Unmanned Aerial Vehicle The latest drone news
Hectre’s award winning Spectre combines software and hardware to help properly size fruit before it heads to the packhouse.
AgriBusiness Global: Hi and welcome to the latest installment of ag tech talk, the podcast that talks about the technology and services transforming the industry today and into the future.
Today we’re talking with Kylie Hall, lead storyteller, otherwise known as the marketing manager, with Hector, the makers of Spectre (a fruit-sizing technology). We’ll talk about in just a minute. I noticed that Hectre is a two-time (two straight years) winner of an award for ag tech — 2021 and 22, I believe? What can you tell me a little bit about what that’s all about.
Kylie Hall: Yeah. So, first of all, Dan really excited to be taking part. I hope the Kiwi accent isn’t too difficult to understand. Yeah. So, Hectre has been awarded the Global Agtech Breakthrough Award two years running now, which is just fantastic something we feel very humble to receive. And so, the Global Agtech Breakthrough Awards are looking at those companies that are doing transformative technology within the ag space. And this something like more than 1,500 entries are received into these awards. So, Hectre was really lucky to be awarded in 2021 and 2022. It’s a great thing to have, and it just enables the industry to see the caliber company we are.
ABG: Great. Congratulations on that. So, let’s talk about the technology itself. What problem is your technology trying to solve?
KH: Yeah. So, Hectre as a business – we have two key spaces that we work in. We work with growers to help them to digitize the orchards. So, that’s orchard management app that we provide, and then we also, have Spectre, which is our fruit-sizing technology, and that’s the one that we will dig into a bit today. That’s the exciting stuff. I think. So, Spectre this type of technology is what we call computer vision AI technology. So, we have pretty much taught a computer how to identify, detect, and size fruit, even when it’s just sitting in a bin at a very, very early stage – as soon as it’s been picked off the tree using very simple technology, like an iPad. So, the problem that we’re looking to solve is two-fold. We help solve problems for growers who are wanting to know at the earliest possible time when they take that very first, bin of fruit off the trees, are they picking to the right size profile, or do they need to leave that fruit on for a little bit longer and reallocate their crews to a different part of the orchard.
So, we’re helping on that front. We also, help growers to really own their own information, because, as a grower, often you’re seeing your produce through to a pack house, and it may not be for quite a few months later, until you actually know what size produce you’ve sent and how much you’ll be paid for it. Size is such a key determining factor of that price.
So, with our technology growers can actually take control of that, and they can actually have size data at their end and know what they’re producing instead of having to wait for the pack house. And the other part of the problem that we solve is actually for the pack house. So, pack houses have truckloads of fruit coming into them every day, and they need to assist things like size to be able to make all of the important follow-up decisions such as: How long I should store that fruit. What markets I can sell that fruit into. What sort of packaging I’m going to need for that fruit. When I’m going to need to package it. How I need to schedule my staff and my pack line, how I need to set my pack line up.
So, all of these decisions have massive impact on cost or on the revenues that they can gain. So, we are bringing them the size data earlier than maybe they’ve been out to get it so, that they can make better informed decisions on all of those fronts.
ABG: Okay, you mentioned artificial intelligence. How does the technology do all that? Obviously, if it’s sizing it, must be viewing it in some way.
KH: Yeah, good, good question. So, perhaps a little bit of an explainer around computer vision, AI computer vision, artificial intelligence. Pretty much what we’re doing there is that if you can imagine children in the first few formative years, their eyes are like a data capture machine. They’re taking all of these photos with their eyes all the time and building up these pictures in their brain about what they see. So, that they can learn what things are, and label them with words in their minds. So, we’re doing something very similar when we talk about computer vision, artificial intelligence, and we are actually teaching the computer how to identify a piece of fruit, how to detect where it is, how to size it, and also, a little bit around how to (determine what) color it is. So, what the color is that it looks like on them as well. So, that that’s what we’re effectively doing. So, we have created models that the computer learns, and we do a whole lot of data collection. We’re taking thousands and thousands and thousands of photos, and we’re feeding that into the computer. And we are teaching it what these photos mean. So, if you imagine we have people that are circling round pieces of fruit and a bin, and showing the computer that this is an apple.
