Classification of LiDAR data based on region segmentation and decision tree

Kai-si Liu; Yan-bing Wang; Hui-li Gong

doi:10.1117/12.2069203

26 November 2014 Classification of LiDAR data based on region segmentation and decision tree

Kai-si Liu, Yan-bing Wang, Hui-li Gong

Proceedings Volume 9262, Lidar Remote Sensing for Environmental Monitoring XIV; 926213 (2014) https://doi.org/10.1117/12.2069203
Event: SPIE Asia-Pacific Remote Sensing, 2014, Beijing, China

Abstract

Aiming at spatial characteristics and echo information of the LiDAR point cloud data, design a regional segmentation and decision tree combined lidar point data classification method. First, based on the continuity of the LiDAR point cloud to finish the experiment area's region segmentation. Then, statistics each area boundaries and internal the number of dihedral angle cosine, to draw a line chart. Using the intersection's cosine of line chart , and region segmentation's minimum height as threshold to determine the ground point and the non-ground points. Finally, statistics separately all LiDAR point data set's dihedral angle, echo times, echo intensity, mean elevation, four constraint information to build a decision tree to determine which type of feature vesting each divided region. Using classification confusion matrix to assess the classification's accuracy, overall accuracy is higher than 94%. Experimental results show that this method can effectively separate roads, trees, buildings and terrain.

Citation Download Citation

Kai-si Liu, Yan-bing Wang, and Hui-li Gong "Classification of LiDAR data based on region segmentation and decision tree", Proc. SPIE 9262, Lidar Remote Sensing for Environmental Monitoring XIV, 926213 (26 November 2014); https://doi.org/10.1117/12.2069203

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available