How computer vision can halve farm weedkiller use
Support
Download and listen anywhere
Download your favorite episodes and enjoy them, wherever you are! Sign up or log in now to access offline listening.
Apr 23, 2023 ·
31m 21s
Here are some of my learning points from this podcast, with Nadav Bocher, CEO and co-founder of Greeneye, the precision spraying company: - Greeneye’s mission is to transform the present,...
show more
Here are some of my learning points from this podcast, with Nadav Bocher, CEO and co-founder of Greeneye, the precision spraying company:
show less
- Greeneye’s mission is to transform the present, broadcast approach to spraying herbicides –where farmers spray an entire field, regardless of weed incidence.
- Greeneye wants to spray only actual weeds, using AI-enabled cameras first to detect weeds, and then turn individual spray nozzles on and off accordingly. The company estimates that actual weed infestation in an average U.S. Mid-West field is around 10%, indicating the scale of possible savings.
- The system comprises, on a 36m spray boom, some 12 graphics processing units (GPUs), and 24 cameras. Once a GPU detects and classifies a weed, it triggers the nozzle to spray, all in real time. The typical system will comprise a dual mode, allowing simultaneous broadcast spraying of a residual pre-emergence product (used to control weeds preventatively), and spot spraying of weeds with a leaf contact product.
- Nadav says identifying grass weeds in some cereal crops is complex, such as grass weeds in a grass-based crop such as wheat – “That’s tough, we’re not there yet.”
- The company’s focus today is detecting grass and broad-leaved weeds in three key crops, soy, corn and cotton.
- Nadav says its algorithms have been tested in multiple fields in multiple countries, giving it the ability to differentiate the crop from weeds, the key task, rather than worry about actually identifying weed species.
- The company’s system is retrofitted on to any existing sprayer, a different approach to some of Greeneye’s competitors, who are teaming up with equipment manufacturers to sell an entire new sprayer.
- Nadav says their full retrofit cost is about $200-250k, and expects an 18-month payback on that investment. There, he is assuming a $33/acre/year herbicide saving, and a 4,000 acre arable operation – so to be clear this speed of payback is only achieved on a fairly large operation.
- Nadav says Greeneye is now going into its second season of rolling out commercially in the U.S. Mid-West. He expects to sell "dozens" of actual systems this year, moving from a “spraying as a service” model in 2022.
Comments
Robert Rain
5 months ago
As far as I know there are already many healthcare providers and businesses that have found ways to apply CNN-based CV systems to real-world problems. And this trend is unlikely to stop anytime soon, especially with services like https://www.oxagile.com/competence/computer-vision/ continuing to develop further in the field of computer vision.
Information
| Author | Gerard Wynn |
| Website | - |
| Tags |
Copyright 2024 - Spreaker Inc. an iHeartMedia Company
