Algorithmic clustering of LiDAR point cloud data for textural damage identifications of structural elements

• Published in: Measurement : journal of the International Measurement Confederation Vol. 108; pp. 77 - 90
• Main Authors: Hou, Tsung-Chin, Liu, Jen-Wei, Liu, Yu-Wei
• Format: Journal Article
• Language: English
• Published: London Elsevier Ltd 01.10.2017; Elsevier Science Ltd
• Subjects: Accuracy; Algorithms; Change detection; Cluster analysis; Clustering; Computing time; Damage detection; Data clustering; Data mining; Environmental effects; Intensity; Lidar; Point cloud data; Rain water; RGB; Sensors; Structural damage; Structural health monitoring; Structural members; Three dimensional models; Structural health monitoring; Damage detection; Point cloud data; RGB; Intensity; Data clustering

•KM/FCM performed better than SC/DBSCAN for the cases in this study.•Intensity PCD was shown to be more appropriate for detecting metal rusting.•Intensity PCD was less affected by environmental factors.•Intensity PCD can more accurately reflect the textural changes.•The best clustering number for da...