Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

Thermographic techniques and adapted algorithms for automatic detection of foreign bodies in food

: Meinlschmidt, P.; Margner, V.


Cramer, K.E. ; Society of Photo-Optical Instrumentation Engineers -SPIE-, Bellingham/Wash.:
Thermosense XXV : 22 - 24 April, 2003, Orlando, Florida, USA
Bellingham, WA: SPIE, 2003 (SPIE Proceedings Series 5073)
ISBN: 0-8194-4932-6
ISBN: 978-0-8194-4932-0
Conference Thermosense <25, 2003, Orlando/Fla.>
Fraunhofer WKI ()

At the moment foreign substances in food are detected mainly by using mechanical and optical methods as well as ultrasonic technique and than they are removed from the further process. These techniques detect a large portion of the foreign substances due to their different mass (mechanical sieving), their different colour (optical method) and their different surface density (ultrasonic detection). Despite the numerous different methods a considerable portion of the foreign substances remain undetected. In order to recognise materials still undetected, a complementary detection method would be desirable removing the foreign substances not registered by the a.m. methods from the production process. In a project with 13 partner from the food industry, the Fraunhofer-Institut fur Holzforschung (WKI) and the Technische Universitat are trying to adapt thermography for the detection of foreign bodies in the food industry. After the initial tests turned out to be very promising for the differentiation of food stuffs and foreign substances, more and detailed investigation were carried out to develop suitable algorithms for automatic detection of foreign bodies. In order to achieve-besides the mere visual detection of foreign substances-also an automatic detection under production conditions, numerous experiences in image processing and pattern recognition are exploited. Results for the detection of foreign bodies will be presented at the conference showing the different advantages and disadvantages of using grey-level, statistical and morphological image processing techniques.