Research Article
3D Point Cloud Simplification Based on k-Nearest Neighbor and Clustering
Algorithm 1
Simplification of 3D point cloud based on the clustering algorithm and Shannon’s entropy.
| Input | (i) | : the data sample (point cloud) | (ii) | : the array in which cluster indexes are stored | (iii) | : the number of clusters | (iv) | : the number of clusters to delete () | (v) | : minimal entropy | (vi) | Begin | (vii) | Decomposing the initial set of points into small clusters denoting , using the k-means algorithm | (viii) | For | | For | | Calculate global entropy of a cluster by using all data samples in according to equation (7), Note this entropy | | If then | | | | pos ⟵ j | | End if | | For | | | | End for | | End for | (ix) | End for, | | | | End. |
|