Rapid and high-precision wheat moisture detection is of great significance in ensuring wheat quality throughout transportation, storage, and procurement. The microwave method can achieve rapid detection of wheat moisture content, but the detection precision is influenced by spherical wave multipath, transmission diffraction effects, container, and wheat density. To improve the precision of detecting wheat moisture content, this paper presents a detection system that utilizes the 3D-printed dielectric metasurface lens antenna (DMLA). By utilizing the characteristic of the DMLA transforming spherical waves into plane waves, the error caused by spherical waves is reduced. The scattering parameters of wheat at varying moisture levels (6.2%-15.6%) and different densities are obtained using antenna and container error calibration algorithms in the frequency range of 23.6-24 GHz. An inverse algorithm is used to analyze the scattering parameters of wheat to determine the dielectric properties. The relationships between the dielectric properties, moisture content, and density are analyzed, and a densityindependent calibration function is proposed to reduce the effect of density on the precision of moisture content detection. A linear regression model was established based on the density-independent calibration function, with a coefficient of determination (R2) of 0.992. In the results of predicting moisture content with the calibration function, the mean absolute error (MAE) is 0.143% and the root mean square error (RMSE) is 0.178%. The experimental results demonstrate the system''s ability to achieve high-precision moisture content detection, with significant application potential in rapid grain quality detection, online monitoring, as well as grain drying and processing equipment.
(8.0+极速模式)