Nondestructive Wheat Moisture Detection System With Metasurface Plane Wave Antenna and De-Embedding Algorithm

The precise measurement of wheat moisture content is a key issue to ensure the quality and safety of wheat. However, high-precision, rapid, and nondestructive detection of wheat moisture content remains challenging. In this article, we proposed a system with single-layer low-frequency metasurface plane wave lens antenna (MPWLA) for nondestructive wheat moisture detection in the frequency range of 5.8–5.9 GHz. Based on the propagation characteristics of plane waves in multilayer dielectric, antenna and container error de-embedding algorithms are proposed to reduce the measurement errors introduced by the parameters of the antenna and container during the measurement process. According to the de-embedded scattering parameters of wheat, its dielectric constant and loss factor are obtained. The calculated dielectric constant and loss factor showed a good linear relationship with moisture content of wheat at 25 °C (±0.5 °C). Based on the relationship between the dielectric properties and moisture content of wheat, a linear regression equation is established to predict the moisture content of wheat. Experimental results show that the root-mean-square error (RMSE), mean absolute error (MAE), and maximum relative error (MRE) of the predicted values were 0.19%, 0.16%, and 4.0%, respectively.