Machine learning techniques for microwave food contamination detection
keywords ANTENNA DESIGN, MEASUREMENTS, SIMULATIONS, COMPUTATIONAL ELECTROMAGNETICS, ALGORITHMS, ELECTROMAGNETISM, FOOD INSPECTION, FPGA ACCELERATION, GPU, NVIDIA, ACCELERATOR, MACHINE LEARNING, MICROWAVE IMAGING
Thesis type EXPERIMENTAL AND SIMULATIONS, MASTER THESIS, MULTIDISPLINARY
Description Foreign body contamination in food is one of the major sources of complaints against food manufacturers, and can lead to injury, loss of brand loyalty and large recall expenses. Different technologies, such as X-ray or infrared techniques, are currently applied to detection systems used for food inspection, but physical contamination, with e.g. wood, plastic, metal and glass, is still present in food.
The main objective of this Master Thesis is to investigate the use of machine learning techniques to identify contaminated food products along the packaging line. The input data are measured via a custom microwave system placed on a testing packaging line available at the antenna and EMC lab, DET. Depending on the background of the Candidate, the acceleration of the selected ML algorithms with software or dedicated hardware design methods will be considered in the Thesis.
Main reference: J. A. Tobon Vasquez, R. Scapaticci, G. Turvani, M. Ricci, L. Farina, A. Litman, M. R. Casu, L. Crocco, F. Vipiana, “Non-invasive In-line Food Inspection via Microwave Imaging Technology”, IEEE Antennas and Propagation Magazine, Special issue on EM Imaging for Food, Vol. 62, No. 5, Oct. 2020, pp. 18-32, [DOI: 10.1109/MAP.2020.3012898]
Notes Expected duration: 6 months.
Deadline 03/12/2022 PROPONI LA TUA CANDIDATURA