“Accelerating the “aha” moment” is the name of the game at the annual Defrag Conference, which brings together some of the “best thinkers, builders and starters” to showcase internet-based tools that “transform information into layers of knowledge.”
Among the great thinkers and builders at this year’s conference was our own Brandon Basso. Brandon demoed our copters and delivered a keynote presentation on “Closed-Loop Farming,” explaining how aerial imaging is enabling more efficient and targeted data gathering for monitoring and managing crops.
Using drones, farmers can quickly and easily gather visual data to assess the health of their crops. In addition, they can use near-infrared mapping to get more detailed information that they could obtain from unassisted visual analysis. The means they can target treatment to the crops that specifically require it, rather than applying a seasonally-driven one-size-fits-all model – an approach known as precision agriculture. Given that farmers lose roughly $25B a year through improper irrigation, insects and weeds, the potential return on investment is significant.
"It's great to participate in a conference like Defrag where we have the ability to interact with the broader tech community,” said Brandon. “Everyone has a genuine interest in making the world a better place through their respective industries, from precision agriculture to augmented reality. Hats off to Eric and Kim Norlin, and all the speakers and participants for another great Defrag!"
Our kudos to Brandon for creating an “Aha moment” around the evolution of UAV technology for agricultural applications.
Comments
John Stuart, i have been researching what you mention on your reply, have you been able to update the post you mention (considering your reply is from 2013 LOL)
i am very interested on the topic :)
One of the biggest challenges to the DIY solution to this is the lack of robustness in the generated NDVI data.This refers for one to the indicative power of NDVI imagery, generated from modified point and shoot cameras, of actual plant health.
However what is also required is inter-temporal robustness, ie the ability to faithfully compare data from different days, weeks and months. When you can generate a time series over the growing season you can start to detect trends and reverse-engineer the efficacy of interventions taken (eg fertiization, irrigation, pest control). I am working on a methodology to address this and I'll post it up as a blog (and mentionit in the Agricultural UAVs group) when its ready.
A PhD from UC Berkeley opens doors that no DIY project ever will.
Marc, we're working on getting the talks online. A similar talk is here. We are evaluating a few different toolchains and 3rd party image processing software. The end solution will be fully integrated though!
Will the talks be available online? Or generally more information?
My 1st question at the moment is if there is a toolchain from the pictures taken over crop fields to the input of modern farming machines (be it John Deere, Claas or whatever standard out there.. all in the best case :-)?
If it takes 5 programs and a lot of manual intervention it is not usable in my point of view...