I am segmenting liver from CT data. Its 2d segmentation on multiple imaging. I trained my system using deep learning method. Now the results of segmentation are as follows...
I want to remove the circle parts from my image. I already tried bwareaopen() function from matlab. Using bwareaopen() function I got the whole image white. please help me.
Thanks

댓글 수: 2

KSSV
KSSV 2017년 6월 5일
If you have the locations of those highlighted images, cant you replace those pixel values with black?
Ahmad
Ahmad 2017년 6월 6일
I want to do automatically. sunkara manoj code working perfectly but now I need to smoothen the boundaries.

댓글을 달려면 로그인하십시오.

 채택된 답변

sunkara manoj
sunkara manoj 2017년 6월 5일

1 개 추천

the result is a binary image but the part is extracted.
if true
I=imread('liver.png');
im = imclearborder(im2bw(I));
im_fill = imfill(im, 'holes');
s = regionprops(im_fill, 'Area', 'PixelList');
[~,ind] = max([s.Area]);
pix = sub2ind(size(im), s(ind).PixelList(:,2), s(ind).PixelList(:,1));
out = zeros(size(im));
out(pix) = im(pix);
imshow(out); end

댓글 수: 3

Ahmad
Ahmad 2017년 6월 5일
편집: Ahmad 2017년 6월 5일
Thanks Sankara Manoj its work very well. But now i want to smoothen the borders. so if you have an idea so please share.
KSSV
KSSV 2017년 6월 6일
How about using convexhull?
Ahmad
Ahmad 2017년 6월 10일
I never tried. I dont know the purpose of convex hull.

댓글을 달려면 로그인하십시오.

추가 답변 (0개)

질문:

2017년 6월 5일

댓글:

2017년 6월 10일

Community Treasure Hunt

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

Start Hunting!

Translated by