Evaluation of morphological filtering in automatically classifying UAV‐derived point cloud

  • Affiliations:

    1 Khoa Trắc địa Bản đồ và Quản lý đất đai, Trường Đại học Mỏ-Địa chất, Việt Nam

  • *Corresponding:
    This email address is being protected from spambots. You need JavaScript enabled to view it.
  • Received: 28th-Jan-2017
  • Revised: 16th-Mar-2017
  • Accepted: 28th-Apr-2017
  • Online: 28th-Apr-2017
Views: 1853
Downloads: 517
Rating: 5.0, Total rating: 51
Yours rating


In Vietnam, in recent years, UAV has been being applied in civil generally and in surveying and mapping particularly. One of the most advantages of UAV is to provide a 3D dense cloud based on stereo image pairs taken by UAV, from which it is possible to obtain Digital Surface Model (DSM) or Digital Elevation Model (DEM), which is popularly used in topographical surveying. However, the very first step of generating a DEM is to classify ground and non-ground points. Recently, several automatic point could classification algorithms have been proposed, each method has its own advantages and limitations, therefore it is not easy to select appropriate one. This study conducted an experiment on filtering UAV-derived point cloud using morphological filtering, which is available in Agisoft PhotoScan Professional software. The results showed that, in area without dense shrub, morphological filtering can efficiently separate ground and non-ground points, Root-Mean-Squared Error (RMSE) of DEM generated from automatic classified point cloud and reference points was 10.4cm. However, in dense shrub areas, it is not efficient method, RMSE was 39.6cm. This was almost overcome when applying manual filtering in these areas. Therefore, it is necessary to combine morphological filtering and manual filtering to efficiently filter the point cloud.

How to Cite
La, H.Phu, Nguyen, M.Quang, Hoang, T.Anh, Dao, K.Van and Tran, T.Anh 2017. Evaluation of morphological filtering in automatically classifying UAV‐derived point cloud (in Vietnamese). Journal of Mining and Earth Sciences. 58, 2 (Apr, 2017).

Other articles