Using Drones to Spot Disease and Weed Infestations in Sugar Beet

About the project

An agronomist and drone operator teamed up to find better ways of detecting plant stress in sugar beet and spotting disease and weed infestations with drones.

A sugar beet grower, David, had heard about the potential of drone technology from a friend who had been using drones to provide counts of plant and tree populations. A few weeks before harvest, David, a farmer and agronomist, approached Agremo for advice on the use of drones with sugar beet, specifically about how they could be used to detect plant stress.

Drones can provide plant counts, data on the location of certain weeds and diseases, or identify irrigation problems by identifying areas of water stress. Thus, they represent an exciting opportunity for farmers to improve crop management and allow a more targeted use of inputs.

Using drone imagery combined with Agremo plant stress analysis, David was able to produce a map indicating areas of weed and disease infestation within the crop. David could consider spraying only the affected areas of the field or using VRT (variable-rate technology).

Customer requirements and challenges

David wanted to see exactly how bad the losses from plant stress were. This included finding areas of bare soil, as well as areas of disease and weed infestation, and any other issues which could impact the plant’s health and the field’s performance.

Information about the type and distribution of weeds on the field is crucial for farmers and autonomous agriculture robots. To address this challenge, they can resort to UAV drones. [1].

Weeds are a major concern for sugar beet producers – especially during the first 8 weeks, weeds such as Chenopodium can do significant damage to a growing crop. David needed to know how much of the entire field had been affected by weeds and disease, and which areas to focus his time and treatments on.

Field conditions / Existing processes

David reported that the season had been tough in terms of plant stress and that he had tried different approaches to maintain yields in the face of plant stress in the past. David usually walked across the fields to monitor his crop during the growing season, and whilst he was usually able to detect signs of plant stress, he did not always identify the underlying causes in time to provide effective treatment with an appropriate post-emergence herbicide or fungicide.

One field, however, was causing him particular concern.

“I have 9 fields in total and I’m expecting good yields overall, but from what I can see on the ground, there’s one [field] that I’m less optimistic about.”

How Agremo approached the project

After consultations with the support team, it was decided to perform a plant stress analysis with the Agremo app. This was chosen because plant stress is often an early indication of disease or weed infestation within a crop. The data gathered can give an early indication of problems and allow a decision to be made on a more specific analysis.

Detecting Plant Stress with Drones

Essentially, drone-based plant stress analyses indicate the percentage and the exact location of areas with any kind of stress, e.g. diseases, weed, areas without plants, and so on. In David’s case, the goal was to show him the total stress area of his sugar beet field.

The process is broken down into three stages.

Step 1: Collect the data

First, David hired a local drone operator to map the entire field. Today, agriculture drone mapping is something most drone operators are familiar with or have at least heard of.

Step 2: Generate a map with stitching software

Images taken by the drone need to be converted into a 2D map that the Drone data software can analyze. This was done by uploading to the stitching platform DroneDeploy.

Step 3: Analyze the field

To analyze his field, David uploaded the 2D map to the drone data analysis platform Agremo. The platform currently offers 12 different analyses, from plant counting to different plant health metrics. Agriculture drones do not replace the farmer’s or agronomist’s decision-making process and experience but provide the data to make more informed decisions which can ultimately improve the bottom line and profitability of the crop.

Agremo always approaches crop issues and problems in the same manner that a doctor approaches a patient with an illness. The first step is to carry out a general examination (i.e. a stress analysis), before recommending a more specific examination, be it an X-ray or scan (i.e. a weed or disease analysis).

Design and Plan

Collect data

Analyze data

Deliver and Apply

The process and the solution

What happened in the field

Agremo suggested weed and plant disease analysis to ascertain the exact locations and extent of the problems caused by Cercospora and Chenopodium. Plant disease analyses can be used to identify the precise location and size of problem areas, optimize treatments and preventive measures (fungicide, insecticide, etc. and spot diseases before they affect the current yield goal. Weed analyses can be used to spot weed-infested areas in time and optimize weed control measures by applying the right amount of herbicide on the right spots.

What went as planned

Most forms of plant stress lead to visual changes. Drones can pick up different leaf colors or blank spots on the field, as well as cover ground more quickly than by crop walking. Multispectral sensors on drones can identify patterns that the human eye cannot. The drone view of the sugar beet field exactly shows the extent of the damage.

What was unplanned

The data indicated that almost 50% of the field showed signs of stress. “I was both shocked and amazed when I got the results. I was shocked that the damaged area was so large — more than half of it. But Bosko, the agronomist from Agremo gave me a good tip”, said David. Before the plant stress report, David reported that he had a problem with the fungal disease Cercospora, and with the weed (Chenopodium album), commonly known as lambs quarters. Cercospora beticola is one of the most widespread and damaging sugar beet (Beta vulgaris L.) foliar disease [3]. Competitive weed species such as Chenopodium album can cause up to 80% yield loss in sugar beet [2], and resistance has limited the effectiveness of chemical control methods.