And when you look at it this way, you’re not actually looking at it the right way to size it. You need to turn it around in size it this way. So, we’re teaching it all sorts of gymnastics, if you like, on how it can, first of all, detect a piece of fruits, know what that piece of fruit is, and then to actually put measurements around that piece of fruit and serve it up as really simple to understand that.
ABG: Okay. So, we have a camera. We have some, hardware and software. Iit’s a combination of things, right?
KH: Correct, correct. So, the Spectre app works on two different types of hardware. We have it just sitting in a standard iPad, and nothing fancy is required for the iPad. Just a normal iPad, and, in fact, some of our customers get reconditioned, second-hand ones for whatever they need.
They can be standing in an orchard or in the pack house. Just lift that iPad up over the bin of fruit. Take a photo, and within seconds our Spectre app is detecting the fruit in that bin, sizing it up, and giving them the size results within seconds.
ABG: Okay. You’ve talked about apples, and you know what I don’t get to in the weeds here. But is it? Cherries are mentioned on your websites, and maybe one or two other types of fruits. Is this applicable to either now or maybe down the road with row crops?
KH: It’s technology that has very broad applications. And this is definitely the space that we’ve started them, because that is where a lot of our customers were, and that is where the problems came from – was our customers sort of feeding into us and industry in general, telling us about things big problems they had around a lack of early size data. So, apples is where we started. We then had the citrus industry come to us and ask for the same sort of technology. So, we’ve built models for oranges, lemons, and mandarins. We then had the cherry industry come to us and say, hey, if you think those apple people have problems you should see the sort of problems we have to deal with, and of course they have this crazy, tight timeline between pick, pack, and produce on the shelf that they are working with. So, getting size data early for them becomes even more critical.
And then, of course, we’ve had lots of other fruit crops with an industry approach us for different models, which we now looking at. But along the way we’ve also, had other crops, too, so, one model that we’ve already developed is onion model, which we did for a fantastic onion packer in New Zealand who was helping us with testing out some of our technology. So, yeah, we were really keen to sort of collaborate with them and get that up and running. So, definitely, the future has very wide and broad applications for this type of technology.
ABG: Okay, Here’s a question about the technology. Growers on their on their fields and sort of every member of the supply chain is like, you know. “Gosh, another software solution. I’ve got accounting software. I’ve got drone issues.” Everybody wants to sell them something. How do you convince someone that this is something they need to integrate into other solutions? Is it something that stands alone? How does that work?
KH: Yeah, that’s a that’s a great question, and I think it’s a responsibility that tech companies really need to take on board is that from our perspective, our company lot of what we do is based around … simplicity. So, how can we make things simpler and easier? And we really bake that into our technology solutions. So, we try to make it as easy as possible. But I think coming with that hand-in-hand is being able to then collaborate with other tech providers to come up with integrations that enable growers and packers to be able to bring all of the data together when they want to in the type of way. So, yeah, that’s definitely something that we’re doing a lot of work on this year in that space. We already have a number of integrations in different areas of the supply chain, whether it might be onboarding integrations to help make things a lot easier and simpler for growers who are onboarding new staff all the time. So, we’re working with those set of integrations. We also work with some packhouses, different bits of software as well. So, we integrate on that way to. And this this is going to be more and more and more of that, and it’s certainly something that we see as really important, and it’s definitely part of part of our road there.
And so, how do we persuade them to buy our technology? That’s a really fair question. We would really have to bring this technology in a way that is simple to use. And this is really complex technology. There are something like 54 tools and technologies that sit behind Spectre. So, we are taking all of that complex technology and really boiling it down through a bucketload of work to make it simple. Taking (something) complex and making it simple is really complicated, and it takes a lot of investment to do that. So, we really have a space in the market. We have become known for that simplicity, which is fantastic, so, that that that helps us.
