what does this line means in the following code? I=I(200:400,250:450,:); and when i change the ratio there comes an error of "index exceeds matrix dimensions"

1 view (last 30 days)
I=imread('images/Persepolis_Taureaux.jpg');
I=I(200:400,250:450,:);% about M1:60second % M2:70second % M3:60second
tic
S=regcovsmooth(I,10,4,0.2,'M1');
toc

Accepted Answer

Andrew Newell
Andrew Newell on 18 Apr 2017
Edited: Andrew Newell on 19 Apr 2017
The function imread retrieves a 3D array consisting of 2D images in various colors or grayscale. The number of these images depends on the size of the third dimension (see Multidimensional arrays). They may, for example, be the red, green and blue components of the total color. The command I=I(200:400,250:450,:) is taking a rectangular sample out of each of these images. As for how regcovsmooth deals with it, this is software from GitHub that I'm not familiar with.
  5 Comments

Sign in to comment.

More Answers (2)

msahar
msahar on 18 Apr 2017
Edited: msahar on 19 Apr 2017
I am assuming that you are particularly asking about regcovsmooth function. This function is used for preserving image smoothing via region covariances. The output S is the structure component extracted from input image I.
Please read the research paper at the link given below,
  3 Comments

Sign in to comment.


Image Analyst
Image Analyst on 18 Apr 2017
regcovsmooth
It probably wants a gray scale image, not a color image. At least that's what I'd try to get rid of the error message.
tic
I = rgb2gray(I);
S = regcovsmooth(I,10,4,0.2,'M1');
By the way, please pick a more descriptive variable name than I. It looks too must like 1 (one) and l (lower case L).
  5 Comments
Image Analyst
Image Analyst on 19 Apr 2017
The code has virtually no comments and that makes it very hard for someone other than the author to support it. Either that was their intent, or they just don't write very good code. I'd have to spend quite a bit of time to understand what that code does, and you can do that just as well as we can. I suggest you contact the author.

Sign in to comment.

Community Treasure Hunt

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

Start Hunting!