Flesh out the member functions for performing connected components clustering (clusterByProximity and connectivityMatrix). This will involve:

1. Completing the connectivity matrix member function.
2. Parsing the matrix to find connected components.
3. Giving each component a different color for plotting purposes.
4. Returning the label of each point in a vector list.

Be careful when you try to run this on arbitrary data. Because the connected components matrix is memory intensive, it can only be done with point clouds that are relatively small (on the order of 20 thousand points). For more than that many points, the best option is to utilize Matlab's internal pointCloud member functions to request the local neighborhood set.

Notes: Matlab functions that will be useful include: pdist, squareform. You will also need to figure out how to implement the algorithm through judicious use of temporary variables.

Explore & Deliverables: Apply to the set of points from the Week #1 data file. The data file clearly has distinct connected components. Show that your code does grab them properly.

Back