Fraunhofer Institute IFF

App exposes ingredients

An app developed by Fraunhofer researchers allows users to look directly inside objects and displays their ingredients. This allows apples, for example, to be examined for pesticide residues.

Checking apples for pesticide residues using a smartphone - this will soon be possible.

© Fraunhofer IFF

The 'HawkSpex (R) mobile' app from the Fraunhofer Institute for Factory Operation and Automation IFF in Magdeburg will enable consumers to check such information in future. To do this, users need to open the app on their smartphone, point the device at the object to be checked - for example the apple - and then receive the desired information.

No hyperspectral camera necessary

There are already systems that can be used to carry out such measurements. However, the user usually has to clamp additional parts, such as a prism, in front of the integrated camera. To enable consumers to do without a hyperspectral camera, the researchers at the Fraunhofer IFF illuminate the object with light of different colors. Instead of the camera measuring the light intensity in the different colors, the display illuminates the object successively in fractions of a second in a series of different colors. So if the display only shines red light on the object, the object can only reflect red light - and the camera can only measure red light. Intelligent evaluation algorithms ensure that the app manages with the limited computing power of a smartphone and compensates for the limited performance of the camera and display.

The first laboratory version of the app, for which a patent application has also been filed, is ready. Before it is released to users, the researchers are developing various initial applications to teach the system using comparative measurements. The researchers hope that the "HawkSpex (R) mobile" app could be launched on the market around the end of 2017.

However, comparative measurements are not always necessary. This is because some questions are not about specifying individual ingredients, but only about measuring different distributions of substances or materials. When buying a car, for example, the app compares whether the paint has exactly the same color in all areas - or whether it has been repainted.

Advertisement

Users expand the range of applications - similar to Wikipedia

The researchers are using an approach modeled on the online encyclopedia Wikipedia. "When the app comes onto the market at the end of 2017, dedicated users will be able to contribute to the big picture and create new applications, such as assessing the contamination of lettuce heads with pesticides, by teaching the system to deal with this type of issue," says Prof. Udo Seiffert, Head of Competence Field at the IFF. This means that users measure treated and untreated heads of lettuce of different varieties with the app and send the data to the IFF. Researchers check the measurements and activate the application for all users.

The app is also interesting in the commercial sector. It can be used to tap into areas where a precision measuring device would not be worthwhile, for example to check the quality of food, test the effectiveness of cosmetic products and for applications in agriculture.

  • Xing Icon
  • LinkedIn Icon
Advertisement
Advertisement

You might also be interested in

Advertisement

Special

Robotics&AUTOMATION 3/2021

Is an industrial robot without a cage really safe? What needs to be considered in the planning phase of human-robot collaborations? And what does a gecko have to do with robotics? The latest e-paper from Robotics&AUTOMATION provides the answers.

read more...
Advertisement
Advertisement
Advertisement

Fraunhofer IFF

Secure cobot installations

When it comes to safety, cobots place high demands on the knowledge of their human colleagues. A digital service developed by European research institutions aims to help companies share knowledge and carry out safe cobot installations.

read more...
Advertisement
Advertisement
Advertisement

Machine Learning

Apply according to plan

Procedures and processes in manufacturing can be optimized through the use of machine learning. However, machine learning is still far from being a standard tool. Fraunhofer researchers have now developed a generally applicable process model.

read more...
Subscribe to our newsletter
Advertisement
Back to home