1207 
THE PHOTOGRAMMETRIC POTENTIAL OF LOW-COST UAVs IN FORESTRY AND 
AGRICULTURE 
G. J. Grenzdörffer a> *, A. Engel b , B. Teichert c 
a Rostock University, Chair for Geodesy and Geoinformatics, J.-v.-Liebig Weg 6, 18059 Rostock, Germany - 
goerres. grenzdoerffer@uni-rostock. de 
b Kampfmittelräumdienst Hamburg, Großmoorbogen 8, 21079 Hamburg - aengell980@googlemail.com 
c University of Applied Sciences Dresden, Department of Surveying and Cartography, Friedrich- List- Platz 1, 01069 
Dresden - teichert@htw-dresden.de 
Commission IICWG I/V - ThS-23 
KEY WORDS: Digital airborne imaging systems, UAV, Agriculture, Accuracy assessment, Forestry 
ABSTRACT: Micro-UAVs (Unmanned-Airborne-Vehicles or drones) with a total weight below 5 kg are interesting alternative 
carriers for agricultural and forestry applications. Compared to standard airborne aerial surveys UAVs are much more flexible and 
weather independent. As a result micro-UAV surveys will pave the way for affordable, current and accurate geo-information. 
Practical tests with two different systems at several locations revealed that both systems were capable of acquiring images in a 
systematic manner. However the necessary post processing effort in order to obtain photogrammetric products suitable for a GIS was 
quite high. The photogrammetric potential for direct georeferencing of micro-UAVs is quite high, but until now has not been fully 
exploited. This is primarily due to the fact that the manufacturers of UAVs are not aware and familiar with the special requirements 
of photogrammmetry and GIS data acquisition, e.g. metric cameras, systematic aerial surveys, precise values of the exterior 
orientation. 
1. INTRODUCTION 
Remote sensing applications for agriculture and forestry often 
require images with a high temporal resolution, e.g. 
Grenzdorffer, 2003. This is difficult and / or costly to obtain, 
either by satellite imagery or by conventional airborne data. 
Therefore, unmanned drones equipped with GPS and digital 
cameras, so called Unmanned Aerial Vehicles (UAVs), have 
become a focus of research. The autonomous navigation of an 
UAV is realised using GPS, inertial measuring techniques and 
the utilisation of other sensors. Only the programmable autopi 
lot enables serious photogrammetric work, thus enabling sys 
tematic, rapid and efficient mapping of areas of interest. Re 
motely piloted vehicles (RPV) with a video downlink are not 
suited for photogrammetric work because the navigation and 
the image triggering can not be done systematically. The UAV 
technology is mainly driven by the military (>80 %) (UAS, 
2007) but there are also developments in the civilian sector, e.g. 
Eisenbeiss, 2004. In Germany so called Micro-UAVs with a 
total weight of less than 5 kg may be used in the uncontrolled 
airspace below 300 m. Other restrictions however apply, espe 
cially in urban areas, which means that applications in sparsely- 
inhabited areas e.g. for forestry, nature conservation and agri 
culture are at the forefront of micro-UAV research. 
1.1 Applications in agriculture and forestry 
In forestry and nature conservation UAVs may be used for 
many applications, such as (Horcher und Visser, 2004): 
• Forest fire detection 
• Monitoring for legal restrictions and evidence in case of 
violations / infringements 
• Locating harvest sites and inspecting forestry operations 
• Monitoring and change detection within natural forests, 
where trespassing is difficult or undesirable 
In agriculture UAVs may be used for: 
• Field trials and research, e.g. Annen und Nebiker, 2007 
• Determination of the biomass, crop growth and food 
quality, e.g. Herwitz et al., 2004 
• Precision Farming, e.g. to determine the degree of weeds 
for site specific herbicide applications. Also a reduction 
of fungicides due to site specific applications in potato 
production is of interest, e.g. Grenzdorffer, 2003 
Reidelsturz et al., 2007 
• Senescence monitoring of cereals and maize for harvest- 
and logistic optimisation. 
All of the mentioned fields of application are characterised by a 
relatively small extent of less than 1.500 ha and the necessity of 
a rapid data availability and data analysis. The required 
absolute positional accuracy is on the other hand quite low (< 1 
m). Beside a visual analysis of the images many of the above 
mentioned applications require images to be integrated as 
georeferenced and orthorectified products in a GIS for analysis 
with other GIS data. 
2. STATE OF THE ART 
Internationally there is little work on the issue of UAVs and 
agriculture. Some examples may be found e.g. in the US a 
RPV-system named Crop Condor 
(http://www.calmarlabs.com/condor.html) was developed. In 
* Corresponding author