Large group of stacked boxes. Martin Barraud/Getty Images
From food safety to electronics protection, industries increasingly rely on sensors to monitor product conditions. These systems, however, often involve expensive electronics that raise costs and complicate recycling.
Now, researchers at the University of Vaasa believe there may be a simpler and more sustainable option.
Doctoral researcher Jari Isohanni has studied how packaging itself can serve as a condition monitor. His work explores the use of printing inks that change color when exposed to shifts in temperature or humidity.
By combining these inks with artificial intelligence, Isohanni shows how packaging can signal changes with near human-eye accuracy without needing electronics.
Smart packaging has drawn attention worldwide, but the challenge has been in recognizing small and rapid color changes reliably.
Machine vision methods often fall short, detecting shifts only after it is too late. Isohanni’s doctoral research in computer science set out to find which techniques perform best in different scenarios.
“My research showed that traditional, simple computational methods work well for recognizing significant color differences. However, for subtle changes and varying conditions, the most effective methods were convolutional neural networks that are based on artificial intelligence,” Isohanni explains.
Until now, no study had compared the performance of different recognition approaches.
Isohanni filled that gap by showing that simple methods suit large differences, while convolutional neural networks excel in subtle, fast-changing conditions.
Functional printing inks change color in response to environmental factors, offering industries new possibilities for monitoring.
Detecting these changes quickly could prevent spoilage, waste, and damage.
“The color change in printing ink is so subtle or rapid that it cannot be recognized effectively enough with current machine vision methods. By the time the ink’s color change is mechanically detectable, the process may already have progressed too far or damage may have occurred,” Isohanni illustrates.
With artificial intelligence, recognition becomes faster and more accurate. This makes it possible to use printed indicators not just in factories but also on consumer goods, offering real-time feedback on product freshness or safety.
Isohanni stresses that printed indicators come with major cost advantages. They can be applied directly on packages alongside regular labels with little extra expense. Unlike electronic sensors, they do not add to recycling challenges.
“Expensive electronic measuring devices cannot be placed on, for example, a lettuce package, as it would constitute a large portion of the product’s price or could cause additional challenges for recycling. Printed indicators solve this problem,” Isohanni says.
His findings open opportunities for multiple sectors. The food industry could track shelf life with greater precision.
Health care providers could ensure proper storage of medicines. Logistics firms could verify conditions during transport. Electronics makers could detect early signs of heat or moisture damage.
The University of Vaasa research highlights a clear path toward affordable, environmentally friendly smart packaging.
By combining color-changing inks with machine learning, industries gain a tool that strengthens quality control while giving consumers reliable information about the products they buy.
MasterCard
