Hi
I have the following result of segmentation
How can I improve the results?
1) I want to improve the appearance to get the vessels more smooth.
2) I want to extract the oval-like part of the vessels in the middle of the image.
Any suggestions?
*****************
Here is a link to the image
https://copy.com/GT6n8EF82BKw

 Respuesta aceptada

Evan
Evan el 31 de Jul. de 2014
Editada: Evan el 31 de Jul. de 2014
For your first question:
Is this strictly for appearances? If so, does medfilt2 do what you want?
rawImg = rgb2gray(imread('seg_vessel.jpg'));
nbhd = [2 2]; %play around with neighborhood size to get the smoothness you want
smoothedImg = medfilt2(rawImg,nbhd);
figure;imagesc(rawImg); colormap(gray)
figure;imagesc(smoothedImg); colormap(gray)

1 comentario

Nisreen Sulayman
Nisreen Sulayman el 31 de Jul. de 2014
Editada: Nisreen Sulayman el 31 de Jul. de 2014
thanks Mr. Evan, I've tried medfilt2 ... can you suggest another thing ... the above code didn't gave the desired effect... I tried to use nbhd=[3 3] unfortunately it separates the vessels path

Iniciar sesión para comentar.

Más respuestas (2)

SRI
SRI el 31 de Jul. de 2014

0 votos

inputImage = imread('seg_vessel.tif');
grayImage = rgb2gray(inputImage);
binaryImage = im2bw(grayImage,level);
[B,W,D,T] = bwboundaries(binaryImage);
imshow(W)
Hi Kindly Look into the code, this might be the result which you had expected
SRI
SRI el 31 de Jul. de 2014

0 votos

inputImage = imread('seg_vessel.tif');
grayImage = rgb2gray(inputImage);
level = graythresh(grayImage);
binaryImage = im2bw(grayImage,level);
[B,W,D,T] = bwboundaries(binaryImage);
imshow(W)
Sorry Use this one

3 comentarios

Nisreen Sulayman
Nisreen Sulayman el 31 de Jul. de 2014
Editada: Nisreen Sulayman el 31 de Jul. de 2014
I haven't got the desired result this is what I got ...
I want to get ride of small line and elongated parts which give "coarse look" of the image above
SRI
SRI el 1 de Ag. de 2014
Hi
You need to crop that oval part alone in this image
Image Analyst
Image Analyst el 1 de Ag. de 2014
You can use imopen() to snip off pointy parts or use imclose() to blur out the boundary.

Iniciar sesión para comentar.

Preguntada:

el 31 de Jul. de 2014

Comentada:

el 1 de Ag. de 2014

Community Treasure Hunt

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

Start Hunting!

Translated by