Contents
Labeling - Lab 4
% Read the an image and display. Note the image is type logical ie binary BW = imread('text.png'); figure(1) imshow(BW);
Label the image.
Each indiviual object is given a unique value
figure(2) lab=bwlabel(BW); imshow(lab) colorbar; % We can also show this as a false color map figure(3) imshow(label2rgb(lab)); title('False Colour Map');
Another way
The labelled image is not that useful in practice so lets delete it
clear lab % Instead use this command CC = bwconncomp(BW); % Each object is stored seperatly is a structure CC CC
CC =
Connectivity: 8
ImageSize: [256 256]
NumObjects: 88
PixelIdxList: {1x88 cell}
Construct the labelled image
% load the target image BW=imread('target.png'); % threshold to convert to a binary image, and perform labelling BW=im2bw(BW,0.1); CC = bwconncomp(BW); % create an array filled with zeros of the same size BW2=zeros(CC.ImageSize); %cycle through each object, adding it into array BW2 for p=1:CC.NumObjects %loop through each image BW2(CC.PixelIdxList{p}) = p; %set the image figure(4) imshow(BW2,[0 CC.NumObjects]) %display with fixed intestiy range pause(.5) %pause for 0.5 seconds end
Find the cirles
s = regionprops(BW, 'Eccentricity'); BW3=zeros(CC.ImageSize); for p=1:CC.NumObjects %loop through each image if s(p).Eccentricity < 0.3 BW3(CC.PixelIdxList{p}) = 1; %set the image end end figure(5) imshow(BW3,[0 1]) %display with fixed intestiy range
Now for some morphological operations
% Create a structure array se=strel('disk',11) se figure(6) imshow(imerode(BW,se)) title('Erode') figure(7) imshow(imdilate(BW,se)) title('Dilate') % Lets try a different structure array se=strel('square',11) se figure(8) imshow(imerode(BW,se)) title('Erode 2') figure(9) imshow(imdilate(BW,se)) title('Dilate 2')
se =
Flat STREL object containing 357 neighbors.
Decomposition: 4 STREL objects containing a total of 32 neighbors
Neighborhood:
Columns 1 through 13
0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 1 1 1 1 1 1 1 1
0 0 0 0 1 1 1 1 1 1 1 1 1
0 0 0 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 1 1 1 1
0 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 1 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 1 1 1 1
0 0 0 1 1 1 1 1 1 1 1 1 1
0 0 0 0 1 1 1 1 1 1 1 1 1
0 0 0 0 0 1 1 1 1 1 1 1 1
0 0 0 0 0 0 1 1 1 1 1 1 1
Columns 14 through 21
1 1 0 0 0 0 0 0
1 1 1 0 0 0 0 0
1 1 1 1 0 0 0 0
1 1 1 1 1 0 0 0
1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 0
1 1 1 1 1 1 0 0
1 1 1 1 1 0 0 0
1 1 1 1 0 0 0 0
1 1 1 0 0 0 0 0
1 1 0 0 0 0 0 0
se =
Flat STREL object containing 357 neighbors.
Decomposition: 4 STREL objects containing a total of 32 neighbors
Neighborhood:
Columns 1 through 13
0 0 0 0 0 0 1 1 1 1 1 1 1
0 0 0 0 0 1 1 1 1 1 1 1 1
0 0 0 0 1 1 1 1 1 1 1 1 1
0 0 0 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 1 1 1 1
0 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1 1 1
0 1 1 1 1 1 1 1 1 1 1 1 1
0 0 1 1 1 1 1 1 1 1 1 1 1
0 0 0 1 1 1 1 1 1 1 1 1 1
0 0 0 0 1 1 1 1 1 1 1 1 1
0 0 0 0 0 1 1 1 1 1 1 1 1
0 0 0 0 0 0 1 1 1 1 1 1 1
Columns 14 through 21
1 1 0 0 0 0 0 0
1 1 1 0 0 0 0 0
1 1 1 1 0 0 0 0
1 1 1 1 1 0 0 0
1 1 1 1 1 1 0 0
1 1 1 1 1 1 1 0
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 0
1 1 1 1 1 1 0 0
1 1 1 1 1 0 0 0
1 1 1 1 0 0 0 0
1 1 1 0 0 0 0 0
1 1 0 0 0 0 0 0
se =
Flat STREL object containing 121 neighbors.
Decomposition: 2 STREL objects containing a total of 22 neighbors
Neighborhood:
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
se =
Flat STREL object containing 121 neighbors.
Decomposition: 2 STREL objects containing a total of 22 neighbors
Neighborhood:
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
1 1 1 1 1 1 1 1 1 1 1
Combining the two methods
tmp=imdilate(BW,se); figure(10) imshow(imerode(tmp,se)) title('Dilate then Erode') % this is close to the oringal. By combining these we can do things like % fill holes in our mask, remove small objects and noise. Erode is useful % for seperating objects that are close together.
