Publications - Rene Gislum

Three-dimensional area coverage planning model for robotic application

Vahdanjoo, M., Gislum, R., Grøn Sørensen, C. — Thu, 01 Feb 2024 11:50:52 +0100

Due to the increasing global population, the supply of nonrenewable resources such as fossil fuels are limited.
Therefore, it is urgent to implement measures to control and reduce energy consumption in agriculture and other
sectors. Improving energy efficiency and reducing pollution in agriculture can bring both financial and environmental benefits. By optimizing in-field coverage planning of agricultural vehicles, fuel consumption can be
reduced. Recent developments in automated coverage planning algorithms, both two-dimensional and three dimensional, aim to optimize and automate operations. The elevation property of agricultural fields has a significant impact on coverage path planning, which means that 3D coverage path planning can be highly advantageous for field operations. In this study, a 3D coverage path planning model was developed that effectively
determines the optimal driving direction in a field based on energy consumption criteria. A digital elevation
model (DEM) was utilized to construct an analytical model that provides improved characterization of the
topographical features of three-dimensional landscapes. The aims of this paper encompass the formulation and
execution of a 3D coverage planning strategy for material input operations that aims to reduce energy consumption, reduce overall traveled distance, generate a map to display variable application rates for fertilizers
using remote sensing techniques, and the evaluation of soil compaction risks in an agricultural field. The results
of this study demonstrate the effectiveness of our optimization model. In the first sample field, the model was
able to generate a solution that reduced the required energy by up to 9.5%, while in the second sample field, it
achieved a reduction of up to 15.6% in required energy. Overall, this study highlights the importance of
considering multiple objectives in coverage path planning and demonstrates how our proposed optimization
model can contribute to significant energy savings and reduced environmental impact in agricultural fields.

Estimating Fuel Consumption of an Agricultural Robot by Applying Machine Learning Techniques during Seeding Operation

Vahdanjoo, M., Gislum, R., Grøn Sørensen, C. — Fri, 01 Mar 2024 11:50:52 +0100

The integration of agricultural robots in precision farming plays a pivotal role in tackling the pressing demands of minimizing energy usage, enhancing productivity, and maximizing crop yield to meet the needs of an expanding global population and depleting non-renewable resources. Evaluating the energy expenditure is vital when assessing agricultural machinery systems. Through the reduction of fuel consumption, operational costs can be curtailed while simultaneously minimizing the overall environmental footprint left by these machines. Accurately calculating fuel usage empowers farmers to make well-informed decisions about their farming operations, resulting in more sustainable and productive methods. In this study, the ASABE model was applied to predict the fuel consumption of the studied robot. Results show that the ASABE model can predict the fuel consumption of the robot with an average error equal to 27.5%. Moreover, different machine-learning techniques were applied to develop an effective and novel model for estimating the fuel consumption of an agricultural robot. The proposed GPR model (gaussian process regression) considers four operational features of the studied robot: total operational time, total traveled distance, automatic working distance, and automatic turning distance. The GPR model with four features, considering hyperparameter optimization, showed the best performance (R-squared validation = 0.93, R-squared test = 1.00) among other models. Furthermore, three different ML methods (gradient boosting, random forest, and XGBoost) were considered in this study and compared with the developed GPR model. The results show that the GPR model outperformed the mentioned models. Moreover, the one-way ANOVA test results revealed that the predicted values from the GPR model and observation do not have significantly different means. The results of the sensitivity analysis show that the traveled distance and the total time have a significant correlation with the fuel consumption of the studied robot.

Betingelse for indsamling af data til brug for omregningsfaktor mellem NDV/NDRE og planteoptag af kvælstof

Wed, 23 Aug 2023 11:50:52 +0200

Variations of seed size and seed number per spikelet and their effects on seed germination in hulled oats

Yuan, Z., Gislum, R., Jing, Z., et al. — Fri, 01 Mar 2024 11:50:52 +0100

Seed yield components and ear characteristics are critical parameters for seed yield improvement of cereal crops. In this study, 23 hulled oat (Avena sativa) cultivars were planted in four experimental sites during 2018–2020. We determined the ear parameters including the number of four spikelet types, the ear length, the seed weight per ear, and the size of the left and the right seed in the double-seed spikelet. In addition, five oat cultivars with the highest percentage of the triple-seed spikelet were selected to evaluate the germination performance of the left, the middle, and the right seed. The results showed that the double-seed spikelet was present in 73% of the ear, occupying the highest proportion than other spikelet types. The negative correlation showed the trade-off between total seed number per ear and average seed weight. According to the pathway analysis, the number of double-seed spikelet had the significantly highest direct effect on the seed weight per ear with the path coefficient of 0.93. In the double-seed spikelet, the length and the thickness both had significantly positive correlations to the seed weight of the left and the right seeds. The middle seed, which had the smallest seed size, had the lowest germination rate, germination index, vigour index, seedling weight, and length than the right and left seed among three seed positions in the triple-seed spikelet. We conclude that the double-seed spikelet number is a key ear characteristic when breeding high-yielding cultivars.

