Feasibility Study of Terahertz Imaging for Non-Visible Defect Detection in Hazelnuts

The successful application of millimeter-wave (mmW) imaging to the quality assessment of soft fruits [1], such as apples and peaches, has motivated the exploration of higher-frequency techniques in the terahertz (THz) range. This study aims to evaluate the effectiveness of THz imaging in more challenging scenarios, particularly for certain types of seeds like hazelnuts. We investigate whether a THz imaging system can be used to assess the quality of hazelnuts affected by spoilage or fungal infections that are not visible to the naked eye, even after shell removal and prior to their use in food manufacturing processes, such as hazelnut cocoa cream production. Using reflection coefficient in the frequency range 105-175 GHz and rotational scanning, we evaluate the system's ability to detect spoilage or fungal infections. Tests include hazelnuts with and without defects, as well as samples with inserted metal intrusions. The initial results show the system can distinguish between different conditions, highlighting its potential for non-destructive food quality inspection.