K-NN and K-means algorithms in Lidar point cloud classification

  • Affiliations:

    Khoa Công nghệ thông tin, 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: 20th-June-2017
  • Revised: 20th-July-2017
  • Accepted: 30th-Oct-2017
  • Online: 30th-Oct-2017
Views: 1108
Downloads: 635
Rating: 5.0, Total rating: 63
Yours rating


The k-NN (k - Nearest Neighbor) algorithm is common algorithms for data mining. K-means is a clustering algorithm belonging to the unsupervised classification, with the idea of grouping objects into k clusters with the focus of each clustered change after each iteration. k-NN is the supervised classification, which calculates the distance from the object to the center of the cluster, finds the smallest distance value, and assigns the object to the corresponding class. This article focuses on the applied research of two K-means and k-NN algorithms into the Lidar cloud point classification - high accuracy remote sensing data and large number of points. With a test data set of 485 points measured in Nghe An, the classification result based on the elevation point value indicates that the error value classified with two algorithms still accounts for the high K -means (31.5%) and the k-NN algorithm is 48.4%.

How to Cite
Nguyen, P.Huu Thi 2017. K-NN and K-means algorithms in Lidar point cloud classification (in Vietnamese). Journal of Mining and Earth Sciences. 58, 5 (Oct, 2017).