Improved open-ended coaxial probe for temperature-dependent permittivity measurements of foodstuff at radio frequencies

Emerging non-invasive microwave and millimeter-wave imaging technologies for food inspection

The microwave and millimeter-wave (MMW) imaging technology is gaining increasing interest for food inspection. It allows for noninvasive, contactless, and fast scanning capabilities, while being cost-efficient and safe to human. This review paper introduces the fundamentals in the interaction of electromagnetic wave with food materials and the current MMW sensing and imaging systems used for foods. Then we present emerging technologies in MMW imaging for inspecting food quality and safety, aiming to meet the modern food industry's demand. According to the most recent technological advancements, it is expected that high-performance antenna, ultrawide bandwidth signal generation, nano-scale semiconductor technologies, radio frequency identification with inductance-capacitance resonator, and machine learning could significantly enhance the capabilities of MMW imaging systems for food inspection.

Cost-Effective Open-Ended Coaxial Technique for Liquid Food Characterization by Using the Reflection Method for Industrial Applications

A cheap technique based on an open-ended coaxial probe together with a vector network analyzer was set up. The vector network analyzer NanoVNA, a very tiny handheld device, is the affordable component that gives the instrumental chain a cost-effective perspective. The open-ended coaxial probe is a cable with an SMA gold-plated termination. User-friendly programs can be used to calibrate the instrument, carry out the measurements, and save data on PC. Simple liquid solutions (sodium chloride, citric acids, and saccharose) and more complex liquid food (milk, egg products, and fruit juice) were investigated. In addition, the temperature on the electric measurement of milk was measured to evaluate a possible influence for refrigerated storage products. The reflection parameters, such as the real and imaginary parts of S11, were used to build univariate and multivariate models. The best results in terms of coefficient of determination and related error were 0.997 (RMSE 0.05%) for sodium chloride and 0.965 (RMSE 0.71 °Brix) for fruit juice considering the univariate model, and 0.997 (RMSE 0.04%) for sodium chloride and 0.981 (RMSE 4.44%) for yolk using multivariate analysis. The proposed solution is non-destructive, cheap, rapid, and very attractive for potential lab and industrial applications.