Estimating grassland biomass using multispectral UAV imagery, DTM and a random forest algorithm

Sutter, Michael; Aebischer, Philippe; Reidy, Beat (17 May 2021). Estimating grassland biomass using multispectral UAV imagery, DTM and a random forest algorithm In: Sensing – New Insights into Grassland Science and Practice : Proceedings of the 21st Symposium of the European Grassland Federation. Grassland Science in Europe: Vol. 26 (pp. 71-73). Zurich: European Grassland Federation EGF

[img] Text
EGF2021_Proceedings.pdf - Published Version
Restricted to registered users only
Available under License Publisher holds Copyright.

Download (517kB) | Request a copy

A prerequisite for efficient pasture management is the regular estimation of the dry matter yield (DMY) by means of a rising plate meter (RPM). With the latest generation of unmanned aerial vehicles (UAV) equipped with a real-time kinematic (RTK) positioning system and a multispectral camera, it should be possible to measure sward heights and to estimate dry matter yields. To investigate this possibility, we developed an algorithm enabling a digital terrain model to be calculated from the digital surface model of grassland. DMY is estimated using a random forest estimator. Initial estimates at a previously unseen site achieved a root-mean-square error (RMSE) of 332 kg DM ha-1. The results demonstrate that UAVs enable DMY predictions with an accuracy level close to RPM measurements. The underlying algorithm will be further developed and adapted to a wider variety of pasture types and meadows.

Item Type:

Conference or Workshop Item (Paper)

Division/Institute:

School of Agricultural, Forest and Food Sciences HAFL
School of Agricultural, Forest and Food Sciences HAFL > Agriculture
School of Agricultural, Forest and Food Sciences HAFL > Agriculture > Grasslands and Ruminant Production Systems

Name:

Sutter, Michael;
Aebischer, Philippe and
Reidy, Beat0000-0002-8619-0209

Subjects:

Q Science > QC Physics
S Agriculture > S Agriculture (General)
S Agriculture > SB Plant culture

ISBN:

978-3-00-068789-1

Series:

Grassland Science in Europe

Publisher:

European Grassland Federation EGF

Language:

English

Submitter:

Michael Sutter

Date Deposited:

28 Jun 2021 11:37

Last Modified:

28 Jun 2021 11:37

Uncontrolled Keywords:

grassland, machine learning, random forest, NDVI, remote sensing, dry matter yield

ARBOR DOI:

10.24451/arbor.15032

URI:

https://arbor.bfh.ch/id/eprint/15032

Actions (login required)

View Item View Item
Provide Feedback