count a vector with continuous non zero elements

2 views (last 30 days)
Sam
Sam on 16 May 2014
Edited: the cyclist on 16 May 2014
Is there a way to count non-zero elements in vector x below x=[0 1 1 1 0 0 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 0 1]
so that the output y will be y=[3 5 7 1]

Answers (2)

Azzi Abdelmalek
Azzi Abdelmalek on 16 May 2014
x=[0 1 1 1 0 0 1 1 1 1 1 0 0 0 0 1 1 1 1 1 1 1 0 1]
ii=strfind([0 x 0],[0 1])
jj=strfind([0 x 0],[1 0])
out=jj-ii
  2 Comments
Sam
Sam on 16 May 2014
thanks Azzi, can you explain to me what is this doing?
Azzi Abdelmalek
Azzi Abdelmalek on 16 May 2014
ii=strfind([0 x 0],[0 1]) % find the indices corresponding to the switch from 0 to 1

Sign in to comment.


the cyclist
the cyclist on 16 May 2014
Edited: the cyclist on 16 May 2014
Here's a similar approach:
y = diff(find([0 x 0]==0))-1;
y(y==0) = []
The first line identifies the zero locations, and then the "distance" between them. This distance is the length of the string of ones.
The second line removes the instances where that distance is zero (i.e. no 1's), since you are not interested in those.
Note that the reason you need to do this operation on "[0 x 0]" instead of just x is to be able to identify a string of ones at the beginning and end.

Community Treasure Hunt

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

Start Hunting!