Sugar beet phenotyping in breeding trials using UAV
Sugarbeet field phenotyping with high accuracy and reliability for assessment of resistant sugarbeet to biotic stresses is required but lacking. The small dimensions of breeding trial plots and the need for frequent revisits hamper the use of airborne images, especially with UAVs, able to deliver assessments at a very high spatial resolution pixel size with a very flexible temporal resolution. High-resolution hyperspectral remote sensing imagery is described to be appropriate to produce optical indices related to the crops health status.
This project aims at applying the UAVs potential for the quantitative assessment of a specific plant disease within breeding trial plots.
Description of tasks
Disease symptoms will be assessed during two successive years through several specific artificially inoculated breeding trials. Inoculated trials will be compared to non-inoculated ones in a replicated plots experimental design. Selected varieties will show a gradient of sensitivity to the disease. In situ visual observations (disease symptoms quantification) will be performed several times after crop inoculation. In the meantime UAV images (with embedded RGB, multispectral and hyperspectral imaging sensors) and similar ground-based spectral information will be acquired.
This project is expected to deliver a rapid operator-independent tool to assess in a standardized way diseases at the breeding trial plot level.
The specific expected scientific results are
Evaluation of different sensors, platforms and selection of the most suitable combination for the disease detection and quantification;
Development of a data acquisition and processing methodology for the assessment of the resistance to the disease for sugar beet breeding trials using UAV imaging, and validation of the produced algorithm.
The potential users are breeding companies and institutions, companies involved in fungicide treatment screening (at field level) or in disease/fungicide application forecast, agricultural research centres, crop phenotyping companies, agricultural UAV companies, agricultural remote sensing sector, agricultural sensor and robotics companies.
This project is coordinated by SESVanderHave (SV). SV presents important field trial potentialities and has also large experience in the follow up and assessment of in situ sugar beet breeding field trials. SV will provide its sugar beet field trials and its breeder expertise in variety and disease assessment. These expertizes will be useful to lead the study and to collect field observations on the disease as reference data for the study.
The expertise of VITO for UAV (equipment, flights, pre-processing and processing) and hyperspectral plant stress detection will be used to acquire, process and analyse images over the trials.
The expertise of CRA-W in ground-based multispectral and near infrared (NIR) hyperspectral applications as well as in chemometrics will provide support in the use of hand-held/embedded field devices to validate the acquired spectral information in relation with the visual observations.
Vincke, D. , Durenne, B. , Mingeot, D. , ESCARNOT, E. , Jacquemin, G. , Ben Abdallah, F. , Curnel, Y. , Mauro, S. , Geerts, P. , Lateur, M. , Planchon, V. , Baeten, V. , Vermeulen, P. & Goffart, J.-P. (2017). Plant phenotyping activities at the Walloon Agricultural Research Centre. Poster in: Plant Phenotyping Forum: integrating European plant phenotyping community (EPPN), Tartu - Estonie, 22-24 November 2017.
Vincke, D. , Delalieux, S. , Amand, O. , Ben Abdallah, F. , Raymaeckers, D. , Vermeulen, P. , De Bruyne, E. , Baeten, V. & Goffart, J.-P. (2018). BEETPHEN Sugar Beet Phenotyping in Breeding Trials Using UAV. Lecture in: BEO Day, Beersel, 30 January 2018. Link
Vermeulen, P. , Vincke, D. , Baeten, V. , Jacquemin, G. , Rabier, F. & Goffart, J.-P. (2018). Research on new methods of plant phenotyping at CRA-W. Lecture in: 12th EU-VCU Expert Seminar, Gand-Belgium, 27th June 2018.