Opdatere trappemodel om dækningsgrad af efterafgrøder ved kontrol i forhold til etableringstidspunkt

Mon, 01 May 2023 11:50:52 +0200

Kamera og kunstig intelligens til vurdering af sygdomstryk i sukkerroer

Mortensen, A. K., Hansen, A. L., Gislum, R. — Sun, 01 Jan 2023 11:50:52 +0100

Formålet med forsøget er at undersøge, om et høj kvalitets farvekamera sammen med kunstig intelligens kan anvendes til automatisk genkendelse af bladsygdomme i sukkerroer.

Operational, Economic, and Environmental Assessment of an Agricultural Robot in Seeding and Weeding Operations

Vahdanjoo, M., Gislum, R., Grøn Sørensen, C. — Wed, 01 Mar 2023 11:50:52 +0100

The development of robotic‐based agricultural machinery systems has significantly increased in recent years. Many autonomous systems have not yet been measured based on sustainability and economic performances, even though automation is regarded as an opportunity to increase safety, dependability, productivity, and efficiency. The operational aspect, economic viability, and environmental impact of replacing conventional machinery with robotized alternatives are the primary focus of this study. The robot considered in this research is designed for extensive field‐work, where PTO and external hydraulics are required. This robot is equipped with two 75 (hp)Kubota diesel engines with a total engine gross power of up to 144 (hp). Both robotic system and
conventional machinery were described, and different scenarios were used to examine various operational and environmental indicators, as well as individual cost elements, considering various field sizes and working widths of implements used in seeding and weeding operations. The findings demonstrate that the robotic system outperforms conventional machinery in terms of operational
efficiency by as much as 9%. However, the effective field capacity comparison reveals that the conventional system has a field capacity that is up to 3.6 times greater than that of the robotic system.
Additionally, the total cost per hour of the robotic system is up to 57% lower than that of the conventional system. The robotic system can save up to 63.3% of fuel during operation, resulting in the same percentage reduction in CO2 emissions as the conventional system, according to a comparison of fuel consumption.

Hvordan kan vi bruge satellitbilleder i frøavlen?

Gislum, R. — Sat, 01 Oct 2022 11:50:52 +0200

Fond lukker efter mangeårig værdifuld støtte til frøavlen

Boelt, B., Gislum, R. — Fri, 01 Apr 2022 11:50:52 +0200

Er det nødvendigt at justere kvælstofmængden til frøafgrøden?

Gislum, R. — Tue, 01 Mar 2022 11:50:52 +0100

A chemometric method for the viability analysis of spinach seeds by near infrared spectroscopy with variable selection using successive projections algorithm

Lakshmanan, M. K., Boelt, B., Gislum, R. — Wed, 01 Feb 2023 11:50:53 +0100

This paper proposes a chemometric method for evaluating the viability of spinach seeds using near infrared (NIR) spectroscopy and successive projections algorithms (SPA). An essential step of the procedure is to apply the SPA to optimize the choice of variables for multivariate classification. Variable selection using SPA has been described as an optimization problem in which a cost function is minimized. Selecting the correct variables makes the chemometric models more complete, precise, accurate, and less complex. The NIR spectra were processed using the Savitzky-Golay and multiplicative scatter correction techniques. After that, the best wavelength subset was selected using SPA. Different classification techniques are then applied to the dimension-reduced data to determine the seeds’ viability. The results show that the proposed method is less complex compared to existing canonical variance methods (1.7% miscalculation error in the proposed way) and is also easier to implement.

Omregningsfaktor fra NDVI og NDRE til kg N optag/ha

Gislum, R., Holst, N., Mortensen, A. K., et al. — Thu, 01 Sep 2022 11:50:53 +0200

Potential Impact of Learning Management Zones for Site-Specific N-Fertilization: A Case Study for Wheat Crops

Franco, C., Mejía, N., Pedersen, S. M., Gislum, R. — Mon, 13 Jun 2022 11:50:53 +0200

This paper proposes an automatic, machine learning methodology for precision agriculture, 1 aiming at learning management zones that allow a more efficient and sustainable use of fertilizer. 2 In particular, the methodology consists in clustering remote sensing data and estimating the impact 3 of decision making based on the extracted knowledge. A case study is developed on experimental 4 data coming from winter wheat (Triticum aestivum) crops receiving site-specific fertilization. A first 5 approximation to the data allows measuring the effects of the fertilization treatments on yield and 6 quality of the crops. After verifying the significance of such effects, clustering analysis is applied 7 on sensor readings on vegetation and soil electric conductivity in order to automatically learn the 8 best configuration of zones for differentiated treatment. The complete methodology for identifying 9 management zones from vegetation and soil sensing is validated for two experimental sites at 10 Denmark, estimating its potential impact for decision making on site-specific N-fertilization.

