 Thanks everyone that chose coming to this session rather than the other ones Voting with your feet, hopefully you choose me. No, it's all good So thanks to Graham and and more for their presentation. That's really interesting and obviously we were before I start We obviously like to try and incorporate those seris imagery into other other products as well So open to Collaborations and working together with people So if we just go full of it So seris imagery Why would you choose seris rather than some other company that that might do imaging for your farm? We're we've we we start in 2013. We're a startup We we do like imaging of mostly of horticultural crops, but we also do row crops unirrigated crops enough in We have been experimenting Australia in that but more so in other in America. So so we we started in So we've found in 2013 And probably the first bit of imaging we did in Australia was 2015. So so Australia's pretty not exactly The it's a smaller market in America But it's a really important part of the seris story because we've we've been imaging for a long time in Australia And we we deliver some really good value to to our to our customers here so So yeah, so first off we were our founder was up here A student and he was in Brazil when he was doing a lot of rainforest imaging for For his for his PhD and he was like well I wonder if we could apply these imaging techniques we're using here in crops because at that point people weren't they were using satellite It wasn't that great a quality. Obviously that was a long time before For Sentinel so his thing was well maybe this this imagery really useful when we're trying to work out how to how to get through a drought So that was that was where we started we partnered up with With some guys in UC Davis in America and and we come up with some with some products from from the from imagery so so our Recently we were meant to feature the New York Times Talking about how we we help different crop growers So so I guess a goal of ours will be to feature in the Adelaide advertiser sometime the next 12 months Yeah, so we we image about probably maybe five crops in Australia So there's lots of potential for for working in other other areas So our our approach to imagery and we're open to using other imagery as well in our platform But at this point we do we have a camera. We build ourself that goes on the on the wing of a plane and we We do out we fly over the the clients product and then create imagery like that So so we feel this is really good because it allows us to create really high quality image So we can we can vary the height depending on what what crop we're using and we can deliver like a range of different products So I guess if you wanted to sum it up the seris difference to to other to other providers we we Put a lot of emphasis on trying to collect really great data with their own sensors We've got thermal camera and and all our the bands are separated out So the red green and the infrared and that are separated into different bands So then we then combine that up into a product that's specific for For for a crop so we produce self in not necessarily doing just doing imagery we Our approach is we really want to try and find a value proposition for a particular crop So we've got a couple there. I'll go through with you up and so Yeah, our emphasis on really getting great data to start off and we can and using controlling the whole process by Using planes by by creating your own camera and that allows us to be really To be able to develop to be able to deliver really good stuff, so I want to just go through and Show you a couple of products we create so so under irrigation management We that's one of our one of our key Products so this is what we call this this image here is what we call our water stress image This is a this is a this is a pivot that you can see there. So the the blue areas Low Low water stress so in other words the crops and handling it all right and the ready areas where it's struggling a bit more and interestingly you can see there the Just where that the pivots actually in progress as we fly over so so that timing of flights is something We can really hit with well depending on the cloud and that we can we can usually be a little bit better than that than a satellite so so yeah, so so one of our little Value propositions we could say is for over 20% of the pivots we fly we find something that That can be actioned or where where you're not maximizing the value and I'm sure the other imaging products Identifying issues as well. So it just depends on the management on how much value you can find but generally In any big farm that we we image like there's always something that you can see there And then once you know about a problem the farmers got to go on here to go and have a look and work out Whether there's something they can change to to improve it or whether they just Whether there's whether it's a soil type rate of issue, but usually there's something they can change so When you've got a big property, it's really hard to get over everything You like it and you've usually got people working for you so the imagery gives you another layer of checking so We find that customers get a lot of value out of just being able to get that status update like in season with the imagery So just wanted to go Just on on the pivot water stress Something we've we've been experimenting with in America is trying to make sure that the wedding front of the of the pivot is actually Hitting the that the potatoes are not getting too wet basically so without thermal imagery So this this one in the the grad the purple one there is a is a what we call a raw thermal imagery So this is an output we do for potatoes So that's basically the thermal camera Not really processed much As it comes off the plane and that provides a lot of value in terms of like we can see where the slightly dark Are his are his wetter and I haven't actually got one for an orchard But I'm like sharp really quickly in that in with that with that layer So it's a good way to get over an area really quickly in fine leaks So thermal is something we really do really well from the plane It's a little bit harder to get from satellite because it's your further away and you've got more atmospheric Variations so so we see that as one of our really great value ads compared to other imagery, but and also obviously there's a lot of Synergy and working with other other imagery as well that maybe is higher highest Temporal resolution in terms of like you might be flying more