Airborne LiDAR point cloud filtering method based on active learning

• Main Authors: HUI ZHENYANG, LI DAJUN, HU HAIYING, WANG LEYANG, LU TIEDING
• Format: Patent
• Language: Chinese; English
• Published: 13.08.2019
• Subjects: CALCULATING; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; HANDLING RECORD CARRIERS; IMAGE DATA PROCESSING OR GENERATION, IN GENERAL; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS

The invention discloses an airborne LiDAR point cloud filtering method based on active learning. The airborne LiDAR point cloud filtering method comprises the steps of S1, obtaining point cloud data and removing low-order noise points; S2, adopting multi-scale morphological operation to automatically obtain and mark a training sample set; S3, performing feature extraction on the training sample set and establishing an SVM model; S4, classifying the candidate sample set by adopting a training model, wherein the candidate sample set is divided into a candidate ground point set and a candidate non-ground point set; setting the oracle as an S-shaped function of the distance from the candidate point set to the fitting curved surface, respectively selecting q points from the candidate ground point set and the candidate non-ground point set in each iteration, adding the q points into the training sample set, updating the training model, performing iteration all the time until the number of the point clouds in the ca