In 2050, they are saying that there isn’t
going to be enough food on the planet to feed the population. Not only that, but
the amount of nutrients that are going into the soil that’s not good it’s
running off into waters reducing the amount of herbicides and pesticides that
are part of our world and yet being able to feed the world as it’s going to grow
by 2050 that’s a major problem. So we’re hoping to help solve that. We’re working
with agriculture consultants and their job is to go out to a field and tell the
farmer, “hey you need more fertilizer. Your yield count is gonna be low this year
because of this natural disaster…” So what they do is they drive their trucks out
on these long dirt roads but they can only look at about ten percent of what’s
there at any one time. We’ve been working with them to use drones to go out there
and they can photograph a hundred percent of the field as frequent as they
need it. So the drones gather the data but the problem is you get massive
amounts of images – we’re talking about gigapixels
of image so lots and lots of images. They have to be stitched together. Right now,
the typical solution is go home upload it on a really slow connection – most of
these farmers are in rural areas – and then you have to wait 24 to 48 hours to
get an image. So by the time you try to make a decision, things have changed.
ViSOAR Ag Explorer is going to allow farmers to gather data with aerial
drones and then be able to take it to their truck and stitch it within ten
minutes – all like 600 images from particular scan and then keep doing that
throughout the day. And they have the maps in hand and make the decision they
on the field. For us, the research is still very connected to the business.
Through the NSF funding with the I-Corps we were able to find that there was an
need for a solution within the agriculture market. And then, with our SBIR funding we were able to take the ideas and actually create a product for the
market. The magic sauce that is ViSOAR is the ability to do reading of images
fast and writing them fast. It’s not just about getting the data but actually
getting the information out of the data. All this amount of imagery is
basically a stack of pixels so you can think of like the pixels that represent
the individual weeds and the pixels that represent the entire map and so when
you’re looking at the data it’s actually only giving you the pixels that you need.
So even though there’s like a Giga pixel map that goes this big like Google Earth
you’re only looking at a small part of it like your town and we’re only sending
the pixels that go with that particular part that you’re interested in. So we can
go down centimeter accuracy which is what you need in order to identify weeds.
This help farmers use less inputs get more output and then increase the amount
of food and products they get out of their fields.