Home Bots & Brains Machine Learning detects contaminants in food

Machine Learning detects contaminants in food

by Pieter Werner

A team of students from the Alta Scuola Politecnica, a joint two-year excellence program of the Politecnico di Milano and Politecnico di Torino, in collaboration with Wavision s.r.l., a Politecnico di Torino spin-off, are developing an innovative solution to address food contamination in packaged products. The project, known as Wavisionproject, employs microwave sensors and Machine Learning algorithms to detect contaminants that could compromise product quality and consumer health.

In the face of increasing volumes of packaged food production and the corresponding rise in contamination risks, this technology aims to enhance consumer safety and maintain customer confidence in food brands. The principle behind the Wavisionproject’s technology is novel: it leverages the dielectric contrast between the food product and any foreign bodies, detected through alterations in microwave signals. This approach differs significantly from current methods like X-ray detection, which primarily identify contaminants based on density differences.

The project is currently progressing in five key areas. The first goal is to refine the prototype’s setup using cost-effective components without sacrificing efficiency. Additionally, the team plans to expand the dataset to bolster robustness tests and enhance the capability to identify contaminants. They also intend to conduct a theoretical analysis of biological contaminants to pinpoint the most prevalent types. Advanced Machine Learning models are being evaluated to improve detection accuracy and reduce calibration time. Lastly, the introduction of a trained Neural Network model is anticipated to manage anomalies in the industrial production chain effectively.

Although the project has made considerable strides, certain research questions remain unresolved. Nevertheless, the innovative approach of the Wavision team aims to address several limitations of existing contaminant detection methods, potentially leading to cost reductions, waste minimization, and enhanced consumer safety. This technological advancement could be a significant milestone in ensuring safer products and bolstering consumer confidence in the food industry.

The Alta Scuola Politecnica (ASP) is a prestigious two-year program established in 2004 by Politecnico di Milano and Politecnico di Torino. It annually selects 150 talented individuals from various countries, with more than a third being women, based on merit and skills from those who have completed their Bachelor’s degree and are enrolling in a Master of Science program at either institution. The program operates primarily in English.

Misschien vind je deze berichten ook interessant