Where there any issues

For a weed and disease analysis, ground truth data was needed. It was carried out using a smartphone to take a picture of a square meter where the weed was present and another where the disease was present. This allows the software to compare areas of the field with and without weed and disease infestation with healthy areas of the crop and provide an accurate picture of the infestation.

Try Agremo Analytics

Free - no credit card required

What were the decisions informed by Agremo analysis

The Agremo analysis allowed David to plan more effective weed and disease control over the season.  Plant stress reports can provide a broad picture of problems in a field, however, they don’t give the specifics on the exact weed or disease but provide insights into the extent of the damage (location and size of the problem areas). Once this is done, you can follow up with a specific weed or disease analysis to provide more details on the levels of infestation and locations within the field to be able to make decisions regarding specific control measures.

Summary ROI on precision agriculture

According to the reports, the area of the weed-infestation was twice as big as the diseased area. The map indicates the exact locations in the field.

Compared to traditional methods such as crop walking, drone-based plant disease and weed analyses are:

  • Fast and save time and resources
  • Able to cover around 500 acres in less than two hours
  • Accurate, with an error rate lower than 2%. With drones, you can analyze every inch of your field
  • Proactive and allow effective in-season corrective measures or can be used to formulate a (VRT) Variable Rate Treatment for the crop.

The data collected allows for more site-specific control of both weeds and disease, saving you time and money. In this case, the farmer could consider applying controls to affected areas of the field or using VRT (variable-rate technology). Research indicates that by applying site-specific weed control using a GPS-guided sprayer herbicide to control broad-leaved weeds in sugar beet, they can be reduced by 41% over a four year period [4].

Early observations of weed density can help to predict yield loss and aid in determining weed control strategies. Data and field maps can be used to formulate VRT prescriptions for chemical control or even for mechanical controls such as precision hoeing of weeds [5]. Drones can also be used to evaluate the effectiveness of control programs and to identify issues such as weed resistance.  It’s been shown that results obtained using drones and image processing and the results obtained by observation can be very close to each other [6]. However, a drone has the advantage in terms of time saved.

Drone data can provide accurate and useful information on these areas and will allow farmers to make informed decisions on the targeted treatment of a field or part of a field.

Want to start your own mapping journey? Take a look at the different Agremo analyses or join the community on Twitter and Facebook to meet other digital agriculture fans!

References

  • P. Lottes, R. Khanna, J. Pfeifer, R. Siegwart, and C. Stachniss, UAV-Based Crop and Weed Classification for Smart Farming. 2017.
  • R. Gerhards, K. Bezhin, and H. J. Santel, “Sugar beet yield loss predicted by relative weed cover, weed biomass and weed density,” Plant Prot. Sci., 2017, doi: 10.17221/57/2016-PPS.
  • Cercospora beticola Sacc. damages the plant by harming the leaves. The number of spots seen in very small numbers and small rounds initially increases rapidly and covers the whole leaf surface. As a result, the leaf dries completely and dies, and it is these changes in appearance that can be picked up by drone imagery. Cercospora causes visible leaf spots and often leads to significant yield losses since certain Cercospora species show resistance to common fungicides. Hence, farmers often need to spray more than once and rotate between different fungicides, and if possible avoid unnecessary applications. With good data on the scale and location of an infestation, a farmer can decide whether an application is justified or not (G. N. Skaracis, O. I. Pavli, and E. Biancardi, “Cercospora Leaf Spot Disease of Sugar Beet,” Sugar Tech. 2010, doi: 10.1007/s12355-010-0055-z.)
  • C. Timmermann, R. Gerhards, and W. Kühbauch, “The Economic Impact of Site-Specific Weed Control,” Precis. Agric., vol. 4, no. 3, pp. 249–260, 2003, doi: 10.1023/A:1024988022674.
  • C. Kunz, J. F. Weber, and R. Gerhards, “Benefits of precision farming technologies for mechanical weed control in soybean and sugar beet - Comparison of precision hoeing with conventional mechanical weed control,” Agronomy, 2015, doi: 10.3390/agronomy5020130.
  • M. M. Özgüven, “Determination of Sugar Beet Leaf Spot Disease Level (Cercospora Beticola Sacc.) with Image Processing Technique by Using Drone,” Curr. Investig. Agric. Curr. Res., 2018, doi: 10.32474/ciacr.2018.05.000214.
Analyze your field

No credit card required