Millions of tonnes of food end up in the trash every year, even though it has not spoiled.
The Fraunhofer Institute of Optronics, System Technologies and Image Exploitation IOSB in Karlsruhe, the Fraunhofer Institute for Process Engineering and Packaging IVV in Freising, as well as the Technical University of Deggendorf and the Weihenstephan-Triesdorf University of Applied Sciences have developed a scanner to determine the freshness of food.
The sensor consists of commercially available hardware and measures the spectrum of the reflected light in the near infrared by point measurement. Here, for example, the firmness of tomatoes was examined - a quite reliable criterion for freshness. Through machine learning, the scanner was trained to assess the firmness and thus the freshness from the course of the spectrum.
At present the method is applicable for homogeneous food. In principle, it can also be transferred to other issues, for example to detect counterfeits in food or to distinguish wood, textiles, plastics or minerals from each other.