Essay on Airborne Lidar Data For Land Cover Classification

783 Words Apr 1st, 2015 4 Pages
Numerous studies have conducted on the use of airborne LiDAR data for land cover classification, especially with the evolution of airborne LiDAR technology. Land cover classification could be conducted using LiDAR data (height and intensity) or incorporating LiDAR data with multispectral aerial/satellite imagery.

A number of experiments have shown the use of LiDAR-derived height surfaces (e.g., DSM and nDSM) with LiDAR intensity data in classification. Brennan and Webster (2006) created five surfaces from LiDAR data, included: DSM, DEM, intensity, multiple echoes, and normalized height. A rule-based classification technique is used for segmenting and classifying these surfaces to land cover classes. The overall accuracy was 94.00% and 98.00% for ten and seven classes, respectively. Im et al. (2008) conducted a sensitive analysis on eight different LiDAR-derived surfaces from height and intensity on three different sites. An object-based classification technique was applied to eight surfaces to classify the terrain into five land cover information and achieved more than 90.00% overall accuracy. Other experiments have used statistical analysis such as skewness and kurtosis to separate non-ground from ground points from the LiDAR height and/or intensity values (Bartels et al., 2006; Yunfei et al., 2008; Liu et al., 2009).

On the other hand, the fusion of multispectral imagery and LiDAR data is conducted to maximize the benefit of different reflectivity behaviour from various…

Related Documents