often But you might have series as well to to let you see that high resolution picture So this pivot here you can see there that if you look down there that the area that hasn't been watered The longest has been watered. You can see that starting a little bit dry the area That's and it's the that's a really good Spread of irrigation so with with our flights with Potato grills in the US have found that they can actually keep track of like where their water logging is so water logging can be a is an issue that Takes a lot of value out of potatoes because you get it really takes away from your quality if they're sitting around in water log for too long So getting that interval that irrigation interval. They've found that the serious water stress product is really good for For not allowing the pivot to stay wet for too long I think I've gone So the other so this is a this is a another product we do for Irrigation crops So you can see at the top we've got a like what we call a water stress And I'll have a high-resolution photo of it later So you can see that a slight variation through the through the through the paddock based on This is this is really a measuring thermal canopy temperature so with with our thermal kind of we can see pick up slight variations in in in Heat which will which will tell you like if the canopy is slightly warmer So as as irrigations applied If if something's happening with a valve, and it's not quite right then you'll It'll start to the transpiration will start to slow down in the plants So you can if you can see that that's following in irrigation Valve then you're about to pick up areas that are That are not being irrigated properly So before the before the plant start to droop with it with our water stress people can see that there's something something not quite right And the bottom one down here is the canopy via so that's basically That's an NDVI shot there. So you can see that the slight variations through that Actually, yes, you can see there. That's possibly a soil type difference there, but Obviously there may be some actionable change you can make to that area there where it's and you can see there Slightly higher bigger through that area there. So if you're getting a lot poorer quality crops through there I'll show you an example later Then obviously then you could make changes based on that So so in there in the blot you can see there this this this row here with the water stress It's been picked up as slightly Lower bigger I'm low a slightly more stressed in the water stress and you can see there that that's about the same size in in the In the NDVI so therefore the growth is about Volumize it same but is there something going on with this row? So it gives people a quick check of their of the irrigation and if there's a blocked Pipe or whatever then that there might be something happening there with stopping the water rowing it flowing up there So we'll pick up with that without irrigation quicker than then What you will with By by no seeing in the field So here's another value proposition from from from the water stress imaging So interesting just as that question on where we could put where how how imagery works in with the Internet of Things devices They're actually technology to work really well together So for instance here take this image if you had your a sensor there Or if you had your sensor here you're going to get a quite different reading like in terms of Soar moisture or if you had like a sat flow meter or whatever So obviously having an image as well to be able to give you that data on where that Sensor fits into the whole property Is is really a value add So so coming back to this example So you can see there this this area here is getting a little bit wetter than the rest of it So it's probably a soil type difference. So this is an example. Definitely that we see all the time in Australia and some of our growers have managed this by by reducing the irrigation here, so maybe snapping off like Changing the pipe over or click clipping off every second emitter And so then you get a more even output of yield so then the crop will become more even and you know You better house at the same time and and Yeah, in this case here 25 to 30 percent improvement in grape quality and that's pretty common to see that sort of improvement So this one is another American example. So you can see that this is our water stress product. So this is there's four Different once again, we've taken the thermal and we've reduced it into the four categories So it's easy to see where the where the variation is So in this area here, it's pretty clear that something was going wrong because The top area there's a little bit more water. Probably there may be the but this whole area is Slightly less than what it should be compared to these other areas in terms of water stress And so making small changes to the irrigation they they could even that up and and it was an adequate and especially at a crucial time of year if the water stress if the irrigation slightly off then you can bump it up and And it really does make a difference to the nut to you to yields so so those are Horticultural crops treat nut crops so coming across to to other examples this one here is a tomorrow crop and so imagery Probably provides a real quick value added especially when there's Pat with pivots and things like pivots and tomatoes like the paddocks are changing all the time so you can you can pick up differences in the in the You can pick up different paddocks really easily without having to change equipment over so you can see here that 40 Okay, back sorry So our water stress picked up that there's a big difference between The water that was actually even though the farmer thought he was applying the same amount to every the whole crop When you looked in the in the in the water stress you can see that it's getting a lot less this end So small changes to the valve meant that even though it's still a little bit uneven. It's a lot better So that's now Whether whether that was a soil type and so he changed around a little bit and added more into this higher block And even it up, but that's a quick yield improvement that can be made with using that the thermo imagery So this is the potato pivots So this is another example of a value add so you can see there that is these couple nozzles are probably Maybe not working probably so that shows up really strongly in the thermo imagery that straight after the irrigation They're just not looking as well as the rest of it So you can see there as well. This