Scanner determines whether food can still be consumed
Is it still good or does it have to go? This is the question asked by millions of people every day when they stand in front of the fridge. Out of uncertainty, many choose to throw away and dispose of foods that are often still in excellent condition. The WWF Germany environmental foundation recently drew attention to the fact that ten million tons of still edible food ends up being thrown away unnecessarily every year in Germany alone. A new pocket-sized scanner could curb this development. It recognizes whether a product is still stable or not.
The Fraunhofer Institute recently presented a small scanner that uses infrared light to determine the actual freshness of food. The device sends the data to the smartphone, where it can be viewed using an app. For example, consumers and supermarket operators can check whether food is already spoiled or not. The Fraunhofer-Gesellschaft recently presented the scanner in a press release on the institute's website.
Scan instead of throwing away
In future, the affordable and portable scanner should ensure that less food is thrown away that is still edible despite minor blemishes or the expiry date. Regardless of whether the goods are packaged or not - the so-called near infrared sensor provides information about the degree of freshness of the scanned product. In addition, the user receives information about how many and which ingredients the food contains.
Ease of use and high mobility
"Infrared light is sent precisely to the product to be examined, then the spectrum of the reflected light is measured," explains project manager Dr. Robin Gruna how the process works. Based on the absorbed wavelengths, the device can draw conclusions about the chemical composition of the goods. Near infrared spectroscopy has long been used in the laboratory, adds physicist Julius Krause from the development team. What is new, however, is that this technology is now also available at low cost and on the move.
More insight for consumers
As the developers report, the small scanner can do more than just determine durability. For example, he also checks the authenticity of a product. "Food is often counterfeited, for example salmon trout are sold as salmon," explains Krause. Such information can also be determined with the device. For example, a spilled olive oil can also be recognized in this way.
The limits of the scanner
The development team also points out the limitations of the new device. It currently only evaluates the product quality of homogeneous foods, i.e. foods that do not consist of several ingredients. Fruit, vegetables, unprocessed meat and yoghurt, for example, are no problem for the device. According to the current state of research, products such as pizza and other ready meals are not so easy to check. In a further project, however, this should also be possible in the future with the aid of imaging spectroscopy, fusion approaches with color images and spectral sensors.
How does the scanner evaluate the food?
According to the development team, the scanner evaluates the infrared spectra using intelligent algorithms that search for corresponding patterns and laws in the measured values. “We can increase the recognition potential through machine learning,” reports Gruna. The project manager explains this process using the example of minced meat. The scanner measures the infrared spectra of the minced meat and supplements the data with statistical methods for microbial spoilage. From this, the device can derive the remaining edibility of the meat. In extensive tests, the scanner had achieved a good agreement with the actual number of bacteria.
Mobile phone analysis
The scanner sends the data to the smartphone via Bluetooth. In the next step, the data is then evaluated from a cloud-based database and displayed via an app.
When will the device be available?
The device is to be used in supermarkets as early as 2019. The next step is to test how consumers accept the device. The project manager is already planning further areas of application. For example, the system could be used to differentiate and classify plastics, wood, textiles or minerals. "The area of application of the device is versatile, it only has to be trained accordingly", summarizes Gruna. (vb)