Estimation of Agricultural Crop Yield Using Digital Image Analysis and Colour Cameras

PhD defence, Friday 18 May 2018, Anders Krogh Mortensen.

2018.05.18 | Trine Berndt Turtiainen Scheelke

Anders Krogh Mortensen

Example of crop distinguishing and recognition using deep learning. Left: Original image of oil radish sown as a catch crop in a barley field plot. Right: Pixel-wise classification using deep learning.

During his studies, Anders Krogh Mortensen researched yield estimation of agricultural field crops using colour cameras. To optimize yield and lower the negative environmental impact, more knowledge per area is needed in agricultural fields. Anders Krogh Mortensen studied image-based algorithms for distinguishing and recognising crops in colour images. The estimated crop coverage was used to estimate the biomass and nutrient content of the crops. Both traditional image-based algorithms and state-of-the-art deep learning-based algorithms were explored to distinguish and recognize crops in images. Anders’ research further demonstrated how image-based measurements can be used to improve simple growth models. The new research findings contribute to increased knowledge per area in agricultural fields in terms of crop coverage and yield estimation. The findings may contribute to better strategies for fertilization and harvest times, both economically and environmentally.

The PhD degree was completed at the Department of Agroecology, Science and Technology, Aarhus University.

This résumé was prepared by the PhD student.

Time: Friday 18 May 2018 at 11.00
Place: Auditorium, AU Flakkebjerg, Forsøgsvej 1, 4200  Slagelse
Title of PhD thesis: Estimation of Above-Ground Biomass and Nitrogen-Content of Agricultural Field Crops using Computer Vision
Contact information: Anders Krogh Mortensen, e-mail:, tel.: +45 60 94 16 19

Members of the assessment committee:

Professor Hans Werner Griepentrog, Institute of Agricultural Engineering, University of Hohenheim, Germany
Associate Professor Jens Michael Carstensen, Department of Applied Mathematics and Computer Science, Technical University of Denmark
Senior Scientist Henrik Skovgård (chair), Department of Agroecology, Aarhus University
Main supervisor:
Associate Professor René Gislum, Department of Agroecology, Aarhus University
Professor (Docent) Henrik Karstoft, Department of Engineering, Aarhus University
Language: The PhD dissertation will be defended in English

The defence is public.
The PhD thesis is available for reading at the Graduate School of Science and Technology/GSST,
Ny Munkegade 120, building 1520, rooms 128-134, 8000 Aarhus C.

PhD defence
Comments on content: 
Revised 17.05.2018