SERVICES

Precision Agriculture

Agriculture surveys are specialized geomatics products, focused on using multispectral and hyperspectral imagery to classify vegetation and identify crop stress, enabling precise application of agrochemicals and water resources

In early days unmanned aerial vehicles may not have been used in the agriculture industry but if there is one industry that has gained a lot because of these drones, it is agriculture. Our drones are increasingly being used by farmers to help with:

  • Identification of crop stresses and yield-limiting issues;
  • Irrigation efficiency and water management;
  • Early detection of plant diseases and pests;
  • Monitoring of growth rates and yield prediction;
  • Tracking overall crop health (vigor) through the growing season;
  • Detecting the early onset of noxious and invasive weeds;
  • Optimizing inputs such as fertilizers; and
  • Assessing canopy coverage and plant counts.

The application of UAV technology in agriculture around the globe is increasing at a phenomenal pace. A number of studies from around the world suggest the adaptation of UAVs for agricultural purposes will be the largest single non-military beneficiary of this fast growing technology.

Farmers are able to protect their plants from various diseases as our drones help in identifying symptoms of diseases through imaging of the field from a height. These drones also collect important data about the soil with the help of NDVI (Normalized Difference Vegetation Index) imaging. The use of UAVs in fields has helped farmers a great deal as they need not carry out manual inspections of the crop to protect it. These drones are lightweight and they can fly over the field to reach every part of it. The high resolution images taken by our drones tell a farmer a lot about the health of the crop and the quality of soil to take appropriate action.

UAViation-Agriculture1

What Is Precision Agriculture?

Precision agriculture is a farming management concept based on observing, measuring and responding to inter- and intra-field variability in crops.

The goal of precision agriculture is to more efficiently apply a farm’s limited resources to gain maximum yield. A primary method for doing that is to minimize variability of crop health within and across fields. To learn more about precision agriculture, read this excellent overview published by The Economist

Due to its nature, precision agriculture requires a LOT of data to work. The three main types of data include:

  • Geo-tagged images: visible and multi-spectral aerial images taken of fields, over time; this is where drones play
  • Equipment data: real time feedback & logs provided by sensor-equipped manned and unmanned equipment such as seeders, spreaders, tractors and combines
  • Management data: crop yield and other data provided by farm operators

Where Do Drones Fit in Precision Agriculture?

Drones are really just a new, high-precision way to obtain geo-tagged images from the air. Compared with other aerial survey methods, drones generate more precise and more frequent data about the condition of crops. This data is used in many ways to improve the performance of a farm’s operation.

For surveying fields of less than 50 hectares in size, drones are cheaper than manned aircraft surveillance, manned scouting and satellite imaging.

Drones are used to gather a variety of image-based data about the condition of crops, fields and livestock including:

  • plant height
  • plant count
  • plant health
  • presence of nutrients
  • presence of disease
  • presence of weeds
  • relative biomass estimates
  • 3D / volumetric data (piles, patches, holes and hills)

For livestock operations, drones can be used to monitor the location, status and movement of animals over time with more frequency and at a lower cost than other means.

Drone data is used to do farming jobs more effectively and efficiently, including:

  • Crop Scouting – replace men with drones
  • Crop Health Monitoring – biggest ROI, by far
  • Field Surveying/Scouting (before planting)
  • Nitrogen Recommendation
  • Yield Monitoring
  • Plant Stress Monitoring
  • Drought Assessment
  • Senescence Analysis
  • Leaf Area Indexing
  • Phenology
  • Tree Classification
  • and more

Drone Advantages Over Other Aerial Imaging Systems

By some measures, 80% of the global drone industry revenues touch agriculture in some way.

But why would farmers – some of the most risk-averse people on Earth – adopt such a new technology?

Perhaps it’s because agriculture drones offer clear advantages over other crop monitoring methods including satellite imaging, manned scouting and manned aircraft. These advantages include:

  • Cheaper Imaging: for fields less than 50 hectares in size, drones are considerably less expensive than satellites or manned aircraft surveillance.
  • Greater Precision: drone cameras take centimeter-level images that reveal much more detail about a crop’s condition.
  • Earlier Detection of Problems: because drones survey more frequently, weeds, pests and other abnormalities are detected earlier.
  • Total-Field Scouting: instead of riding an ATV around the perimeter to scout perhaps 5% of a field, now every field can be scouted 100% using drones.
  • 3D/Volumetric Data: drone images can be used to calculate the volume of piles, holes, hills and patches. These can be compared to Infrared images to detect density issues like hot spots in a crowded beet field, or to identify contour problems such as north slope shade issues.
  • More Frequent Index Reporting: drones offer a cost-effective way to monitor crops more frequently for key indices like CCCI (canopy chlorophyl content index), CWSI (crop water stress index) and NDVI (normalized difference vegetation index).

How Flying Cameras Measure Crop Health

Most agriculture drones depend on multi-spectral imaging to spot problems with a crop’s health; specifically, they look at changes over time in visible light (VIS) and near-infrared (NIR) light reflected by crops. These images are taken over time by drones, manned aircraft or satellites.

It is possible to detect plant health from these images because plants reflect different amounts of visible green and NIR light, depending on how healthy they are. By measuring the changes in visible and NIR light reflected from a crop, we can spot potential health issues.

This image explains the general idea:

To monitor changes in plant health over time, drone images are processed to calculate a tracking index called NDVI (normalized difference vegetation index), which is a measure in the difference between light intensity reflected by the field in two different frequencies:

NDVI is the ratio of near infrared (NIR) reflectivity minus visible red reflectivity (VIS), divided by NIR plus VIS:

NDVI = (NIR-VIS)/(NIR+VIS)

Here is what you see when you compare a normal camera image of a winter wheat field to a NDVI-processed image of the same field:

  • Color images of the border between the wheat-after-wheat image.
  • VI image of the border between the wheat-after-wheat image. The stark contrasts between field can be easily seen from the aerial imagery.

VIS and NDVI images of winter wheat field (courtesy Agribotix)

Notice how the NDVI-enhanced image (right) does a great job separating the healthy wheat stalks (green) from dying edges (red) and the dry earth (black/brown).

There is some debate over whether NDVI is the right index or whether the simple difference between light spectrums (NIR – VIS) is more useful.

NDVI is the most popular index calculated using drone data, but there are many others. Some may be more or less important to your farm, depending on your situation. Some of the more popular indices include:

  • CWSI (crop water stress index): measures temperature differentials to detect/predict water stress in plants. Requires a thermal imaging sensor and the use of a nearby weather station.
  • CCCI (canopy chlorophyl content index): detects canopy nitrogen levels using three wavebands along the red edge of the visible spectrum. Requires visible and near infrared cameras.

If you want to dive deep into this emerging science of agriculture image processing, here’s a list of the latest research.

NDVI – Normalized Difference Vegetation Index