Endelig rapport i pilotprojekt om biomasse

Wed, 01 Jun 2022 11:50:53 +0200

Roboter im Ackerbau - Entwicklung aus der Nische

Gislum, R., Jørgensen, R., Mortensen, A. K., Mikkelsen, B. — Sat, 01 Jan 2022 11:50:53 +0100

Predicting heavy metal contents by applying machine learning approaches and environmental covariates in west of Iran

Azizi, K., Ayoubi, S., Nabiollahi, K., Garosi, Y., Gislum, R. — Tue, 01 Feb 2022 11:50:53 +0100

The cuurent study was performed to predict spatial distribution of some heavy metals (Ni, Fe, Cu, Mn) in western Iran, using environmental covariates and applying two machine learning methods comprised Random forest (RF), and Cubist. In this respect, a combination of different input environmental variables (remote sensing data, topographic attributes, thematic maps and soil properties) were used in modeling under four scenarios (I: remote sensing data (RS); II: RS + topographic attributes resulted from digital elevation model (DEM); III: RS + topographic attributes + thematic maps; IV: RS + topographic attributes + thematic maps +soil properties). The maps of Euclidean distance from mines and roads as well as the geology map have been used as thematic maps. A total of 346 soil samples were taken using stratified random sampling from the surface layers (0–20 cm depth) of the studied area and selected heavy metals (Ni, Fe, Cu, Mn), and soil properties were measured in the laboratory. RF and Cubist models were used to predict soil heavy metals in four scenarios. The results indicated that the best prediction accuracy was achieved for the fourth scenario (IV) when all input variables were combined to predict selected heavy metals. Moreover, two models showed different capability for various metals. According to our results, the random forest model had a high accuracy in predicting Ni (R² = 0.67) and Cu (R² = 0.60), In contrast, the Cubist model had a higher accuracy in predicting Mn (R² = 0.55). For predicting Fe, both models provided a similar accuracy (R² = 0.73). This study proved the high capability of machine learning methods to use easily available environmental data to predict studied heavy metals in the large scale that are essential for decision making in sustainable management in agricultural and environmental concerns.

Kommentarer til oplæg til brug på workshop i pilotprojekt biomasse og efterafgrøder

Gislum, R., Hansen, E. M., Thomsen, I. K. — Mon, 15 Nov 2021 11:50:53 +0100

Assessment of grass lodging using texture and canopy height distribution features derived from UAV visual-band images

Tan, S., Mortensen, A. K., Ma, X., Boelt, B., Gislum, R. — Fri, 01 Oct 2021 11:50:53 +0200

Lodging is a major limiting factor for the yield, quality and harvesting efficiency of selected crops worldwide. This study presents an efficient, robust and non-destructive assessment of lodging severity for four different grasses for seed production, using images collected by an unoccupied aerial vehicle (UAV) in two field plot experiments across five growing seasons. Canopy texture and height related features were extracted from individual plot images and evaluated for estimating lodging severity. Histograms of oriented gradients (HOG) were used as texture features, and three canopy height distributions features (CHV1, CHV2 and CHV3) were proposed. Each canopy height distribution feature divides the plots into subplots and estimates the average height of each subplot. CHV1 concatenates average height of the subplots into its feature, while CHV2 concatenates the difference in average height between all subplots, and CHV3 concatenates the difference in average height between adjacent subplots. The plots were classified using support vector machines into three categories according to the lodging severity. The results showed that the HOG and height distribution features can be used for grading lodging severity in UAV images with high accuracy (71.9% and 79.1%, respectively). However, the HOG features showed a negative relationship to the ground sample distance (GSD), while the CHV1 had a constant accuracy across the GSDs. Combination of the two features did not significantly improve the classification accuracy. The present results have potential to generate lodging severity maps for application in precision farming and thereby to increase grass seed yield and harvest efficiency at farm scale. It should be noted that results and methods from the current study might not be transferred to other crops due to crop specific lodging characteristics and effect of yields.

Data og teknologi i greenkeeperarbejdet

Mon, 10 May 2021 11:50:53 +0200

Greenkeepere i Danmark har stadigt flere muligheder for at bruge data og ny teknologi i arbejdet med at pleje golfbanen. Men hvordan og hvor meget skal det bruges. Og hvad kan det bruges til? Vi tager en diskussion med Réne Gislum, der er lektor ved Århus Universitet, Institut for Agroøkonomi, Thomas Pihl, der er chefgreenkeeper i Furesø Golfklub og Torben Kastrup Petersen, banechef i Dansk Golf Union.