is the wedding front. So tracking that wedding front and seeing Actually just in front of it is actually dried out enough. So we're not we're not putting water on top of already waterlogged areas is a really Potato grills have found that to be a really big value add for for our imagery So most of the the examples I've given you so far have been like the from the thermal so the water stress imagery So this is this is what we call our chlorophyll index So it's similar to NDVI, but we've got a different logarithm we used to work it out and we find that Although that's just a red and green image. So it looks the same. We found that it picks up Slightly more growth. So at the higher end. So we're in areas like this where it's actually really growing really well You'll see more variation using the chlorophyll imagery compared to NDVI. So this This patch here. This is that NDVL versus chlorophyll the same Flight in the same time so you see with NDVI. It's uniformly fairly green And so that's obviously growing really well, but when we look at our chlorophyll We're seeing more variation and you can see that the high top end of the pivot is better So when you when you come to optimizing in yields and once you've dropped out the the obvious problems You can pick up with irrigation and having really good data quality so that you can actually See change and it be repeatable is really important So yeah, so the nitrogen we find it's really useful for tracking that in-season growth and in the case of pivots There's probably the option of adding in extra fertilizer depending on where where the crops up to so in this case here Obviously they've picked up a change and then applied extra to here Based on what the imagery showed so obviously having that high resolution data allows you to get more Insights and all allows you to Be able to Sort of target your application, I guess So another tool we've got with it with a platform, but obviously we're we could Our imagery could also be used in other platforms to do this. We create What's called a bare soil? There's two things is zone creation. So we create we we do something called bare soil imaging So when when a pivots Empty before they start before we start planting it We find that doing we can actually pick up the zone differences quite from soil type So you can create a zone from from the From from what the imagery looks like from from from the from that From the bare soil image. So that's a really great way of seeing soil type variations Guess what I'm trying to say. Sorry And Obviously later in the year quite often that the growth patterns will be really strongly Correlated to what the bare soil was so being able to see it before you plant it Especially if it's not a pivot use all the time can really add value So we can do the VR a maps like the verbal rate map in our app So so the way that the imagery gets delivered is people at the moment they They they contract with us and then we deliver it straight to a web app But we are also open to talking and importing our imagery into other other platforms Yeah, so this is this is what we call this is a this is a product. We call color infrared So as I assume that the bare soil imagery This and we can create soil maps from that that will give you an application Rate map before you start the pivot something we've been working on the last couple years is Most of the time without without imagery people will create the image someone like myself will go out and talk to the grower and They'll identify areas where we can make changes or where we can where we can where they can change a management Basically to to to make use of the information the imagery encapsulates. We've also been adding in In some crops the ability to we automatically detect like based on what we know about pests and the crops Where there might be potential issues. So this is what we call our Anomaly detection so we do the image and then we run it through our program and Identify areas that compared to this image in the last image. There's a change that's worth following up. So so in this case here We've identified three little plots that Looking at the image we think considering what we know about aphids and what we know about this crop There's a possibility that there's aphids at that spot and quite a lot of people will go and look at it And we're always improving that and trying to validate that better. So that's that's our automatic automatic anomaly detection So these three images as well, there's we've tried to We we're picking up different things based on yes, so this we can apart from just a point anomaly We also identify areas that I've got we think there's a problem in the whole block So in this case here vermicelli and wilt we think is perhaps causing this pattern and This case here. We think perhaps rot is a problem up here and there So so that's that all comes like it's delivered as the imagery is delivered basically so So in terms of later in the year after a season's finished We use our imagery in various ways to try and help people make bigger scale management decisions in this case here This is something we've been doing for quite a few years as tree County So basically we we have a logarithm that runs through the image and identifies each tree center So each those little dots is a tree center And from that when you've got an older orchard in almonds or citrus or whatever then quite often There's missing trees for various reasons. So this this area is So those areas obviously trees so you then you can count that as a tree and Use use that in your planning to work out how many replants you need to get Okay, yeah tree counts. So here's a here's a couple examples of tree counts. So you can see that Depending on the actual types of tree we can get pretty good error rates So just rush it is but this is this is this is another error in terms of delivering this at a bigger scale This is this is this is our anomaly detection, which we call field rank so we can put each field against her like a This is where we think they're at so green a red is obviously not so good green better Oranges check something out. So when you've got a great big when you've got a big pivot farm having this information saying Oh, I think these pivots are where they rank. This is where the pivots rank can be really useful And can be stress we create this stress Water stress image and then we can we combine it up to give you a feel for where that where the where your strong areas And the crop on that will often match really well with yield. So that's thanks for that