Utilizing Non-Destructive Sensors to Improve Fruit Quality and Grower Decision Making: An Interview with Dr. Ioannis Minas

Hunter Weber

March 16, 2023 at 4:19 pm | Updated March 23, 2023 at 3:54 pm | 11 min read

 

Assistant Professor Dr. Ioannis Minas from Colorado State University joined us recently to discuss his research and give a behind-the-scenes look at how he uses the F-750 Produce Quality Meter. Our Director of Applied Science, Galen George, and Dr. Minas discussed how traditional fruit quality testing methods are being replaced by modern technology. Dr. Minas has been able to delve deeper into his research and better understand how to measure the maturity of produce using NIR spectroscopy.

NIR spectroscopy is used to non-destructively test fruit’s dry matter and brix in seconds, while traditional methods are inaccurate or require wasteful and tedious techniques. In this interview, we wanted to gain insight into how Dr. minas uses spectroscopy and how he plans to help get it into the hands of more growers. Don’t miss out on this formative interview. Watch or read how NIR spectroscopy is changing the agriculture industry for the better.

Subscribe to the Felix instruments Weekly article series.


By submitting this form, you are consenting to receive marketing emails from: . You can revoke your consent to receive emails at any time by using the SafeUnsubscribe® link, found at the bottom of every email. Emails are serviced by Constant Contact

Watch the full interview here

Dr. Ioannis Minas Insights

Introduction

Galen: Okay. Welcome. Thank you for joining me today. My name is Galen. I am the director of Applied Science at Felix Instruments. And I’m very excited to be joined today by a special guest. His name is Dr. Ioannis Minas, a professor of pomology at Colorado State University. I’ll let him introduce himself and give a little more background on his research, what he does, and where his interest lie. Then, we’ll talk briefly about how you are utilizing our instruments.

Ioannis Minas, Assistant Professor of Horticulture and Landscape Architecture, Western Colorado Research Center, Colorado State University, January 6, 2016

Dr. Ioannis: Thanks for having me this morning, Galen. My name is Ioannis Minas. I’m originally from Greece, but I joined CSU in 2015. My background comes from pomology, post-harvest, and biology. But here at Colorado State University, I work in many aspects of tree fruit production. We are focusing on peach production, Colorado’s major tree fruit crop. We are working on rootstock, staining systems, and cold hardiness. Because of my background, I’m interested in fruit quality.

This is where I get to use your technology in order first to understand how we can improve fruit quality in the orchard. That’s one of the major aspects of my programs as well, but also how we can put technology into the hands of the growers. And peach growers are like many other growers, low tech or no tech industry. So we should explore how we can use technology to give them the information they need to make the best decisions for their crops.

Why Farmers Stick to Traditional Methods

Galen: So you bring up the most interesting point with this technology: our F-750 and F-751 sensors for fruit quality. There’s always seemed to be a lot of interest on the post-harvest side of the supply chain where typically they already have QA labs where they’re doing destructive testing, and that’s just a normal part of everyday life for them. But farmers, especially on the smaller side of operations, have always relied on traditional means and tribal knowledge to assess the maturity of their fruit and when they should be harvesting. So what do you see as the path forward to getting this technology into the hands of the farmers?

Dr. Ioannis: I agree. And that’s the main thing. There are two worlds: the world focusing on pre-harvest, farming and harvesting the fruit, and the world managing post-harvest. This is more distinct in Europe, where I’m coming from. And I know this system very well. Here in the US, you find most, like, many of the growers, are doing their packing. So you can find a vertical situation where the farmer is also the packer, shipper, and post-harvest special. So in the first part, I agree with you. It’s very hard for the farmers to move out of their tradition. In most cases, they are three, four, or five generations on the same farm. So they know what they’re doing and rely on their senses most of the time. Saying: We think we should be harvesting two days from now. We think that we have been doing everything right. Most of the time, these claims are not coming with any data to back up the claims.

I come from a family like that. All of my family are peach farmers in Greece. So I grew up in the north, and I understand that very well, and how difficult it is to adopt a change into the practice you have been making money from for so many years. I have also realized that by doing so, you are not realizing what you are wasting or the cost you are paying from making decisions that, in my view, are based on no data.

This is where I started working with non-destructive sensors and F-750 back in 2016 to see, okay, can we get data from a non-destructive concept? How good this data is? Can growers use the technology too? I see that the post-harvest people are more interested in the packing house and stuff, they’re much easier adapters, but still, you need practical information to get them working with the sensor. The farmers, at the same time, are more relaxed about moving in the right direction.

