1:n
stats{i} = regionprops('table',bw,'Centroid',...
imageIdx = repelem(1:n,cellfun(@(c)size(c,1),stats))';
statsTbl = [table(imageIdx), vertcat(stats{:})];
Example of output
>> head(statsTbl)
ans =
8×4 table
imageIdx Centroid MajorAxisLength MinorAxisLength
________ ________________ _______________ _______________
1 256.5 256.5 834.46 834.46
1 300 120 81.759 81.759
1 330.47 369.83 111.78 110.36
1 450 240 101.72 101.72
2 256.5 256.5 834.46 834.46
2 300 120 81.759 81.759
2 330.47 369.83 111.78 110.36
2 450 240 101.72 101.72
Paul Costache
Paul Costache on 15 Aug 2021
Worked perfectly. Thank you!

Sign in to comment.

More Answers (0)


Community Treasure Hunt

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

Start Hunting!

Translated by