The Use of Remote Sensing to Determine Nitrogen Status in Perennial Ryegrass (Lolium perenne L.) for Seed Production

Gislum, R., Thomopoulos, S., Gyldengren, J. G., Mortensen, A. K., Boelt, B. — Tue, 01 Jun 2021 11:50:53 +0200

Sufficient nitrogen (N) supply is decisive to achieve high grass seed yields while overfertilization will lead to negative environmental impact. From the literature, estimation of N rates taking into account the crop’s N status and its yield potential, seems promising for attaining high yields and averting adverse environmental impacts. This study aimed at an evaluation of remote sensing to predict final seed yield, N traits of the grass seed crop and the usability of nitrogen nutrition index (NNI) to measure additional N requirement. It included four years’ data and eight N application rates and strategies. Several reflectance measurements were made and used for the calculation of 18 vegetation indices. The predictions were made using partial least square regression and support vector machine. Three different yield responses to N fertilization were noted; one with linear response, one with optimum economic nitrogen (EON) at ~188 kg N ha−1, and one with EON at ~138 kg N ha−1. We conclude that although it is possible to make in-season predictions of NNI, it does not always portray the differences in yield potential; thus, it is challenging to utilize it to optimize N application.

Analyser i pilotprojekt om biomasse på baggrund af data fra forsøgsår 2020

Sat, 01 May 2021 11:50:53 +0200

The Use of Multispectral Imaging and Single Seed and Bulk Near-Infrared Spectroscopy to Characterize Seed Covering Structures:

Mortensen, A. K., Gislum, R., Jørgensen, J. R., Boelt, B. — Thu, 01 Apr 2021 11:50:53 +0200

The objective of seed testing is to provide high-quality seeds in terms of high varietal identity and purity, germination capacity, and seed health. Across the seed industry, it is widely acknowledged that quality assessment needs an upgrade and improvement by inclusion of faster and more cost-effective techniques. Consequently, there is a need to develop and apply new techniques alongside the classical testing methods, to increase efficiency, reduce analysis time, and meet the needs of stakeholders in seed testing. Multispectral imaging (MSI) and near-infrared spectroscopy (NIRS) are both quick and non-destructive methods that attract attention in seed research and in the seed industry. This review addresses the potential benefits and challenges of using MSI and NIRS for seed testing with a comprehensive focus on applications in physical and physiological seed quality as well as seed health.

Effekten af præcisionsjordbrug og omregning til efterafgrøder

Tue, 01 Dec 2020 11:50:53 +0100

A simulation of variable rate nitrogen application in winter wheat with soil and sensor information - An economic feasibility study

Sun, 01 Aug 2021 11:50:53 +0200

Variable rate N management strategies are often based on information about soil texture or from canopy sensors, mounted on ground-based vehicles or satellites. However, disentangling the effect of each information type on N management strategy with experimental studies is often difficult, as
results are only valid for the specific experimental conditions as well as the weather conditions for
specific years. An alternative for this is to use deterministic crop growth models. This study examines
whether ‘static’ soil profile information or ‘dynamic’ canopy sensor type information provide the
best basis for decision making concerning N-application at the subfield level. The DAISY model was
used for simulating crop growth on six soil profiles found in a heterogeneous loamy sand field and a
five-year crop rotation. A range of management descriptions and simulations were made using
5x500 years of synthetic weather data with each crop in the rotation set at the first year of the five
parallel simulations. Simulated growth was used as a proxy for a ‘dynamic’ canopy sensor based
information system. The net revenue was then calculated for a range of price relations between
fertilizer (model input) and wheat yield (model output), including wheat price adjustments according
to protein content. Based on regressions and backward induction analysis, the N application that
maximizes the expected net revenue were estimated for four information cases; Case 1) Uniform
application, assuming no prior information, Case 2) application based only on soil type , Case 3)
application based on canopy sensor information only and Case 4) application based on combined soil
and canopy sensor information. Findings from this study indicate that decisions with soil information
alone provide an annual net revenue (without considering cost of collecting information and variable
rate technology) of variable rate application (VRA) between 3.88 and 13.30 € ha-1 across price and
soil variation. This net revenue is approximately doubled with applications based on only canopy
sensor information and again this is approximately doubled with applications based on both soil and
canopy sensor information. The results may guide developers to decide on what type of information
should be included in their decision support systems.

Så meget af det tilførte N fjerner frøafgrøden

Gislum, R. — Thu, 01 Oct 2020 11:50:53 +0200

Revurdering af Landbrugsstyrelsens kontrolmodel 2019 (trappemodellen)

Wed, 02 Sep 2020 11:50:53 +0200

Virkemidler til reduktion af kvælstofbelastningen af vandmiljøet

Eriksen, J., Thomsen, I. K., Hoffmann, C. C., et al. — Wed, 08 Jul 2020 11:50:53 +0200

Linket i højre side henviser til DCA rapporten. Dokumentet over linket er selve leveringen til ministeriet med følgebrev.