What we have been experiencing over the last couple of years is that even farmers will adopt a technology if they get the information they are interested in. So we, as researchers, are focusing on some direction. They focus on information that can be easily interpreted and support their decision-making process. So this is where we are now. We’re trying to figure out if we have reliable data and how this data can be transformed for the grower to support their decision-making.

Traditional Fruit Quality Testing Methods

Galen: I don’t know if this is proprietary information or anything, but if you could shed some light on what is traditionally done to assess the maturity in the field versus what you are assessing as far as internal quality or external color, whatever you’re doing with the 750 and how they kind of correlate to each other and how you’re going to be able to show the farmer that, hey, you know, you’re traditionally looking at the color of the peach, but actually what I’m doing with this is measuring, directly quantifying chlorophyll and that’s a better indicator of maturity. So anything like that, that you have insights you feel you can divulge?

Dr. Ioannis: Yeah. Traditionally, the grower will visit the farm and have a chart of the sequence of the ripening of the cultivar. And every season, they will visit the orchard. They will walk in, and they will try to feel some fruit. Some growers will say if I touch 10 fruit and they are soft, I’m going and picking. Other people will focus on background color assessment visually. They’re not even doing any destructive assessment. It’s very hard to find people doing any systematic assessment before harvest. They are mostly based on the background color change of the fruit, which is very hard on the fully red-colored cultivars, which are the majority right now, either in the U.S. Or Europe. And that’s what has been the case.

Now we have changes in environmental conditions, storms, and unpredictable conditions. You may see newer varieties coming, but we don’t have much data on how they ripen across many locations. So that’s another problem that we have. Especially when they have limited labor and have to make a plan on where to locate their labor and what sequence they need to follow. They might fall short in one block with more yield, and they will go one day later. And that one day, in some cases, maybe big waste in the soft ward, or they might go a little bit earlier, and the fruit are already green. They’re not ready or a big proportion of the fruit. So the watchers would not be efficient there, or they would be harvesting green fruit that is equally damaging the product quality that the consumer will have or even affect the efficiency of the packing house. So if the packing house has to deal with too many green or soft, that is also a big problem.

We are using F750 and other parameters. Of course, we are measuring dry matter brix but we have created robust algorithms that they can hold across seasons and across locations. Also, we are using a complex index that is a little bit indoor IP. Now we’re using Spectra and weather information to justify, to predict harvest time.

Galen: Oh, wow. That’s awesome.

Dr. Ioannis: Yes. For peaches, it will be a big solution. This year, we will have real growers in our large indoor pilot in Colorado and Utah. We are trying to see with them how good our prediction is, how this may be helping them, how to report the data, and what information they need. Do they want to know when you should initially go and start picking? Because different growers have different practices. Other growers focus on bulk harvesting, and big supermarkets like dealing like a commodity. Other growers are focusing more on fruit quality. They want to make careful harvests because they want high-quality fruit and direct consumer sales. So there are different things, but the capacity of spectroscopy, and in that case, F-750, is that it’s giving you that flexibility to work through the data, through different approaches, and deliver some information that can be customized for each particular user. It’s a complicated process. It’s not an easy process. We have invested around six to seven years working on this every season, consistent with the data we are collecting. We have broadened our capacity across 25 beet cultivars and are expanding with these different locations this season.

Sampling Requirements For the Best Fruit Quality

Galen: That’s awesome. So you bring up a couple of other good points about the value of technology like this isn’t necessarily just in the obvious direct. You know you save labor by doing instantaneous measurements instead of destructive testing. But the fact that it’s allowing for better labor planning when it comes to harvesting, that seems like the bigger picture of time savings for the farmer and for ensuring the highest quality crop is getting out to the customer. As far as linking data, that’s a smart approach as everything starts to go through big data. Big data is changing the world and being able to pair as much data, all the spectral data, with other inputs, soil, temperature, weather, and all sorts of other precipitation, moisture, soil moisture, and all these other different types of sensors you can have out in your fields. Being able to pair those together to put together a comprehensive and holistic idea of what’s happening in your orchard, that’s going to be a game changer for everyone and ensure that we can have the highest quality crop every year, and with reduced labor costs as well, and labor input. As far as sampling is concerned, did you run into anything about– I don’t know what your practices were, but with using the 750, the F-751, you mentioned that a farmer might go on the field and only grab 10 fruit or something like that. I know representative sampling and sample sizes are important for this kind of data to get an accurate picture of everything happening over the entire orchard. What have you found as far as your sampling requirements?

