Responsive image





UAV-Based Estimation of Grain Yield for Plant Breeding: Applied Strategies for Optimizing the Use of Sensors, Vegetation Indices, Growth Stages, and Machine Learning Algorithms

Prey, L.; Hanemann, A.; Ramgraber, L.; Seidl-Schulz, J.; Noack, P. (2022)

remote sensing 14, 6345 (24), S. 1-17.
DOI: 10.3390/rs14246345


Open Access Peer Reviewed
 

Non-destructive in-season grain yield (GY) prediction would strongly facilitate the selection process in plant breeding but remains challenging for phenologically and morphologically diverse germplasm, notably under high-yielding conditions. In recent years, the application of drones (UAV) for spectral sensing has been established, but data acquisition and data processing have to be further improved with respect to efficiency and reliability. Therefore, this study evaluates the selection of measurement dates, sensors, and spectral parameters, as well as machine learning algorithms. Multispectral and RGB data were collected during all major growth stages in winter wheat trials and tested for GY prediction using six machine-learning algorithms. Trials were conducted in 2020 and 2021 in two locations in the southeast and eastern areas of Germany. In most cases, the milk ripeness stage was the most reliable growth stage for GY prediction from individual measurement dates, but the maximum prediction accuracies differed substantially between drought-affected trials in 2020 (R2 = 0.81 and R2 = 0.68 in both locations, respectively), and the wetter, pathogen-affected conditions in 2021 (R2 = 0.30 and R2 = 0.29). The combination of data from multiple dates improved the prediction (maximum R2 = 0.85, 0.81, 0.61, and 0.44 in the four-year*location combinations, respectively). Among the spectral parameters under investigation, the best RGB-based indices achieved similar predictions as the best multispectral indices, while the differences between algorithms were comparably small. However, support vector machine, together with random forest and gradient boosting machine, performed better than partial least squares, ridge, and multiple linear regression. The results indicate useful GY predictions in sparser canopies, whereas further improvements are required in dense canopies with counteracting effects of pathogens. Efforts for multiple measurements were more rewarding than enhanced spectral information (multispectral versus RGB).

mehr

Technology Perspective

Herlitzius, T.; Noack, P.; Späth, J.; Barth, R.; Wolfert, S.; Bernardi, A....

Handbook Digital Farming, S. 109-189.
DOI: 10.1007/978-3-662-64378-5_3


Peer Reviewed
mehr

Ukraine-Krieg: Fallen jetzt die GPS-Lenksysteme an Traktoren aus?

Hoffmann, L.; Noack, P. (2022)

agrarheute.com, 05.03.2022.


mehr

Identifying operation modes of agricultural vehicles based on GNSS measurements

Poteko, J.; Eder, D.; Noack, P. (2021)

Computers and Electronics in Agriculture 185, 106105.
DOI: 10.1016/j.compag.2021.106105


Peer Reviewed
 

The operation mode determines the optimal settings for different parameters of agricultural vehicles. A classification of road mode and field mode is essential for adapting settings, e.g. the automatic adjustment of tire inflation during transport on road and operational field work. This study focuses on the development and application of algorithms for automatically detecting the operation mode of agricultural vehicles based on GNSS data. The approach is solely based on the parameters speed, COG and derived values such as acceleration, curve radius and angular speed. Known field boundaries and the current position of the vehicle have been neglected to increase the flexibility and applicability of the algorithm. For this purpose the GNSS data were collected with two GNSS receivers differing with respect to model and correction data source (EGNOS and RTK). Speed, time, heading and derived parameters were included in the development of a decision tree based model to classify the operating mode using the rpart package in RStudio. The prediction of operating mode was carried out with the predict package in RStudio. A confusion matrix was introduced to validate the performance of different models. The algorithms derived from the two training datasets (EGNOS and the RTK dataset) show convincing results in the detection of road and field mode. Both algorithms demonstrated an accuracy of more than 90%. The prediction performance was improved when training and validation data were derived from the same dataset (either EGNOS or RTK dataset). The comparison of two algorithms based on EGNOS and RTK data reveal the advantage of models based on RTK data. It is of great importance that the number of wrong decisions regarding the detection of the operating mode on the road are minimized since road safety plays an important role and the potential harm caused by a wrong decision is substantially higher than in the field. The method reveals a large potential for other applications where the operating mode is relevant.

mehr

Wirklich reif für die Praxis? Teil 1: Wie praxistauglich ist die teilflächenspezifische N-Düngung bereits?

Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

Bayrisches Landwirtschaftliches Wochenblatt (18), S. 36-38.



Hier mehr, dort weniger Maiskörner

Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

Bayrisches Landwirtschaftliches Wochenblatt (22), S. 38-40.



Feldgrenzen auf den Zentimeter bestimmen

Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

Bayrisches Landwirtschaftliches Wochenblatt (27), S. 41-42.



In friedlicher Mission

Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

Bayrisches Landwirtschaftliches Wochenblatt (32), S. 38-39.



Der smarte Kuhstall

Fleischmann, A.; Bauer, B.; Braun, K.; Breunig, P.; Meyer, T.; Noack, P.; Saeed, M....

Bayrisches Landwirtschaftliches Wochenblatt (35), S. 41.



Precision-Farming: teilflächenspezifisch Düngen - so starten Sie

Göggerle, T.; Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

agrarheute, 26. Juni 2020.


mehr

Precision-Farming: Warum teilflächenspezifisch wirtschaften?

Göggerle, T.; Wilmes, R.; Bauer, B.; Braun, K.; Breunig, P.; Fleischmann, A.; Meyer, T....

agrarheute, 19. Juni 2020.


mehr

Flach ins Feld statt spritzen

Bleisteiner, N.; Hamberger, S.; Heinz, M.; Steigerwald, T.; Seubert, K.; Bauer, B....

Landwirtschaftliches Wochenblatt - BWagrar Ausgabe 5-2020.



Digitalisierungsprojekt 'Diabek' kommuniziert mit den landwirtschaftlichen Zielgruppen über eine eigene Webseite sowie Social Media-Kanäle

Newsmeldung HSWT, .; Noack, P. (2020)

Forschungs-News HSWT, 10.03.2020.


mehr

Neues Forschungsprojekt der HSWT automatisiert die Verarbeitung drohnengestützt erhobener Bonituren in der Weizenzüchtung

Newsmeldung HSWT, .; Noack, P. (2020)

Forschungs-News HSWT, 13.02.2020.


mehr

Landwirtschaft 4.0 – Nachhaltige Landnutzung und digitale Methoden: ein Widerspruch?

Noack, P.; Rudner, M. (2020)

Politische Studien 490, S. 27-37.


mehr

Entwicklungen der Technik und Auswirkungen

Noack, P. (2019)

Lichtmesstag 2019 der Landmaschinenschule Triesdorf am 01. Dezember 2019, Altes Reithaus in Triesdorf.



Flach bearbeiten statt Glyphosat

Bleisteiner, N.; Hamberger, S.; Bauer, S.; Heinz, M.; Steigerwald, T.; Seubert, C....

Bayerisches Landwirtschaftliches Wochenblatt (42), S. 28-32.



Mehr Bodenschutz für Fruchtbarkeit und Gewinn

Noack, P.; Volk, L.; Poteko, J. (2019)

Johannitag am 30. Juni 2019 in Triesdorf.



Echtzeiterkennung von Feld- und Straßenfahrt auf Basis von GNSS-Messwerten

Noack, P.; Volk, L.; Poteko, J. (2019)

Johannitag am 30. Juni 2019 in Triesdorf.



Digitalisierung und Smart Farming - Bedeutung und Nutzen für die heutige Landwirtschaft

Noack, P. (2019)

Getreidemagazin 25 (6), S. 8-11.


 

Die Digitalisierung und „Smart Farming“ finden in den letzten Jahren starke Beachtung. Dabei sind die Inhalte hinter den Begriffen nicht neu. Gleichzeitig werden teilweise hohe Erwartungen gesteckt: Einsparung von Kosten bzw. Arbeitszeit, Steigerung der Erträge sowie die Verringerung des Energie-einsatzes und negativer Umwelteinwirkungen. Und das alles am besten gleichzeitig.

mehr

Kompetenzzentrum für digitale Agrarwirtschaft

Hochschule Weihenstephan-Triesdorf

Kompetenzzentrum für digitale Agrarwirtschaft
Neuseser Str. 1
91732 Merkendorf

T +49 9826 654-242
koda[at]hswt.de

Betreuung der Publikationsseiten
Gerhard Radlmayr | Laura Hucke
T +49 8161 71-3350, -5384
publikationen.zfw[at]hswt.de