We’re also bringing this type of technology to scale. When we think about our Spectre, we think about that hand-held version, just having it on an iPad and taking a photo. We have worked with different parts of industry that have had even greater requirements, such as enterprise legal packhouses that have 40 trucks coming in every day, and they’ve got at least 100,000 apples on those trucks that they’re bringing in. They have problems at scale. So, we have been very quick to work with them to find solutions at scale.
So, if we talk about our latest innovation, which is Spectre Top Down, which is taking that really simple, fruit-sizing technology that we’ve created, and then hooking it up with the special camera installation, we can now actually take photos, capture, size and detect more than 5,000 apples on one truck, even before it’s finished unloading at a packhouse.
So, the fact is we respond to what industry needs and that everything we’re creating is born from industry needs. It’s not from our tech stars sitting there going, “Oh, this is really cool. I reckon someone would want this.” It’s really borne out of need. So, that’s super helpful. And the fact that we have we have stayed persistent. When technology has gotten really hard to deal with, we have stayed persistent, and we have driven this through, and we have got to the end, and we have brought the solution to market. And then we’ve worked with the industry to finetune it and make sure that it is commercially valuable.
It’s all of that really, that puts us in a position where we now have the global leading position on this early fruit-sizing technology.
ABG: You mentioned working with the industry. I’m sure as you do that every time you come up with a solution and say, okay, let’s apply it. Then somebody comes along and says, “oh, that’s great. Can you do this?” Looking at your website, there are a bunch of things listed. A lot of our readers are the manufacturers of those types of crop inputs. Can you talk a little bit about what’s going on there?
KH: Yeah, sure. So, we are looking to provide as much of a comprehensive solution to that supply chain as we can whilst also ensuring that we don’t become too distracted and too thinly spread, that we can’t do things well, and we don’t finish.
That’s a really big problem in the tech industry. There are a lot of projects that start that never get finished, because the next exciting thing comes along. So, it takes real discipline to stay on track with those roadmaps to make sure that you get things finished. Spray fits into our orchard management solution. So, it’s where we are working with growers, and we currently provide them with our app which helps them to digitize all of the orchard processes – so, all of their labor management, their harvest management, all of their quality management everything like that they can do through our management app. Spray is another area that we’ve been doing some work in. So, this is all about helping to make it easier for growers to manage their spray programs, to put together these spray plans, to manage the withholding periods – all of those really critical elements that are attached to spray.
It is an area that we recognize is really important to growers, and very critical decisions are made from information around spray. So, thinning is just one of them. Thinning decisions can be made, and whichever way that decision goes can really affect how much of a pack out a grow is going to keep off the trees.
So, we are taking our time in that space because we know it has really critical impact for growers, so, it’s not something that we. We don’t have a cowboy mentality. We don’t have to be first to market. We can watch and see some of the mistakes that are being made. We can learn from those and take those with us.
And we are working with growers and industry to see if this is a space that we want to go all in on and develop that out further. Or maybe one of those areas within our offering where what we mentioned before is being able to work with other solutions, to be able to bring them in and integrate them. So, we will either look to do it all ourselves, or we will look for an integration that will fit that space and fill that back for us, and we don’t have any issue with either?
We don’t feel from a brand or an ego perspective that we have to do it all. That’s not the answer that brings the best value to growers and packers. So, whichever way we feel is going to solve that the best is the way we’ll go.
ABG: Are there some other features that may or I guess, may not come out of it. You know that you’re exploring right now?