Dr. Ioannis: We have developed our sampling protocol for the grower. They are required to scan a number of fruits. Then, because we are monitoring the ripening progress, we can predict harvest time with optimal fruit quality. Of course, we are seeing the variability in the population. Not everything will be harvested at the same time. This is impossible in peach. Really what matters to the peach growers is when they should get into the field first. They have a practice that they go every two or three days in the same block. They do buses. It’s very critical to make this first decision. So to know what proportion of fruit will be ready for that day. So this is what we’re focusing on.

Galen: Yes, the critical first day of when the very first fruit is coming to maturity.

Dr. Ioannis: Not where is the fruit to not delay your first picking or to not go too early.

Galen: Right, okay.

More on Dr. Ioannis’s process

Dr. Ioannis: Because you are losing one day in one block, and at the same time, you should be in another block. You know, and that’s either the crew leader or the farm manager is on a hot spot during decision. Peach season is completely different from apples and the major fruit crop because there are many varieties. Ripening time across varieties is critical to determine because things may overlap. You have to move fast. You have little time from harvest to the packing house, cold storage, and shipping. So, what we’re trying to do is we want to optimize the system with information. Like at the different levels with formation. And eventually, we have been successful also with post-harvest problems sensing.

Galen: Oh, more disorder, internal defects, and disorders that are happening, that are appearing post-harvest. That’s great. That’s another huge application for this kind of technology to be able to assess.

Dr. Ioannis: We are piloting the pre-harvest part, which I think has broader applicability. Really and my goal is personally to get peach growers into information-based decision making, informed decision-making for either their farming style to achieve a minimum and above fruit quality so they can be consistent in their internal quality but also plan their harvesting and sales.

Galen: Gotcha. So, I mean, you’ve done a really good job of thinking about not just the entire supply chain from the farm to the retailer but also as far as your use of this technology, planning out your model building, and incorporating variables that are necessary to incorporate into a model to ensure robustness across all your different hundreds of varieties, multiple different seasons.

As you progress in seasons, things change. Regions, you know, growing regions. All those things present new variables and the physiology of the fruit that alter the instrument’s spectral responses. So I would say that other people are thinking about doing similar things with their crops and what we’ve done with certain crops on our end with kiwi fruit and other crops planning out this process and understanding that it’s not a 30-day. You can build a model in 30 days, and it’ll work for the rest of your life kind of process where it’s more of an evolving process.

Dr. Ioannis: It’s an ongoing complex process. That’s one part of the reason. Even we had so much progress on the technological part of spectroscopy in the horticulture industry. Your company is an example of that. You have built a nice instrument to aid the industry. The fact that you must calibrate and how long that calibration lasts is key for industry adaptation. Also, for future use, what you are using as metrics because you mentioned big data errors or different things. It’s very critical to acquire meaningful data. Every crop is different. Every industry has different perspectives, different goals, and different management. So, it’s necessary for horticulturalists, physiologists, and engineers to work together. It’s not either one side or the other side.

Galen: Yes. It is a big collaborative effort if we want to ensure the future of our food supply. I 100% agree with that. Getting people to agree on important metrics and standards should have always been a huge issue. Hopefully, we’ll get around that at some point in the future here and agree that for certain things, dry matter is important; for other commodities, it’s more about the brix, or the internal hue, or the firmness, or whatever it is. But yeah, I agree with you.

Well, I appreciate you telling me about your project and all the work you’ve been doing. I know that you’re continuing to work with the device. I am excited to see as you push to get this more piloted in the hands of farmers and see their reactions to it and the kind of data you can get. I’m excited to see where this is heading. You’ve got a really good project that you’ve been pursuing here, and I hope to see more success from that. I want to say thank you again for joining us today to talk about technology. If anyone wants to check out anything that Dr. Minas has been doing, you’ve got articles and stuff all about you all over the Internet. Just Google his name, and you’ll find more information about his projects and program.

Dr. Ioannis: Yes, thanks for having me. And I can comment that my direction with you all in the company has been great. You guys are always willing to support any difficulty we may face. Thankfully, we are not facing many difficulties. That’s a good thing. Working from a horticultural perspective has been nice, trying to solve industry problems using technology. I’m also excited to see what the future will bring us, especially with our pilots this year. And I’m excited because the growers seem to be excited. That’s a good thing.

Galen: That’s always good to see. Yeah, that’s excellent. Well, yeah. We’ll get an update here after this year’s harvest. And I’m excited to see what comes with that. So thanks again. And as I said, if anyone wants more information about Dr. Minas and his projects, you need to do is Google his name. Trust me, I’ve done it, and plenty of articles are out there about him. All right. Thank you so much.

Dr. Ioannis: Thank you. Thank you.