Neural Network Code for Image Classifica​tion/Segme​ntation

1 view (last 30 days)
I have written a code in Matlab for neural network classification/segementation the image after train the image. But, it is not working as expected. I get the error in this line "yhat = net(double([r;g;b]));". Is there any mistake in the code? Can you take a look at the code and help me out !!
a = imread('C:\Users\user\Desktop\MATLAB PROCESSING\Matlab Processing\Band432\open land\klang.tif');
b = rgb2gray(a);
figure, imshow(a);
figure, imshow(b);
thr = 0.5;
for i = 1:1:size(a,1)
for j = 1:1:size(a,2)
pix = a(i,j,:);
r = pix(1,1,1);
g = pix(1,1,2);
b = pix(1,1,3);
yhat = net(double([r;g;b]));
if yhat >= thr
result = 1;
b(i,j) = 255;
else
result = 0;
b(i,j) = 0;
end
end
end
figure3, imshow(b)
or any suggestion code for classify the image in 5-class
I really hope can get any feedback because I really not familiar with neural network toolbox
  1 Comment
Greg Heath
Greg Heath on 28 Aug 2015
I don't know. However, searching on
NEURAL IMAGE SEMENTATION
resulted in 32 hits in ANSWERS and 33 hits in the NEWSGROUP.
Hope this helps.
Greg

Sign in to comment.

Answers (0)

Community Treasure Hunt

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

Start Hunting!