Vertical Accuracy of a Ground Filtered VA V-derived DEM Using Cloth Simulation Filtering Algorithm

Abstract

Collecting earth surface data using $U$ nmanned Aerial Vehicles (VA V) become more easy and accurate in the recent time. Based on photogrammetry, point clouds data can be produced from small format aerial photography (SF AP) using Structure from Motion (SfM) to make 3D model called Digital Surface Model (DSM). To create Digital Elevation Model (DEM), ground filtering must be applied to DSM. LiDAR point clouds data is common source data to create DEM. In this research, DEM is created by executing ground filtering algorithm called Cloth Simulation Filtering (DSM) to VA V-derived DSM with different cloth resolution. The aim of this research is to examine vertical accuracy of VA V-Derived DEM using Cloth Simulation Filtering method. The result accuracy from this algorithm calculated according Badan Informasi Geospasial (Geospatial Information Agency of Indonesia) vertical error calculation. The result show different accuracy for each cloth resolution. Vertical accuracy from 0.5, 1, and 2 size of cloth resolution consecutively is 2.34 m, 4.0 m error, and 26.27 m.