for loop store values?

so I am convinced there is a better way for me to approach the following code:

ftse = xlsread('ftse123.xlsx');
L = length(ftse);
s = floor(L/10);
f25 = ftse(:,3);
k=1;
h=1;

for i = 1:10
    j=s*i;
    h=s*(i-1);
    if     h== 0
        h=1;
    else
        h=s*(i-1);
    end
    y{i} = f25(h:j,:);

end

x1=cell2mat(y(1));
x2=cell2mat(y(2));
x3=cell2mat(y(3));
x4=cell2mat(y(4));
x5=cell2mat(y(5));
x6=cell2mat(y(6));
x7=cell2mat(y(7));
x8=cell2mat(y(8));
x9=cell2mat(y(9));
x10=cell2mat(y(10));

I used copy and paste after struggling to save the values x1 etc, within the for loop.

Also noticed that x1 is has a length of 522 and not 523 like the rest?

The code achieves what I want/need but must be long and drawn out.

any advice would be great thanks.

[Stephen23]

"I used copy and paste after struggling to save the values x1 etc, within the for loop."

Copy-and-pasting code is a sign that you are doing something wrong. Using numbered variables is a sign that you are doing something wrong. The solution to both of these: use indexing.

Please upload ftse123.xlsx so that we have something to work with.

[Phil Whitfield]

that's when I posted after I started copy and pasting because I knew it was wrong.

Answer 1

[Stephen23]

I would recommend that you use readtable or importdata and use a table to store, group, and process your data. In lieu of that, here is some simple code to group the third column into ten groups:

>> N = size(ftse,1);
>> V = linspace(1,11,N+1);
>> V = fix(V(1:end-1));
>> C = accumarray(V(:),ftse(:,3),[],@(v){v});

It is much simpler and more efficient to access the cells of C using indexing. Do NOT create numbered variables.

"Also noticed that x1 is has a length of 522 and not 523 like the rest?"

Because ftse has 5225 rows. There is no way to split this into ten equally sized groups:

>> size(ftse)
ans =
        5225           4
>> size(ftse,1)/10
ans =
        522.5

Half the groups will have 252, half have 253 elements. My code spreads these evenly through C:

>> cellfun('size',C,1)
ans =
   523
   522
   523
   522
   523
   522
   523
   522
   523
   522

[Phil Whitfield]

Thanks, is there a way to create a 523x7 and 523x3 matrix of C from the index?

normally I would do :

xx=[x1,x2,x3,x4,x5,x6,x7];
xxx=[x8,x9,x10];

also I appreciate the 522 thing, should've seen that thanks.

[Stephen23]

"Thanks, is there a way to create a 523x7 and 523x3 matrix of C from the index?"

Not easily, because that does not match how many of each size data group that you actually have: five of each. So you could easily get 522x5 and 523x5 matrices:

M523 = [C{1:2:end}]
M522 = [C{2:2:end}]

Answer 2

[Ashraf Rayed]

i have this code, i have to check the leaf area of some pictures in a folder and then i have save the area of those leaves by sequence in an excel sheet. Can you please help to sort out the problem?

D = 'F:\MATLAB\R2018a\bin\rayed mat lab\experimental';

S = dir(fullfile(D,'*.jpg'));

for k = 1:numel(S)

F = fullfile(D,S(k).name);

I = imread(F);

ih=rgb2hsv(I);

ih1=ih(:,:,1);

binary=ih1>0.6;

leafArea = sum(binary(:))

subplot(2, 2, 2);

imshow(binary, []);

message = sprintf('The leaf area = %d pixels.\n pixels = %.1f%%',...

leafArea, defectArea/leafArea*100);

uiwait(msgbox(message))

end