I DCA-rapporten er der foretaget mindre sproglige og redaktionelle rettelser i forhold til versionen, som blev leveriet til ministeriet den 8. juli 2020. Derudover er der i kapitlet Afgasning af husdyrgødning kombineret med højere udnyttelseskrav for afgasset gødning sket en justering af beregningsgrundlaget af hensyn til overensstemmelse med seneste rådgivningsnotat på området. Ændringen påvirker ikke den kvalitative vurdering af klimaeffekten i kataloget. I Bilag 3, Klimagasser, er der desuden tilføjet et forklarende afsnit vedr. den indirekte N2O-emissionsfaktor relateret til nitratudvaskning.

Field scale agronomic and environmental consequences of overlapping Nfertilizer application by disc spreaders

Gyldengren, J. G., Greve, M. B., Skou-Nielsen, N., Olesen, J. E., Gislum, R. — Tue, 01 Sep 2020 11:50:53 +0200

Application of mineral nitrogen (N) fertilizer is not uniform at the field scale when applied with disc spreaders. Some of the distribution variation is due to potentially predictable mechanisms, but the variation is also influenced by stochastic events and multiple interactions, making each combination of field, spreader and application event a unique case. This may be one of the reasons why the agronomic, economic and environmental consequences of uneven fertilizer distribution at the field scale remains poorly understood and insufficiently quantified. Previous studies have addressed the problem statistically or empirically in regularly shaped experimental plots or a subsection of a field. While these efforts provide valuable contributions to our understanding, they neglect critical areas of the field with regards to fertilizer application: Wedges, interfaces between the in-field and the headland and in-field obstacles. In this modelling study, we present an attempt to describe the variation in fertilizer application at the field scale and assess the consequences on winter wheat grain yields, grain N yields and N losses to the environment. By combining GIS and agroecosystem modelling, we assess yield and environmental effects of two spreaders with different working widths (24 and 48 m) in four field polygons selected to represent a relevant span with regard to size and geometry for Danish conditions. The effects of the combinations were assessed for two soils, a coarse sandy soil and a sandy loam. Both accuracy (average N input rate relative to target) and precision (evenness of distribution) was found to decrease in small (4−6 ha) and geometrically irregular fields compared to large (40−50 ha) and regular fields. Increasing the working width from 24 to 48 m increased the variation in small fields, but not in large. Grain yields were negatively affected by distribution variation, while there was a poor correlation with average applied N rate. In contrast, grain N yields were insensitive to distribution variation, but showed strong correlation with average N input rate. N leaching was affected by both the amount and distribution of applied N. Across all field x spreader x soil scenarios, field scale grain yields decreased between 0–877 kg DM/ha and annual N leaching increased by up to 9 kg N/ha when fertilizer application was simulated with a spreader rather than even application. In general, the agronomic results from the large and regular field resembled the results from previous studies much better than results from the smaller and/or irregular fields. If not accounting for the impact of field size and shape distribution in the landscape, the effects of uneven N application, and thereby the potential gains by improving spreader performance, may therefore be underestimated.

Effects of winter wheat N status on assimilate and N partitioning in the mechanistic agroecosystem model DAISY

Tue, 01 Dec 2020 11:50:53 +0100

Nitrogen (N) management in modern farming needs to balance the interests of yield quantity and quality with environmental impact of reactive N lost to the atmosphere and aquatic environments. Mechanistic agroecosystem models are useful tools to analyse the combined effects of management options and natural conditions, including soil fertility and climate, site‐specific optimal N application rates and environmental impact. An important component of the system description is the crop module, including the responses to N status and N stress. To improve the description of crop growth response to N status in the DAISY model system, we implemented an empirical model for N status effects in winter wheat on the partitioning of assimilated dry matter and N between leaves and stems, originally described by Ratjen and Kage (Journal of Agronomy and Crop Science , 2016, 202, 576s). To our knowledge, this mechanism has not been included in any of the most widespread and commonly used mechanistic agroecosystem models. We tested and compared the model performance with and without the new N status response against data from a two‐season winter wheat experiment in Denmark, where crop growth and partitioning of dry matter and N were measured. Implementation of the new N status response function improved model performance and ensured a more robust crop growth description, especially in scenarios with periods of low crop N status or N stress. The mechanism mimics an adaptation strategy, where the crop balances N stress and growth potential by dynamically adjusting the leaf‐stem ratio and thereby N demand. This behaviour opens a discussion regarding the empirical concept of a biomass driven critical N curve.

Expression of starch-binding factor CBM20 in barley plastids controls the number of starch granules and the level of CO2 fixation

Zhong, Y., Sagnelli, D., Topbjerg, H. B., et al. — Wed, 01 Jan 2020 11:50:53 +0100

The biosynthesis of starch granules in plant plastids is coordinated by the orchestrated action of transferases, hydrolases, and dikinases. These enzymes either contain starch-binding domain(s) themselves, or are dependent on direct interactions with co-factors containing starch-binding domains. As a means to competitively interfere with existing starch-protein interactions, we expressed the protein module Carbohydrate-Binding Motif 20 (CBM20), which has a very high affinity for starch, ectopically in barley plastids. This interference resulted in an increase in the number of starch granules in chloroplasts and in formation of compound starch granules in grain amyloplasts, which is unusual for barley. More importantly, we observed a photosystem-independent inhibition of CO2 fixation, with a subsequent reduced growth rate and lower accumulation of carbohydrates with effects throughout the metabolome, including lower accumulation of transient leaf starch. Our results demonstrate the importance of endogenous starch-protein interactions for controlling starch granule morphology and number, and plant growth, as substantiated by a metabolic link between starch-protein interactions and control of CO2 fixation in chloroplasts.

