Calculating the eigenvalues of simple shapes
7 Ansichten (letzte 30 Tage)
Ältere Kommentare anzeigen
John Bach
am 13 Mär. 2023
Bearbeitet: John Bach
am 25 Mär. 2023
Hi there,
I've used the strel function to create a range of shapes and I would like to now calculate the eigenvalues of each shape although I am struggling to do this and would really appreciate any help regarding this.
Thank you in advance,
1 Kommentar
the cyclist
am 13 Mär. 2023
Bearbeitet: the cyclist
am 13 Mär. 2023
Do you have a reference for what the eigenvalue of a binary shape is? I did some googling of keywords, but didn't find something definite. (Maybe this is well known in image processing, but that is not my specialty.)
Are you stuck on the math of it, or the MATLAB coding? Have you written any code?
Akzeptierte Antwort
Walter Roberson
am 13 Mär. 2023
M = double(strel('disk',5).Neighborhood)
E = simplify(eig(sym(M)))
vpa(E)
(the imaginary component is due to round-off error)
M2 = double(strel('octagon',12).Neighborhood)
E2 = simplify(eig(sym(M2)))
vpa(imag(E2))
vpa(E2)
1 Kommentar
Bjorn Gustavsson
am 15 Mär. 2023
The/One benefit of using svd instead of eig is that one get real singular values - which is not a guarantee with eig. Appart from that the soutions should be comparable/similar/identical.
Weitere Antworten (1)
Bjorn Gustavsson
am 13 Mär. 2023
If you have a binary image then why not just run through the svd and see what you get:
I = zeros(256);
I(64:(64+128),64:(64+128)) = 1;
[U,S,V] = svd(I);
figure
subplot(1,2,1)
plot(diag(S))
subplot(1,2,2)
imagesc(U(:,1)*S(1,1)*V(:,1)')
% Or for a funnier example:
I = numgrid('B',258);
I = I(2:end-1,2:end-1);
[U,S,V] = svd(I);
subplot(2,2,1)
plot(diag(S))
subplot(2,2,2)
imagesc(U(:,1)*S(1,1)*V(:,1)')
subplot(2,2,2)
imagesc(U(:,1:4)*S(1:4,1:4)*V(:,1:4)')
subplot(2,2,2)
imagesc(U(:,1:16)*S(1:16,1:16)*V(:,1:16)')
You can also look at the individual eigen-images by something like:
imagesc(U(:,7)*V(:,7)')
HTH
0 Kommentare
Siehe auch
Kategorien
Mehr zu Linear Algebra finden Sie in Help Center und File Exchange
Produkte
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!