How do I find the start position (index) of a cluster of non-zero values in a vector?

Question

Rafael 2015-2-8

评论： Star Strider 2015-2-8

Lets say I have a vector that is populated with zeros apart from in a few clusters where its populated by ones.

I.e. x= [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1 ...]

Does anyone have any ideas/code on how to easily return the indices of the start of each cluster? I.e. so I'd get

start_positions=[11 29 42 ...]

It has to be the start of the clusters, I've already got a code that would be able to find the end of them.

Thanks a bunch!

Answer 1

Star Strider 2015-2-8

You could combine these into one statement if you want to. I kept them separate so you can explore ‘dx’ to see how the code works:

x= [0 0 0 0 0 0 0 0 0 0 1 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 0 0 0 0 0 0 0 0 0 1 1 1 1];
dx = diff([0,x]);
x1idx = find(dx > 0)

produces:

x1idx =
      11    29    42

The leading ‘0’ in the diff call vector makes the indices come out the same between ‘x’ and ‘dx’.

Rafael 2015-2-8

Good and very fast answer, thanks!

Star Strider 2015-2-8

My pleasure!