2 dimensional array algorithm question

[Zarathustra]

I'm sure there must be a simple algorithm for giving the number of
"ones" that are connected, for example in an array like this:
int a[5][5] = {0};
a[0][0] = a[0][1] = a[1][1] = a[2][1] = a[0][4] = a[3][4]= a[4][4] =
a[4][0] = 1;
with all the rest still equal to zero.

1 0 0 1 1
0 0 0 0 0
0 0 0 0 0
1 1 1 0 0
1 0 0 0 1

such that the output in this case, can give 1 group of size 4, 1 group
of size 2, and 2 groups of size 1.

It looks simple, but everything I've tried so far hasn't caught all
possible shapes of groups of "1"s.

Any ideas would be greatly appreciated.

[Paul Mesken]

On 12 Jun 2005 11:16:16 -0700, "Zarathustra"
<hu************@hotmail.com> wrote:

I'm sure there must be a simple algorithm for giving the number of
"ones" that are connected, for example in an array like this:
int a[5][5] = {0};
a[0][0] = a[0][1] = a[1][1] = a[2][1] = a[0][4] = a[3][4]= a[4][4] =
a[4][0] = 1;
with all the rest still equal to zero.

1 0 0 1 1
0 0 0 0 0
0 0 0 0 0
1 1 1 0 0
1 0 0 0 1

such that the output in this case, can give 1 group of size 4, 1 group
of size 2, and 2 groups of size 1.

It looks simple, but everything I've tried so far hasn't caught all
possible shapes of groups of "1"s.

Any ideas would be greatly appreciated.

Jburgy mentioned in thread "Clustering in C" the Hoshen-Kopelman
algorithm. It's very simple. Just use it and then count the number of
cells for each cluster.

Here's a link :

http://splorg.org:8080/people/tobin/...nkopelman.html

[Zarathustra]

Paul Mesken wrote:

On 12 Jun 2005 11:16:16 -0700, "Zarathustra"
<hu************@hotmail.com> wrote:

I'm sure there must be a simple algorithm for giving the number of
"ones" that are connected, for example in an array like this:
int a[5][5] = {0};
a[0][0] = a[0][1] = a[1][1] = a[2][1] = a[0][4] = a[3][4]= a[4][4] =
a[4][0] = 1;
with all the rest still equal to zero.

1 0 0 1 1
0 0 0 0 0
0 0 0 0 0
1 1 1 0 0
1 0 0 0 1

such that the output in this case, can give 1 group of size 4, 1 group
of size 2, and 2 groups of size 1.

It looks simple, but everything I've tried so far hasn't caught all
possible shapes of groups of "1"s.

Any ideas would be greatly appreciated.

Jburgy mentioned in thread "Clustering in C" the Hoshen-Kopelman
algorithm. It's very simple. Just use it and then count the number of
cells for each cluster.

Here's a link :

http://splorg.org:8080/people/tobin/...nkopelman.html

Thanks very much. That was exactly what I was looking for.

[googmeister]

It suffices to use depth-first search or breadth-first
search to label all of the clusters. Union-find is
probably overkill here since your graph is fixed.