Multidimensional particle swarm optimization-based unsupervised planar segmentation algorithm of unorganized point clouds
Title of Paper:
Multidimensional particle swarm optimization-based unsupervised planar segmentation algorithm of unorganized point clouds
Summary:
This paperpresentsanunsupervisedplanarsegmentationalgorithmofunorganizedpointcloudsbased
on multidimensional(MD)particleswarmoptimization(PSO).Arobustobjectivefunctionofthe
unsupervisedplanarsegmentationisestablishedaccordingtoclusteringdistancesofPSOclustering
algorithmandinliersofrandomsampleconsensus(RANSAC)method.Afterthat,MDPSOalgorithmis
adoptedtooptimizetheobjectivefunction,wheretheoptimalnumberandpositionsofthesegmented
planarpatchesaresoughtsimultaneously.Inordernottogettrappedinlocaloptima,amodification
strategy oftheglobalbest(GB)positionofswarmineachdimensionisaddedtotheMDPSOalgorithm.
Thus theunsupervisedplanarsegmentationofpointcloudsisrealized.Experimentalresultsdemon-
strate thehighplanarsegmentationaccuracyoftheproposedalgorithm.
Co-author:
Wang Lin, Cao Jianfu, Han Chongzhao
Papers