KH: Yeah. And in fact, we just (recently) released what we call Speak to Dashboard. So, it’s taking all of the data that we collect with Spectre with that fruit sizing app and it’s serving it up into a really easy format on a laptop so, that people can play with that data. They can group things together. They can look at the fruit coming off a particular block or a certain variety of fruit, and I can see what’s the size of color that the getting out of there. So, that’s been another iteration. We’ve got this great technology. How can we provide even more value from it? And we are always discovering that because often industry will tell us how they using our technology, and we like, “far out. We didn’t even think of that. It’s a great idea. Never thought of using it that way.” So, we learn, we learn a lot from industry.
We are also looking at what other sorts of quality issues that people would like to know about as early as possible, so, they might be information like defects on fruit. So, that’s certainly a space that we’re looking at. We get asked all the time. If we can. Please take our technology and hook it up to a four-wheel drive or a mower in the orchard and capture all of the data while it’s still on the tree — looking at it pre harvest, really sizing. So, we always interested in that space as well.
ABG: So, instead of just a worker standing there with an iPad. Maybe actually, somebody could drive down and get rows and rows as opposed to just a couple of trees here and there.
KH: Yeah. Well, with our technology, so, with Spectre what happens at the moment is that we are providing the earliest possible sizing data for harvested fruits. So, as soon as they’ve picked that first bin, we’re able to give them the data. They’re getting it right at that early, early stage, and what our customers are asking us for in addition to that is that they’d love to get to a point where they could use our technology to actually be able to size the fruit while it’s still on the tree at different stages. So, it’s a space that we have interest in, and it’s probably not going to say too much more than that at the moment.
ABG: Okay, I don’t want to don’t want to get anybody in trouble here, promising something that that doesn’t yet exist.
KH: We know there are a lot of people interested in what we’re doing in this space. So, until we’ve got something to bring out, if that’s what we decide to do. We don’t need to say too much on it.
ABG: Fair enough. What else do we need to know?
KH: Oh, that’s a good question, I think, for most of your listeners probably have an understanding around why size is important, but some of them may not. So, I think it’s probably just something to touch on, and maybe if I could just paint a bit of a picture might be helpful for that. So, if you imagine a fruit packhouse that doesn’t have our type of technology. And let’s say they’re an apple packhouse. They’ve got big open-top tracks pulling into the receiving sites if they pack houses, and they might have like 40 trucks pulling in a day. They got more than 100,000 pieces of fruit on that truck, and before they had technology like ours, they would have QC, quality control staff at that packhouse.
If they had time they would go, maybe take about 50 pieces of fruit off that entire truckload of fruit coming in and they would then take hand calipers and size each of those pieces of fruit records that information, and then try to put that information into something meaningful.
So, if you think about that, that’s like a 0.5% sample size that they’re getting. And then they’re making all of these critical decisions around, “who can I sell this fruit to, what market wants that size, when can I commit that order and say, ‘yeah, I’ve got that?’ How long do I need to store that fruit for? How do I need to set up my packing line and the packaging and everything else that comes with that.”
That is such a small sample size to make that sort of decision from. And what would happen is that because it’s so small, and it’s so time consuming to collect that data, because it’s so, small it’s pretty meaningless. And in fact, it can actually the inaccuracy of it, because it’s so small can lead to really bad decisions.
And those decisions can mean that you go and store that fruit and you think that in a couple of months’ time your going to have the right size fruit to pack for a certain retailer. So, you bring out that fruit. You start running it across a really expensive multi-million-dollar, huge commercial grader machine and that grader machine stops and says, “Hold on a minute. You don’t actually have the size for here. This isn’t that size fruit that you thought it was.” So, for a packhouse, it’s your worst nightmare, because they’re all about efficiency, and getting that pack line to run smoothly. So, they’ve got to stop everything they’re doing. They might have 100 staff spaced around that warehouse ready to do checking and packing, and all these different jobs that they’re doing. Everyone stops. Everything stops. The cost does not stop.
So, you’re paying people to stand there. They can’t be doing their jobs, because this fruit is the wrong size. So, then, you have to make decisions about: What are you going to do with this? And where are you going to get the other fruit that you thought you had from to actually fill this order that you that you’ve promised you can deliver.
