Utopia: Universal Three-dimensiOnal Passport for process Individualization in Agriculture
A digital passport for all food products, a utopia or feasible?
Joost Batenburg and Robert van Liere are the leaders of this project, which runs from fall 2020 to fall 2025. We spoke with them halfway through the research and received an extensive explanation of their shared future vision for the food industry; with 1 click, a complete insight into the history, origin, ripeness, etc. of all food products in a digital passport.
Prof. Dr. Joost Batenburg is a professor of imaging and visualization at LIACS and is also a researcher at the Center for Mathematics and Computer Science (CWI). The Leiden Institute of Advanced Computer Science (LIACS) is an expertise center for multidisciplinary research and education in the field of computer science and artificial intelligence (AI).
Prof. Dr. Ir. Robert van Liere is also a researcher at CWI and part-time professor at Eindhoven University of Technology.
Centrum Wiskunde & Informatica (CWI) is the national research institute for mathematics and computer science in the Netherlands. Founded in 1946, CWI is part of the Institutes Organization of NWO, NWO-I. Although based at Amsterdam Science Park, the institute has strong international ties and enjoys a global reputation for its innovative research.
Partners
The UTOPIA project team combines the expertise of CWI in the field of 3D image reconstruction and machine learning with in-depth knowledge of agricultural products (in this case, especially fruit) from Wageningen University and the Department of Veterinary Medicine at Utrecht University (especially poultry). Additionally, Greefa is involved with extensive experience in high-volume production systems for processing fruit, and collaboration is taking place with Meyn regarding poultry. Greefa is a market leader in sorting technology, developing and producing sorting and packaging solutions for fruit and vegetables. Meyn Food Processing Technology B.V. is a subsidiary of CTB Inc. CTB Inc. is a global designer, manufacturer, and marketer of systems and solutions for grain preservation; poultry, pigs, and eggs production; poultry processing; and various equestrian and industrial applications. There has been a successful collaboration with Meyn for some time now.
Approach
The current processing of food products has prompted the research to approach this differently. "We are used to mass production and treating everything the same way, but this makes it difficult in the first place to grasp the quality of each individual product. Because all products are then treated in the same way, there is also a lot of loss during the processing process and thus waste of food.
Finally, there is little insight into the 'internal condition' of food products (think of the ripeness of fruit), this information is actually desired for each individual product."
The UTOPIA project will therefore develop a powerful approach for in-line 3D X-ray scanning of agricultural products, which will make it possible to create an individual digital passport for each product with characteristics related to safety, quality, and morphology derived from the 3D internal representation of the product.
Image: The Utopia conveyor belt: each product is scanned and analyzed, and then a digital passport is created. The passport is used to make the right choices and individual actions.
3D
Step 1, create a 3D CT scan from many directions by rotating the product on a specially developed carousel. This collects the data for rapidly calculating the 3D images. "You want to see the inside, we have a lot of expertise in X-rays and can make an individualized CT scan of each product."
Digital passport
Step 2, the database that keeps track of what has been measured plus the images that come from it. Each product has its own dataset about what is known about this product.
Real-time analysis through machine learning
Step 3, this looks at output from laboratory experiments about what might be wrong with these products and teaches an algorithm to detect this from data coming off the conveyor belt. Currently, the focus is on apples and chicken; the partners in the project have access to samples of these products (including defects) and bring them in to be scanned.
Decision support
Step 4, all the knowledge from the various partners comes together here. "For the first time, CT scans of these food products are interpreted, there is a real knowledge gap." Together with the knowledge from the CT scans and input from experts like apple specialist Dr. Rachael Wood (Postdoc at Wageningen University), we then build knowledge about what is happening with the product.
Persuasion through visuals
The research group has learned a lot over the past two years, especially through the development of an industrial CT scanner in 2016, which can be used to try out various things - on a small scale. Together, they are developing concrete demonstrators to bring these smart technologies into the real high-volume environment.
Challenges
The enormous quantity of products is the biggest challenge in providing a digital passport for all products in the future. Robert explains, "6,000 chickens go through a poultry slaughterhouse per hour, and tons of fruit go through a factory per day, so it is impossible to scan every product in this way. We are therefore investigating what is feasible."
Robert adds that reliability must be 100%, and the ideal scenario is not always achievable. "The food chain is a very complex chain, and you have to continuously take that into account."
Joost explains that the research group also encounters other limitations, such as the maximum duration of a scan. "The goal is not the most perfect image, but especially the feasibility is leading in this. You may not get a perfect image, but it would be sufficient for this purpose."
Future
"For now, this is still a utopian vision; we are far from achieving it, and it is an enormous but necessary step for the entire production industry. Everyone in the food chain wants to move in this direction," Joost and Robert state.
This approach will pave the way for a fully individualized quality assurance and processing of agricultural products, raising the bar for food safety and waste reduction through early detection of ailments and diseases. It will also increase value through individualized processing processes.
