From the family farm in Dumont, Minnesota, Kirk Stueve is the tip of the spear for Ceres Imaging in the Midwest, helping corn and soybean farmers solve problems with images, and improving our imagery technology at the same time.

Here’s his latest example of Ceres imagery shedding light on a perennial issue in managing corn. The short video was captured by Kirk in the Ceres Imaging app.

  • The problem:
    • Nitrogen availability varies across individual fields and from season to season.
    • This is a common problem near Dumont and in other parts of the Midwest.
    • The result when treating every part of the field the same is typically uneven crop growth, inefficient use of fertilizer, and/or depressed yields.
  • The solution: 
    • Use in-season imagery and nitrogen calibration ramps in different soil types to monitor the nitrogen availability of the current season.
    • Target any deficient areas of the field that have historically high yield potential with higher rates of N via an in-season Y-DROP application.
  • The result: 
    • 10 days after application and a couple of nice rain showers, the problem areas have higher vigor more consistent with the rest of the field.
    • The south headland was not treated and demonstrates some canopy improvement, but it is lagging behind other parts of the field.
    • This provides some qualitative evidence of success, but the hard numbers collected from the check strips after the fall harvest will render the final verdict.
  • Kirk's application details:   
    • The “before” image was flown at v10-v12.
    • 28% UAN was applied via Y-DROPS when most of the field was ~v12 or more. The south headland, other more narrow parts of the field perimeter, and several check strips were not treated with the Y-DROP application.
    • Part of what viewers are seeing is the natural canopy closure of the field, but it does appear to be less pronounced on the south headland (where no Y-DROP application occurred), which suggests the crop experienced a benefit from Y-DROP application.
    • Correlations to yield in the fall will render the final verdict.

 

Inquiring minds on Twitter wanted to know more details about this example.

Here’s a Q&A with the Ceres team, including information on how Ceres calibrates its cameras to get images that solve problems.

Q. Tom Oswald @notilltom: Impressive video but my gut wonders if there’s more to the picture. I know there’s this thing called calibration of imagery but haven’t spent much time in that space.

A. Kirk Stueve @EyeSkyBootField: Your gut is very intuitive Tom! I'm going to post some validation pictures from the field soon to give a visual sense of the consistency between the two different snapshots. Stay tuned.

This question is a great one, so we turned to the team in our imagery department for a longer response you can find at the end of this post.

Q. Louie Nigg @LouieDN: How much rain in that same time period

A. Kirk Stueve @EyeSkyBootField: 1.7 inches in two different storms

 

Q. Tom Kennelly @KennellyTom: How was picture taken? Sat, or plane

A. Kirk Stueve @EyeSkyBootField The 6-22 before and 7-05 after images were taken via airplane by #CeresImaging

 

Q. Michael @jdfarmernd Was it a variable rate app driven by same imagery?
Looks like it really evened out nicely!

A. Kirk Stueve @EyeSkyBootField Spot on Michael! It was a variable rate application. I also left several check strips with more/less/no nitrogen for various trials. It will be interesting digging into the data this fall!

 

Ceres Imaging Calibration in focus

Calibration and Atmospheric correction
Ceres imagery is captured from planes flying at typical altitudes from one to three kilometers above the ground. To deliver useful imagery corresponding to plant physiology at ground level, we perform both camera calibration and atmospheric correction.

Camera Calibration
Each Ceres camera system is calibrated before it flies over customers fields. We use standardized calibration sources to translate raw images captured by our cameras to meaningful physical units.


The result is that all our camera systems produce equivalent values, which also have a meaningful physical interpretation.


For our VNIR imagery, the cameras are calibrated in our lab using a Labsphere©-calibrated radiance source. Using these measurements, we can convert raw images into radiance, whose units are W/(m2Sr). This is essentially the "brightness" of an image at a particular wavelength of light, as measured at the camera.


For our thermal imagery, the cameras are calibrated in our lab from calibrated black body thermal source. This allow us to convert the raw thermal imagery into an image whose units are degrees Celsius measured at the camera.

Atmospheric Correction
The calibration steps result in meaningful values measured at the camera, a few kilometers above the surface. But what is really needed are values at the plant level. The process of converting at camera values to at plant values is known as atmospheric correction. Ceres uses a method of atmospheric correction which synthesizes publicly available meteorological data, standard atmospheric modeling, and simple physical laws.

For each image produced, we collect a variety of meteorological observations from public sources. We select the most realistic meteorological values for a particular image based on the distance from the field, the time of acquisition, and the local weather patterns.

Precision agriculture Company news Midwest

Back to blog

The difference between Ceres AI and other technologies I've used is the help I get from their expert team.
Jake Samuel, Partner
Samuel Farms
With Ceres AI we can take a more targeted approach to applying fertilizer and nutrients.
Brian Fiscalini, Owner
Fiscalini Cheese Company
These flights can cover way more ground and provide more insight than a dozen soil moisture probes — and it's cheaper to implement.
Patrick Pinkard, Assistant Manager
Terranova Ranch
The average Ceres AI conductance measurement from its imagery over the season has provided the best correlation with applied water.
Blake Sanden
University of California Cooperative Extension