J. Yi: Deep learning for non-invasive diagnosis of nutrient deficiencies in sugar beet using RGB images


Bio Information

Jinhui Yi is a computer science Ph.D. student at the University of Bonn under the supervision of Professor Juergen Gall in the Computer Vision Group (and PhenoRob Cluster). His research interests include domain adaptation and scene understanding.  He obtained his Master’s degree in Electrical Engineering and Information Technology at the Gottfried Wilhelm Leibniz Universität Hannover, Germany. His master thesis focused on deep reinforcement learning for semantic image analysis. Before that, he received his Bachelor’s degree in Electrical Engineering at South China University of Technology, China.


Presentation Abstract

In order to enable timely actions to prevent major losses of crops caused by lack of nutrients and hence increase potential yield throughout the growing season while at same time prevent excess fertilization with detrimental environmental consequences, early, non-invasive and on-site detection of nutrient deficiency is required. Current non-invasive methods to assess the nutrient status of crops deal in most cases with nitrogen (N) deficiency only and optical sensors to diagnose N deficiency such as chlorophyll meters or canopy reflectance sensors do not monitor N but instead measure changes in leaf spectral properties that may or may not be caused by N deficiency.

In this work, we study how well nutrient deficiency symptoms can be recognized in RGB images of sugar beets. To this end, we collected the Deep Nutrient Deficiency for Sugar Beet (DND-SB) dataset, which contains 5,648 images of sugar beets growing on a long-term fertilizer experiment with nutrient deficiency plots comprising N, phosphorous (P), and potassium (K) deficiency as well as omission of liming (Ca), full fertilization and no fertilization at all. We use the dataset to analyse five convolutional neural networks for recognizing nutrient deficiency symptoms and discuss their limitations.

The authors would like to explicitly thank all staff members of the Crop Science Group (INRES, University of Bonn, Germany) who have been running the long-term fertilizer experiment Dikopshof for decades, as well as Angelika Glogau, Lilli Wittmaier, and Adolf Kiener for their help with nutrient analysis.


Video

Ask the Speaker

You can post your questions for the speaker of this talk here until Nov 8. Your questions are visible to all other registered participants of the conference and will be visible together with your username that consists of the beginning of your email address. Participants can “like” your question. Questions with more like are more likely to be discussed during the live Q&A session on November 10, 2020.