Changes in spatiotemporal protein and amino acid gradients in wheat caryopsis after N-topdressing

Zhong, Y., Vidkjær, N. H., Massange-Sanchez, J. A., et al. — Wed, 01 Jan 2020 11:50:53 +0100

Wheat grain nitrogen content displays large variations within different pearling fractions of grains because of radial gradients in the protein content. We identified how spatiotemporal mechanisms regulate this. The protein gradients emerged clearly at 19 days after anthesis, with the highest N content in aleurone and seed coat, followed by outer endosperm, whereas the lowest was in middle and inner endosperm. Laser microdissection, qRT-PCR and LC–MS were used to dissect tissue from aleurone, outer endosperm, middle endosperm, inner endosperm and transfer cells, measure gene expression and levels of free and protein-bound amino acids, respectively. The results showed that different FAA transportation pathways worked in parallel during grain filling stage while the grain protein gradient did not follow spatial expression of storage proteins. Additionally, two nitrogen (N) topdressing timings were conducted, either at the emergence of top third leaf (standard timing) or top first leaf (delayed timing), finding that delayed N topdressing enhanced both amino acids supply and protein synthesis capacity. The results provide insight into protein synthesis and amino acid transport pathways in endosperm and suggest targets for the enhancement of specialty pearled wheat with higher quality.

Bemærkninger til krav for indberetning af data i pilotprojekt om biomasse

Gislum, R., Thomsen, I. K. — Thu, 19 Sep 2019 11:50:53 +0200

Drone dataflow - a MATLAB toolbox for extracting plots from images captured by a UAV

Mortensen, A. K., Laursen, M. S., Nyholm Jørgensen, R., Gislum, R. — Tue, 01 Jan 2019 11:50:53 +0100

The use of unmanned aerial vehicles (UAVs) with mounted cameras has seen a rapid increase in popularity within precision agriculture research for surveying fields. Typically, the collected images are stitched into georeferenced maps using photogrammetry software. This paper proposes a MATLAB toolbox for easing the workflow. The toolbox consists of two modules: (1) detection of ground control points in images collected by UAVs; and (2) extraction of regions-of-interest from georeferenced maps generated from the images. Use of the toolbox is exemplified on selected plots from a plot experiment throughout a growing season.

The Oil Radish Growth Dataset

Fri, 07 Jun 2019 11:50:53 +0200

The Oil Radish Growth Dataset for Semantic Segmentation and Yield Estimation

Mortensen, A. K., Skovsen, S., Karstoft, H., Gislum, R. — Mon, 17 Jun 2019 11:50:53 +0200

The Oil Radish Growth Dataset for Semantic Segmentation and Yield Estimation

Mortensen, A. K., Skovsen, S., Karstoft, H., Gislum, R. — Tue, 01 Jan 2019 11:50:53 +0100

Data sharing in research is important in order to reproduce results, develop global models, and benchmark methods. This paper presents a dataset containing image and field data from a field plot experiment with oil radish (Raphanus sativus L. var oleiformis) as catch crop after spring barley. The field data consists of fresh weight, dry weight, Carbon content and Nitrogen content from multiple weekly plant samples collected from the plots. The image data consists of images collected weekly prior to the plant samples. A subset of the images corresponding to the plant sampling areas have been annotated pixelwise. In addition to the image and field data, weather data from the growing period is also included in the dataset. The dataset is accompanied by two challenges: 1) semantic segmentation of crops and 2) oil radish yield estimation. The former challenge focuses on data image, while the latter focuses on the field data. Baseline methods and results are provided for both challenges.

Overordnet vurdering af risiko for merudvaskning i pilotprojektordning om biomasse

Thomsen, I. K., Gislum, R., Hansen, E. M., Olesen, J. E. — Wed, 03 Jul 2019 11:50:53 +0200

Beskrivelse af setup for vidensindsamling og måling af NDVI-værdi i pilotprojekt om biomasse

Gislum, R., Nyholm Jørgensen, R., Thomsen, I. K., Hansen, E. M., Olesen, J. E. — Mon, 08 Jul 2019 11:50:53 +0200

Organic herbage seed production

Boelt, B., Topbjerg, H. B., Jing, S., Gislum, R. — Tue, 01 Jan 2019 11:50:53 +0100

