Detect the vertical dark layers in greysacale image

Hi!

I have a grayscale image with dark vertical layers. I want to know the average pixel-width of the dark layers in the following image:

What I found problematic is that the image also contains noise which can be seen when I use the canny edge detection:

Any sugestions which method can be used in this case?

Thanks in advance

Answer 1

[Akira Agata]

By applying findpeaks function (with appropriate option settings) to the average intensity profile, you can detect locations of dark bands.

I'm not sure what is the definition of the "average pixel-width" of dark band, but I believe one possible (and simple) solution would be detectiong width at the half-prominence point for each nitch in the the average intensity profile.

% Read the image
I = imread('arsringsbild2.png');
I = rgb2gray(I);

% Calculate average intensity and dark positions
avgIntensity = mean(double(I));
[val,pos] = findpeaks(-1*avgIntensity,...
  'MinPeakProminence',  40,...
  'MinPeakHeight',      -100);
val = -1*val;

% Visualize the result
figure
subplot('Position',[0.1 0.8 0.8 0.15])
imshow(I)
subplot('Position',[0.1 0.1 0.8 0.7])
plot(avgIntensity)
hold on
scatter(pos,val,'^','filled')
legend({'Avg. Intensity','Detected dark position'},'FontSize',12)
xlim([1 size(I,2)])
ylabel('Average intensity for each horizontal position')

[Aon]

Thank you for a great answer!

Do you know a way to also measure the pixel-width of the light bands?