Southern African Science Service Centre for Climate Change and Adaptive Land Management BMBF
SASSCAL Data and Information Portal
Open Data and Information on Climate Change and Adapted Land Management in Southern Africa

Ground truthing woody plant species to compare with aerial imagery in biodiversity observatories, Namibia (Bachelor thesis)

Map Map


Title
Title Ground truthing woody plant species to compare with aerial imagery in biodiversity observatories, Namibia (Bachelor thesis) ?
Author Esmé Roshané //Hoёses ?
Abstract The main objective of this study was to find out if woody plant species could be identified on drone derived aerial imagery based on ground truthed GPS and crown diameter measurements. In addition, the study explored three different methods to make the connection between imagery with what is actually being seen on the ground. Ariel imagery was acquired for three biodiversity observatories namely Mile 46, Mutompo and Sonop using the eBee drone, a 1 kilometre squared image was obtained for an observatory which used an RGB and (NIR) Near infrared camera with a 5 cm ground resolution and flying was done through an automated procedure. Results for comparing ground truth crown diameter measurements with the imagery showed no difference for both Mutompo and Sonop while a trend was observed in Mile 46. GPS points that could be positively identified on the imagery were the highest in Sonop (72) percent compared to Mile 46 (16) and Mutompo (27) percent using the random selection method. The best method was found to be selecting trees from the imagery before ground truthing as this method has shown 100 percent accuracy to identify trees from imagery based on GPS and crown diameter measurements. In order to positively identify trees on the imagery, knowledge of the trees in the observatory is essential. ?
Dataset
Document Reference Date Type creation ?
Date 2017-12-31 ?
Language English ?
Online Linkage ?
Associated project SASSCAL (Phase 1) ?
Subproject 062 Vegetation Survey of Namibia ?
Dataset Classification
Type PDF ?
Category thesis ?
Geographic Location
Geographic Description Namibia ?
Metadata
Metadata Contact Person Strohbach, Ben, Dr ?
Metadata Date Stamp 2019-10-02 ?
Identifier
Internal identifier sdp_doc_documents_6623 (Link)