Organic farmers are supposed to use organically produced seed. For more than 20 years, organic cool season grass and clover seed has been produced in Denmark. For the past 10 years, the average total production area has been approximately 3,000 ha, with the major species in production being perennial ryegrass (Lolium perenne L.) and white clover (Trifolium repens L.). Organic seed crops are undersown in a spring barley cover crop using different establishment techniques. Perennial ryegrass seed is sometimes mixed with and sown with the spring barley and can also be established after a mechanical weeding of the spring barley. Obtaining a solid and uniform establishment of the seed crop is considered a very important weed control strategy, which in organic grass seed production also consists of a relatively high nitrogen application. In the seed production year, various new techniques for mechanical weeding are being tested in experiments at Aarhus University and in grower fields. One of the major challenges in organic white clover seed production is insect pests (Apion spp. and Hypera spp.). Earlier experiments have shown seed yield decreases of 40% in untreated control plots. In years with high incidence of insect pests, the seed grower might choose a late defoliation of the white clover seed crop to reduce the damaging effects of the insects. However, a recent study has shown that biological control using a parasitoid might be promising as a longterm pest management strategy. Over the past two decades, organic seed production practices have been established and, interestingly, are now being tested and adapted in conventional herbage seed production.

Digitalization in herbage grass seed production and research: Opportunities and challenges

Gislum, R., Gyldengren, J. G., Boelt, B. — Sun, 12 May 2019 11:50:53 +0200

Digital herbage seed production and research offer new possibilities but also new challenges. Technological, software, and algorithm development proceeds extremely rapidly, while the development of decision support systems and related agronomic and biological interpretation lags. These challenges must not prevent us from exploring new technology and methods in herbage seed production and research. We must recognize, however, that increasing the utilization of applied and available nutrients through the use of sensors, local weather data, and predictions, in combination with application algorithms, may not necessarily increase farmer revenue or reduce negative environmental impacts. Advanced field phenotyping holds great potential and is receiving a great deal of attention, based on publication frequency, and breeders will gain new knowledge from these technologies and methods. The use of advanced phenotyping in combination with crop models is an interesting area that undoubtedly will increase our basic knowledge of plant development and performance, which is necessary in order to better understand the interactions between nutrient utilization, optimal application timing for plant growth regulation, weed control, and other field operations. The conclusion is that digital herbage seed production and research are still in the early development stage.

Hvor præcist kan vi ramme frøafgrødens kvælstofbehov?

Gislum, R. — Fri, 01 Mar 2019 11:50:53 +0100

Præcis gødskning: Forskerne i Flakkebjerg arbejder på at bestemme kvælstofbehovet i frøgræs og bestemme mængden af biomasse ved brug af billeder.

Fastsættelse af NDVI-krav i pilotprojektet om biomasse

Gislum, R., Thomsen, I. K., Hansen, E. M., Olesen, J. E. — Wed, 22 May 2019 11:50:53 +0200

Markforsøg med efterafgrøder: Etableringstidspunktets betydning for dækningsgrad

Thomsen, I. K., Hansen, E. M., Larsen, R., Gislum, R. — Tue, 01 Jan 2019 11:50:53 +0100

Importance of seed quality for the fresh cut chain

Deleuran, L. C., Olesen, M. H., Shetty, N., Gislum, R., Boelt, B. — Thu, 30 Aug 2018 11:50:53 +0200

Seed is the most fundamental input in vegetable production on which the effectiveness of other inputs and outputs depends. If the seed is of poor quality, the use of other inputs is less successful and will occasionally be irrelevant. A definition of seed quality will depend upon the use for which the seed is intended. Viability, vigor, uniformity, purity and health of the seeds are quality parameters important for all production systems. The ideal situation for both fresh produce and fresh cut would be seeds that germinate uniformly at a high percentage and subsequently grow relatively fast to the desired size, thus, resulting in leaves of improved quality. It is likely that such leaves will be more able to withstand the rigorous processing that includes harvest, transportation, washing, sanitization, de-watering and packaging. This ideal situation would both be determined by the mentioned properties of the seeds and to a large extend the cultivar of the seeds. The quality of a vegetable product is defined by the interaction of these factors, including environmental factors as well as by management and technology. Novel, fast and non-destructive technologies for seed testing allow for new sorting possibilities of a seed lot. At present, new initiatives using multispectral imaging and single seed NIR spectroscopy is being developed at AU for description of seed lots. Preliminary results on investigations of pak choi seeds are presented.

