Filter löschen
Filter löschen

Edge Detection in a 2D Matrix

19 Ansichten (letzte 30 Tage)
Wayne
Wayne am 31 Jul. 2023
Kommentiert: Wayne am 8 Aug. 2023
Hi there,
I have a 2D matrix which contains contains 0s and 1s as shown in the left image below. I would like to get the outer boundary of this shape. By using the edge function, it gives a nice estimate of the boundary, but some of the points lie on the original shape. What I did was to use the "edge" function directly with only my 2D matrix as input. Are there any ways to get the outer boundary only and without any points intersecting the original shape? Thank you.
  2 Kommentare
Dyuman Joshi
Dyuman Joshi am 31 Jul. 2023
What is the output you want to obtain?
Also, can you attach your data? Use the paperclip button to do so.
Wayne
Wayne am 31 Jul. 2023
Hi @Dyuman Joshi thanks for the reply - i have attached the data. I would like to have a boundary of a single pixel which surrounds the original figure. The "edge" function does that to some extent, but there are still some overlapped regions.

Melden Sie sich an, um zu kommentieren.

Akzeptierte Antwort

DGM
DGM am 31 Jul. 2023
Consider the example:
% a logical image
mask = imread('tinyblobs.png');
imshow(mask)
% generate the perimeter image
paddedmask = padarray(mask,[1 1],'replicate','both');
perim = bwperim(~paddedmask);
perim = perim(2:end-1,2:end-1);
imshow(perim)
% show that the new pixels do not overlap the original blobs
% and that there is no gap between the two
imshow(imfuse(mask,perim))

Weitere Antworten (2)

Matt J
Matt J am 1 Aug. 2023
Bearbeitet: Matt J am 1 Aug. 2023
% a logical image
mask = imread('tinyblobs.png');
outerboundary = imdilate(mask,strel('disk',1))&~mask;
imshow(mask+2*outerboundary,[])
  1 Kommentar
Wayne
Wayne am 8 Aug. 2023
Thank you, appreciate the response!

Melden Sie sich an, um zu kommentieren.


Bruno Luong
Bruno Luong am 31 Jul. 2023
Bearbeitet: Bruno Luong am 31 Jul. 2023
Do you want this
load(websave('data.mat','https://www.mathworks.com/matlabcentral/answers/uploaded_files/1446792/data.mat'))
d=conv2(data,[0 1 0; 1 1 1; 0 1 0],'same')>0;
b=d-data;
figure
imagesc(data);
axis equal
figure
imagesc(b)
axis equal
  1 Kommentar
Wayne
Wayne am 8 Aug. 2023
Yes this was what I wanted, thanks for your help!

Melden Sie sich an, um zu kommentieren.

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!

Translated by