how to calculate width from center line in binary image ?

Hello !

my goal is to find the location of my object which has the smallest width.

I did the skeletonization and then I tried to apply bwdist on my image. Then I multiplied the 2 images. But how do I know which is the location with the smallest width ? (because I need to mark and identify this location).

Thank you

close all;
clear all;
clc; 

folder = "C:\Users\larle\OneDrive\Documents\Image_chromosomes\testing_matlab2\resultat"; 
baseFileName = "entite4.png";
fullFileName = fullfile(folder, baseFileName); 
im = imread(fullFileName);
im = imsharpen(im);%accentue nettete avec masquage flou
im = medfilt2(im);%filtre median en 2D (reduction de bruits)

thresholdValue=220;

%binarisation
binaryImage = im < thresholdValue; 

%skel without extremum
skel= bwmorph(binaryImage,'skel',Inf);
B = bwmorph(skel, 'branchpoints');
E = bwmorph(skel, 'endpoints');
[y,x] = find(E);
B_loc = find(B);
Dmask = false(size(skel));
for k = 1:numel(x)
   D = bwdistgeodesic(skel,x(k),y(k));
   distanceToBranchPt = min(D(B_loc));
   Dmask(D < distanceToBranchPt) =true;
end
skelD = skel - Dmask;
figure(2)
imshow(skelD)
hold all;

edtImage = bwdist(binaryImage);
centerlineImage = double(skelD) .* edtImage;
figure(5)
imshow(centerlineImage)
meanRadius = mean(centerlineImage(skelD))
figure(6)
imshow(meanRadius)

[Adam Danz]

I only see 1 object in your image. Is there supposed to be multiple objects? If not, I don't know what to goal is, "to find the object with the smallest width".

Answer 1

[Image Analyst]

You can use min() to find the min value of the centerlineImage that's not zero. Then use find to find the row and column

minValue = min(centerlineImage(centerlineImage > 0))
[rowOfMin, columnOfMin] = find(centerlineImage == minValue)

Answer 2

[Hammad Moussa]

how to filtre median and feltre gaussien help me