Materials

Download the images you need: image1; image2; image3;

Contents

The main goal of this assignment is to find organs of interest in CT images, using techniques of binary vision in Matlab. The organs of interest are the kidneys, the liver, and the spleen, as shown below.

original kidney image	isolated regions

You should do this assignment in Matlab. For those who don't know Matlab, here is a Matlab script that works for me. You can use it, adjust it, etc.

kidney = imread('kidney.jpg');
imshow(kidney);
k = find(kidney>128);
kidthresh = zeros(512,512);
kidthresh(k) = 1;
figure;
imshow(kidthresh);
d8 = strel('disk',8);
kidopen8 = imopen(kidthresh,d8);
figure;
imshow(kidopen8);
d1 = strel('disk',1);
kidclose1 = imclose(kidopen8,d1);
figure;
imshow(kidclose1);
b = bwlabel(kidclose1,8);
ll = label2rgb(b);
figure;
imshow(ll);

Explanation

After the image has been thresholded, morphological operators are used to break bridges, remove noise, close holes. the morphological functions are imopen, which performs an opening with a disk of size 8 and imclose, which performs a closing with a disk of size 1.

After noise is removed, the function bwlabel finds the connected components of the thresholded image and assigns a unique integer label to each component. Since we want to see the results in colors, the function label2rgb is used to produce the final image with the results.

To Do

• Try the given Matlab code on all 3 images
• This should give good results on kidney.jpg, but not necessarily on the others.
• Try other thresholds and other structuring element sizes for the other two images, trying to get the best result you can for each image.

Turn-In

This warm up assignment can be done alone or in groups. If you do not know Matlab, we will pair you with someone who does. The idea is to get familar with Matlab and see what simple operations can do.