MATLAB Answers

Extract number and infromation from multiple image files ?

5 views (last 30 days)
D. Frank
D. Frank on 25 Jun 2020
Commented: D. Frank on 7 Jul 2020
Hi everybody
I have a folder with many images in a structure of name: "name_id_324_988_true_207_1.6", i want to take the part "324_988_true" to calculate, for example:
ls Defolder;
23l9p9_Sy3lMlT_265_332_false_270_1.32.jpg
38bvq6_dwJCL2R_126_665_true_6_0.5.jpg
i want to take the part "265_332_false" for each image and show the calculated number on the distribution table for example:
folder = 'Defolder';
S = dir(fullfile(folder,'*.jpg'));
N = {S.name};
[~,F] = cellfun(@fileparts,N,'uni',0);
V = str2double(F);
V1 = V';
all i got is a lot of NaN, please help, thank youu

  1 Comment

Sign in to comment.

Accepted Answer

Rik
Rik on 27 Jun 2020
The code below should do what you need. Adapt as needed.
folder = 'Defolder';
S = dir(fullfile(folder,'*.jpg'));
N = {S.name};
%overwrite N so it works on my computer as well
N={'23l9p9_Sy3lMlT_265_332_false_270_1.32.jpg','38bvq6_dwJCL2R_126_665_true_6_0.5.jpg'};
c=cellfun(@MyFun,N,'UniformOutput',false);
celldisp(c)
function c=MyFun(str)
c=strsplit(str,'_');
c=c(3:5);
c{1}=str2double(c{1});
c{2}=str2double(c{2});
c{3}=strcmp(c{3},'true');
end

  12 Comments

Show 9 older comments
Rik
Rik on 7 Jul 2020
Don't do any further processing in that function. The plotting should not be inside that function. You should only use it to extract values from the file names, nothing more. Anything you want to do with those values should happen outside of that function.
Rik
Rik on 7 Jul 2020
Leave this function intact:
function c=MyFun(str)
c=strsplit(str,'_');
c=c(3:5);
c{1}=str2double(c{1});
c{2}=str2double(c{2});
c{3}=strcmp(c{3},'true');
c{4}=(c{1}/45)*c{2};
end
Then you can modify the code after cellfun:
folder = 'Defolder';
S = dir(fullfile(folder,'*.jpg'));
N = {S.name};
%overwrite N so it works on my computer as well
N={'23l9p9_Sy3lMlT_265_332_false_270_1.32.jpg','38bvq6_dwJCL2R_126_665_true_6_0.5.jpg'};
c=cellfun(@MyFun,N,'UniformOutput',false);
c=vertcat(c{:});
for n=1:size(c,2)
c{1,n}=cell2mat(c(:,n));
end
c(2:end,:)=[];
%you can also put more descriptive variable names here instead
plot(c{1},c{2},'*')
function c=MyFun(str)
c=strsplit(str,'_');
c=c(3:5);
c{1}=str2double(c{1});
c{2}=str2double(c{2});
c{3}=strcmp(c{3},'true');
c{4}=(c{1}/45)*c{2};
end

Sign in to comment.

More Answers (1)

Stephen Cobeldick
Stephen Cobeldick on 1 Jul 2020
One simple regular expression does this quite efficiently:
D = 'Defolder';
S = dir(fullfile(D,'*.jpg'));
N = {S.name};
T = regexpi(N,'(\d+)_(\d+)_(true|false)','tokens','once');
T = vertcat(T{:});
M = str2double(T(:,1:2))
B = strcmpi(T(:,3),'true')

Tags


Translated by