So, that was the scenario that was happening. And with this technology, now, when you’ve got that same truck coming through, and you’re sizing like 5,000 apples at one time, even before that chuck is finished.
You’re now up to a much bigger sample size, which is hugely meaningful. It’s like nearly 10,000% more than that initial sample size that you’re getting. So, it’s just hugely massive. Now you can make some good decisions.
ABG: That makes sense. I wanted to ask you this. I should have asked earlier. It kind of fit a different part of our conversation. We were talking about spray issues, and I know that’s still something that’s being looked at. In terms of integrating it, whether it’s the solution with other types of software, or accounting, or whatever it is, regulatory issues are changing all the time. Are there any elements that are part of this that deal with that side of the equation?
KH: That’s a really good point. It’s one of the biggest, onerous burdens that this industry in particular, the produce industry is really faced with. And it’s ongoing. As soon as one regulation comes in, and you’ve been dealing with it, then something else (comes along). So, it’s really tough on all parts of the supply chain. The areas that we help with along that way. We’ve been bringing through lots of different enhancements to our work with things like changes and minimum wage rates – the introduction of minimum wage rates in different states over time. double time, all of those different sorts of labor issues that are coming through. Now in our orchard management app, for instance, we catch all of the time data that workers are working, and we also, catch a piece rate.
So, we need picking piece rate. And then you, of course, have to do some gymnastics once again around, okay, if someone’s picking piece rate, but they haven’t picked enough in a day,
I have to pay them the minimum wage. So, what’s the difference there? How much is the top up that I have to pay them – all of these sorts of complications. So, we have brought to market over the last couple of years technology that helps to address that. We are doing all of that work for them, so, that it’s as simple as when you are clocking in a worker on your orchard you’re capturing the start times, you’re capturing the finishing times. We are capturing through our harvest module all of the fruit that they’ve been picking and tagging it. So, we know who it’s coming from, where it’s come from, what varieties, etc. it is.
A lot of good traceability data through that as well. And then we are also, within our app doing all of the minimum wage calculations in the top up, so, that when it comes to the payroll person, they pretty much click a button and that’s what they have to do. So, previously that sort of requirement would have had probably triple sets of either handwriting or data entry into computers, and so, trying to help make all of that easier is really important. Capturing information like where we fruit is coming from. That’s really important. We do look after a lot of that through our harvest module.
ABG: Sure, I know you have operations, or I should say, people using the solution. And in the U.S. here – in Washington State the big apple growing area. Obviously, in New Zealand, where you’re located. I’m assuming there’s a few other countries that you’re in.
KH: Yeah. As a company, we started back in 2016, and New Zealand is our home country, where we began. We were so fortunate to have amazing growers to work with here. And I’ve been a drop that fun fact in, then, that New Zealand is the world leader, for the most apples produce per hectare on the planet. So, go New Zealand, little country down the end of the world doing great stuff. So, with that with that sort of heritage it’s been a fantastic place for us to be originating from a base from. New Zealand was our first big market. We then we’re very lucky to be introduced to some fantastic growers and packers in Washington, and that began our foray into the U.S. market. We are very lucky to be able to call (several prominent growers) our customers. We feel very, very fortunate for that, not just because they are amazing customers to have, but they are such great quality-focused and innovation-focused businesses to work with. So, when we are trying to bring out great solutions to these problems that they are amazing people to work with. So, big shout out to them. We are also, in Mexico and Chile. We are also, in Australia, and we are now in Italy as well, and working with lots of other growers and packers in many other countries that want to pick up our technology.
ABG: Okay? Well, I want to thank the lead storyteller for Hectre Kylie Hall. Thank you for being with us. If somebody wanted to learn more about the company or the solutions you offer? How should they go about doing that?
KH: Just look us up online. Our website is www.hectre.com, or please feel free to reach out to me directly to [email protected].
ABG: Terrific. Thank you very much for your time.
KH: Thanks so much.
