<img height="1" width="1" style="display:none" src="https://www.facebook.com/tr?id=151207998623800&amp;ev=PageView&amp;noscript=1">

Drone-based Plant Count Increases Accountability

on June 16, 2016 Agriculture

upclose image of tomato plant.png

A grower compares the number of tomato plants in the field to the number supposedly planted

By Anya Lamb, Marketing Manager @DroneDeploy

Earlier this week we announced two new tools to help growers count plants and gain insights about stand establishment. In this post, we’ll take a closer look at how one grower used the new plant counting tool from AgriSens to verify planting work done by a third party.

Toshiro Aoki is an agriculture consultant that works for a private farm in Northern California, and he’s always on the lookout for new tools that can help make growers more efficient. Interested in mapping, he considered several different software vendors, and eventually settled on DroneDeploy because it was the best match for what he wanted to do in terms of cost and capabilities.

More than simply mapping, Toshiro was especially interested in using drones to count plants. When he heard that DroneDeploy was testing a way to count crops, he was excited to help beta test — and had the perfect field in mind.

An outside company had recently transplanted many hundred thousand tomato plants into a 74 acre field, and the plants weren’t doing well. “In tomato transplanting, you get charged for what’s in the field”, said Toshiro. “The way they usually count is to do a rough estimate over 100-ft plots and take an average.” According to greenhouse’s estimate, losses were only 5%, but based on what Toshiro was seeing in the field, he suspected losses were much higher.

“We wanted to compare the number of plants in the field to what we’re being billed, so that we can hold the planter accountable,” said Toshiro.

Collecting data in hours, not days.

 
drone hovering over tomato crop.jpeg
Toshiro used a DJI Matrice 100 drone to map the 74 acre field.

The first step was to fly the field and capture imagery.

“I waited until the plants were at a certain stage — about 2–3 weeks old — where they’re big enough to be visible in the map and small enough that you can tell an individual plant from another,” said Toshiro.

He used a Matrice 100, which allowed him to fly the entire 74 acre area at an altitude of 290 ft in just one 30-minute flight.

 
drone imagery of 2 to 3 week old tomato plants.png
These tomato plants, about 2–3 weeks old are easy to distinguish from one another.
Using traditional methods, estimating the plant count in the field — by counting the number of plants in a number of representative 100 ft by 100 ft blocks — would have taken a day or two and wouldn’t have been as accurate.

Analyzing the Map

 
interactive drone image of tomato crop.png
Click to explore the completed interactive map

Once the drone landed, Toshiro uploaded the imagery to DroneDeploy to generate a map, and then sent that map, along with answers to a few questions, to the AgriSens tool. Later, Toshiro received a report with a count of the number of plants in the field.

 
agrisens tomato.png

The outcome was clear — rather than the 5% loss the planter claimed, the map showed 26% fewer plants than the farm had been billed for. With this information, the grower can now go back to the planter and hold them accountable.

“If we’re paying a company to do a transplant, and there are problems in the field based on how they planted it, we can look at the survival rate. We can say you messed up, we’re out all these plants, and decide whether we need to file an insurance claim,” said Toshiro.

In other words, for the relatively small cost of a drone map and the plant counting analysis, Toshiro and the grower now have a much better shot of getting financial recompense for the lost tomato plants.

“The plant counting seals the deal. It saves us the trouble of having to go out and count the whole field,” Toshiro said.

Where to Learn More

To successfully use the plant count tool, please make sure that your map meets the following imagery requirements:

  • At least 2.5 cm/pixel resolution
  • Individual plants are visible in the map

For complete imagery requirements and step-by-step instructions, please check out our help center.

Anya Lamb

Drone industry researcher and storyteller. Content and product marketing at DroneDeploy.

Subscribe to Email Updates

READ MORE OF WHAT YOU LIKE.