In this context, a cluster is defined as a subset of points where the distance between two neighboring points is less than a given threshold (see image). It is thought for 3D space, but might be easily tuned up (if it does not work already) for higher dimensions using euclidean distance. I believe there might be a more efficient implementation.
INPUT:
- points: Set of N points in 3D space. NX3 array.
- threshold: Maximum distance between two points to belong to the same cluster.
OUTPUT:
- clusters: MX1 cell array, with each cluster, each cluster being an array with the index of the points that make it up. M is the total number of clusters.
- cluster_id: An NX1 array mapping each point to the cluster index it belongs to.
[clusters, cluster_id] = cluster_finder(points, threshold);
