Aarhus University Seal / Aarhus Universitets segl

Image-based mapping of grass clover fields using artificial intelligence – paves the way for improved sustainability and yield of eco-friendly crop

PhD defence, Thursday 8 April 2021. Søren Kelstrup Skovsen.

2021.04.08 | Sunitha Satkunam

Søren Kelstrup Skovsen

Figure 1. Visualization of mapping. Top: Examples of automatic image recognition (green = grass, red = clover). Bottom: Mapping of field-scale clover content based on spatially distributed images across the field.

During his PhD study, Søren Kelstrup Skovsen researched the use of image recognition in mixed crops of grass and clover, commonly used as feed for the dairy industry. Targeted fertilization, based on the local clover content, can improve both yield and sustainability through better utilization of the applied nitrogen. Until this PhD study, however, it has not been economically feasible to map the clover content at a large scale. Søren Kelstrup Skovsen has studied the use of image based estimation of the clover content using artificial intelligence and synthetic data. Not only did this method provide a means for accurate evaluations of the clover content in changing conditions, but did so in an automated manner, applicable for feasible mappings at a large scale.

 

The PhD study demonstrates a viable path for introducing improved fertilization practices across 286 000 hectares of Danish farmland. The inexpensive, yet accurate, data acquisition additionally supports future agroecological and engineering research in mixed crops through more extensive experimental designs.

The PhD study was completed at Department of Electrical and Computer Engineering, Faculty of Technical Sciences, Aarhus University.

This summary was prepared by the PhD student.

Time: Thursday 8 April 2021 at 13:00
Place: The PhD defence will be held online. To receive a link to the event, please send an e-mail to Henrik Karstoft, hka@ece.au.dk.
Title of PhD thesis: Mixed Crop Mapping by Species Discrimination using Data Driven Computer Vision and Simulated Data
Contact information: Søren Kelstrup Skovsen, e-mail: ssk@ece.au.dk, tel.: +45 32023 6663
Members of the assessment committee:
Professor Chris McCool, Department of Agricultural Robotics and Engineering, Bonn University, Germany
Professor Thomas B. Moeslund, Department of Architecture, Design and Media Technology, Aalborg University, Denmark
Associate Professor Stefan Hallerstede (chairman), Department of Electrical and Computer Engineering, Aarhus University, Denmark
Main supervisor:
Professor Henrik Karstoft, Department of Electrical and Computer Engineering, Aarhus University, Denmark
Co-supervisor:
Senior Researcher Rasmus Nyholm Jørgensen, Department of Electrical and Computer Engineering, Aarhus University, Denmark
Language: The PhD dissertation will be defended in English

The defence is public.

Due to the coronavirus situation, the PhD thesis will not be available for reading at the offices of the PhD administration. Instead, interested parties may send an e-mail to gradschool.tech@au.dk to receive a digital copy of the PhD thesis. Please note, it may in certain cases be necessary to make an individual arrangement to read the PhD thesis.

PhD defence
17427 / i43