Florian Männer

Florian Männer earned two bachelor degrees from the Technical University of Munich: one in vocational education (B.Ed.) for the subjects of agriculture and mathematics and the other in agricultural and horticultural sciences (B.Sc). He then completed an M.Sc in agricultural sciences with a major in plant production systems at the University of Hohenheim.

Currently he is at the University of Bonn working on a doctoral dissertation on the topic of remote hyperspectral sensing for analyzing fodder vegetation and the degradation of savanna ecosystems in Namibia.

Since 2022 Florian has also been a project leader and product owner at Fraunhofer IGD in Rostock, where he heads the crop working group of the smart farming department. His research interests range from near and remote sensing technologies across ground-level robotic and tractor platforms to the use of drones and satellites for imaging farmland, grasslands, and moorlands. In terms of methods, he is addressing RGB, multispectral and hyperspectral sensor technology, and two- and three-dimensional imaging with machine learning.

Farm- and grassland

Grassland research

Disease and stress detection in strawberries

• Use of neuronal networks to identify species in images captured by drones for monitoring biodiversity and populations in grasslands and detecting weeds on farmland
• Use of 3D dot clouds and hyperspectral data to determine biomass and fodder quality in order to support the management of grasslands and pastures
• Use of RGB and multispectral and hyperspectral data from captured unitemporal or multitemporal images to detect  damage inflicted by diseases, pests, and wild animals
• Satellite-based analyses of population trends on farmlands, grasslands, and moorlands
• Breeding assessments based on images taken by camera-equipped drones and robots for capturing biomass development, plant heights, flowering times, and estimated yields within the scope of practical plant breeding
• Use of ground- and air-based multispectral images and machine learning algorithms to identify weed species and developmental stages and enable optimal, resource-saving, integrated crop protection
• Monitoring of moorlands by capturing multimodal data from camera images and soil and climate sensors for estimating carbon sequestration, biodiversity trends, and water levels
• Machine learning and analysis of multi- and hyperspectral image data of plant material for analyzing the quality of harvested crops in the pre- and post-harvest process chain