Find easy patterns in a logic vector

Hello everyone,

I am a little bit confused performing an easy procedure.

I have a long logic vector, for example:

R = [1 0 1 1 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 1 0 0 1]

It would be nice to localize this information in smalls groups.

For instance, in this position u have a group of three, here a group of 5, here only one.

Is there any specific command to do that?

Thanks!

Answer 1

[Daniel Shub]

It doesn't handle the edges perfectly, but you might want to start with

diff(find(diff(R)))

The diff( R ) is a logic vector indicating if the ith element is end of a group. The find() gets the indices on which group end. The outer diff() then determines how many elements are between group endings.

[fjnb86]

Thanks Daniel

Very useful

Answer 2

[Image Analyst]

I'm not sure what you mean by "localize" but maybe you mean "connected components labeling." If you have the Image Processing Toolbox, you can identify each groups of connected 1's with a label number:

R = [1 0 1 1 0 0 1 1 1 0 0 0 0 1 1 1 1 1 0 0 1 0 0 1]
labeled_R = bwlabel(R)

In the command window, it shows:

labeled_R =
  1 0 2 2 0 0 3 3 0 0 0 0 4 4 4 4 4 0 0 5 0 0 6

If you want the indexes of each group (object, or "blob"), then you can call regionprops().

indexes = regionprops(labeled_R, 'PixelIdxList');