sliding the rectangle using nlfilter

hai

i am new to matlab and studing about optic disc localization

in that my work is to develop two feature maps namely I horizontal map II vertical map

and then create a rectangle and move over the feature map and calculate the sum of pixels inside the rectangle and move the rectangle all over the image and calculate the sum of pixel values in each point.

my third step is i have to plot the values and draw a graph

now i had generated two feature maps and i have created a rectangle also

i have strucked in making the rectangle to move

please tell me wheather an function is available in matlab for moving the rectangle over the image

i want the rectangle to move automatically and not a dragable rectangle and i have trid with imrectwhicvh is of the form of draggable rectangle

please help me

and the code i have used is

 *r=rectangle('position',[10,10,30,560],'linewidth',2.4,'edgecolor','w');  
 func = @(x) sum(x.cdata(:));
   B = nlfilter(final2,r,func);
   figure,imshow(B);*

please hep me

thanks

hai jan Simon

as you mentioned a have used nlfilter function but i am getting the result as NAN and the time taken for processing also high s0 i used colfilt function but the answer is not coming

what can i do now please tell me thanks

[Walter Roberson]

duplicates are at http://www.mathworks.com/matlabcentral/answers/13866-move-rectangle-over-te-image and http://www.mathworks.com/matlabcentral/answers/13444-to-make-the-rectangle-move-over-the-image

[Jan]

@kanagapushpavalli: Please post the command you have used, otherwise we cannot guess what's going wrong.

Answer 1

[Walter Roberson]

Please read the responses to when you asked this before. http://www.mathworks.com/matlabcentral/answers/13866-move-rectangle-over-te-image and http://www.mathworks.com/matlabcentral/answers/13444-to-make-the-rectangle-move-over-the-image

I recommend against the approach you have taken. Please see my answer to the 13444 post.

Answer 2

[Jan]

Drawing a rectangle will not help. You want to process rectangular sections of the image data.

Please read the documentation of help nlfilter. There you find: "B = nlfilter(A, [m n], fun)" with m and n are the dimensions of the rectangle and fun is @sum in your case.