:: Volume 2, Issue 2 (9-2014) ::
jgit 2014, 2(2): 75-90 Back to browse issues page
Improving the result of LiDAR data filtering algorithms using mathematical morphology
Abstract:   (3483 Views)
Today, aerial laser scanners (LiDAR) have an important role in 3D data acquisition from features. Bare earth information is very important in deferent applications such as DTM extraction, determination of traversable area, etc. Up to now, a lot of algorithms have been developed to automated filtering of LiDAR data. The weakness of most of these algorithms is inability to remove the large buildings. The main aim of this paper is to solve of this problem. Mathematical morphological operators were used for this purpose. First, the LiDAR data was filtered using one of the most efficient filtering algorithms (slope based filtering algorithm). Afterward, the result of filtering stage was improved by perform the proposed algorithm that is based on mathematical morphological operators. The result of accuracy assessment indicate a negligible increase in type I error and significant decrease in type II and total errors. Since, in filtering process, the type II and total errors are more important than type I error, performing this supplementary processing present very good result. Quantitative evaluation shows the output of the improved slope based algorithm with 20º slope threshold present the best result. In this case type I error increased from 4.98% to 5.27%, type II error reduced from 9.043% to 4.44% and total error decreased from 7.03% to 4.85%. Qualitative evaluation indicates the good performance of the proposed algorithm in removing the large buildings which are remained from filtering stage. Slope based filtering algorithm and Mathematical morphological operators were implemented in MATLAB software.
Keywords: Aerial laser scanners (LiDAR), Filtering, Slope based filtering algorithm, Mathematical morphology.
Full-Text [PDF 1621 kb]   (942 Downloads)    
Type of Study: Research |
Received: 2015/09/8 | Accepted: 2015/09/8 | Published: 2015/09/8

XML   Persian Abstract   Print

Rights and permissions
Creative Commons License This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
Volume 2, Issue 2 (9-2014) Back to browse issues page