Manual geo‑rectification to improve the spatial accuracy of ortho‑mosaics based on images from consumer‑grade unmanned aerial vehicles (UAVs)

Tue, 01 Jan 2019 11:50:53 +0100

The global spatial accuracy of ortho-mosaics based on images from consumer-grade unmanned aerial vehicles (UAVs) is relatively low. The use of ground control points (GCPs) improves the accuracy but it requires RTK-GNSS (global navigation satellite system) measurements in the field. The objective of this study was to evaluate manual geo-rectification after ortho-mosaicking as a cost-effective alternative to GCPs or UAVs equipped with RTK-GNSS. Google Earth images and A-B lines used for tractor auto-steering were used for manual geo-rectification with a free geographic information system (GIS) and remote-sensing (RS) software (QGIS). Two different photogrammetry software (Agisoft Photoscan and Pix4DMapper) were used for ortho-mosaicking. Images were captured in three fields at two altitudes (40 and 80 m) with a multi-rotor camera drone equipped with a GNSS without ground reference (Phantom 4). The results showed that flight altitude and photogrammetry software had no impact on the spatial accuracy and that manual geo-rectification significantly improved the spatial accuracy. Root mean squares (RMSEs) for ortho-mosaics without geo-rectification was in the range of 4 to 28 m and the range was 0.6 to 2.5 m and 0.5 to 1.1 m for geo-rectification based on Google Earth images and A-B-lines, respectively. RTK-GNSS tagged UAV images and the use of GCPs gave average RMSEs of less than 0.1 m. It was concluded that manual geo-rectification offers a feasible alternative to GCPs and UAVs with RTK-GNSS when spatial accuracy of about 1 m is acceptable.

Predicting Dry Matter Composition of Grass Clover Leys Using Data Simulation and Camera-based Segmentation of Field Canopies into White Clover, Red Clover, Grass and Weeds

Fri, 01 Jun 2018 11:50:53 +0200

Estimating the nitrogen nutrition index in grass seed crops using a UAV-mounted multispectralcamera

Wang, H., Mortensen, A. K., Mao, P., Boelt, B., Gislum, R. — Tue, 01 Jan 2019 11:50:53 +0100

The application of adequate nitrogen (N) fertilizers to grass seed crops is important to achieve high seed yield. Application of N will inevitably result in over-fertilization on some fields and, concomitantly, an increased risk of adverse environmental impacts, such as ground- and/or surface-water contamination. This study was designed to estimate the N status of two grass seed crops: red fescue (Festuca rubra L.) and perennial ryegrass (Lolium perenne L.) using images captured with an unmanned aerial vehicle (UAV) mounted multispectral camera. Two types of UAV, a fixed-wing UAV and a multi-rotor UAV, operating at two different heights and mounted with the same multispectral camera, were used in different field experiments at the same location in Denmark in the period from 432 to 861 growing degree-days. Seven vegetation indices, calculated from multispectral images with four bands: red, green, red edge and near infrared (NIR), were evaluated for their relationship to dry matter (DM), N concentration, N uptake and N nutrition index (NNI). The results showed a better prediction of N concentration, N uptake and NNI, than DM using vegetation indices. Furthermore, among all vegetation indices, two red-edge-based indices, normalized difference red edge (NDRE) and red edge chlorophyll index (CIRE), performed best in estimating N concentration (R2 = 0.69–0.88), N uptake (R2 = 0.41–0.84) and NNI (R2 = 0.47–0.86). In addition, there was no effect from the choice of UAV, and thereby flight height, on the estimation of NNI. The choice of UAV type therefore seems not to influence the possibility of diagnosing N status in grass seed crops. We conclude that it is possible to estimate NNI based on multispectral images from drone-mounted cameras, and the method could guide farmers as to whether they should apply additional N to the field. We also conclude that further research should focus on estimating the quantity of N to apply and on further developing the method to include more grass species.

Valg af biomasseindeks i pilotprojekt om efterafgrøder/biomasse

Gislum, R., Nyholm Jørgensen, R. — Fri, 09 Nov 2018 11:50:53 +0100

Segmentation of lettuce in coloured 3D point clouds for fresh weight estimation

Mortensen, A. K., Bender, A., Whelan, B., et al. — Thu, 01 Nov 2018 11:50:53 +0100

Monitoring the health and yield of crops during production is an important, but labour intensive component of commercial agriculture, especially in high value crop such as lettuce. This article proposes a novel method for segmenting lettuce in coloured 3D point clouds and estimating the fresh weight. The proposed segmentation method operates by clustering points into leaves and then evaluating their affiliation to a lettuce of interest. From the segmented lettuce point clouds, the volume, surface area, leaf cover area and height predictors are extracted and correlated to the fresh weight. The proposed segmentation and yield estimation methods are evaluated on Cos and Iceberg lettuce point clouds generated from images collected by an agricultural robot in an outdoor field experiment. The results demonstrate that the proposed segmentation method is able to successfully isolate lettuce (F1-score = 0.88–0.91). Analysis of the segmented lettuce models show that the calculated surface areas correlate strongly with measured fresh weight (R^2 = 0.84–0.94). Not only does this validate the segmentation method, it allows an accurate estimate of the lettuce fresh weight (RMSE = 27–50 g) to be produced non-destructively.

Betragtninger vedrørende miljøeffektberegninger af pilotprojektordning om præcisionslandbrug

Nørremark, M., Thomsen, I. K., Blicher-Mathiesen, G., et al. — Thu, 20 Sep 2018 11:50:53 +0200