what are your opinions to modify this code for improvement the result?

1 view (last 30 days)
i want to implement contrast enhancement, thresholding, morphological opening, Edge detection, morphological erosion and skeketonization
i wrote the below code , but the result was not enoughly suitable for me.
so what are your proposals to obtain best result ?. also if i want to do local thresholding how can i do it ? what will change in thresholding step?.
m=imread();
%%%%%1-Contrast enhancement%%%
x=adapthisteq(m);
%%%%%%2-global thresholding%%%%%%
level=graythresh(x);
m1=im2bw(x,level);
%%%%3-Morphological opening%%%%%%%
BW2 = bwmorph(m1,'open');
%%%%%%%%%%4-Sobel Edge detector%%%%%%
w=fspecial('sobel');
edge=imfilter(BW2,w,'replicate');
%%%5-morphological erosion%%%%%%%%
BW3 = bwmorph(edge,'erode');
%%%6-morphological skeketonization%%%%%%
W3 = bwmorph(BW3,'skel') ;
% displaying images%%%
figure,imshow(m);
figure,imshow(x);
figure,imshow(m1);
figure,imshow(BW2);
figure,imshow(edge);
figure,imshow(BW3);
figure,imshow(W3);
thank you
  2 Comments
Walter Roberson
Walter Roberson on 19 Jan 2012
We do not know what _would_ be "suitable enough for you".
Past discussions about this project tend to suggest to us that you are hoping to extract more information than is present in the image.
mmm ssss
mmm ssss on 20 Jan 2012
yes, this is right Mr.Walter.
do you have any modications to the above code?

Sign in to comment.

Answers (1)

Sean de Wolski
Sean de Wolski on 20 Jan 2012
I would rename your variable edge to a name that doesn't overwrite the IPT stock function edge().

Categories

Find more on Image Processing Toolbox in Help Center and File Exchange

Community Treasure Hunt

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

Start Hunting!