This paper investigates the feasibility of using 77-81 GHz frequency-modulated continuous-wave (FMCW) millimeter-wave (mmWave) radar for non-contact chicken egg quality monitoring. The study is organized as two complementary arms. The first arm is a cross-sectional destructive-validation study in which different eggs of the same batch of 24 eggs were measured over 8 storage days and linked to mass, albumen and yolk indices, pH, candling observations, and breakout-derived quality scores. The second arm is a longitudinal nondestructive study in which the same three eggs were tracked for 8 days and linked to non-destructive ground truth data such as: mass loss and candling-derived air-cell metrics. Raw radar data were acquired using a TI AWR1843EVM BOOST with a DCA1000EVM at a fixed 0.40 m antenna-to-egg distance from multiple side and vertical views. A target-locked processing pipeline, quality-control filtering, egg-level feature aggregation, and machine-learning regression were used to connect radar features to ground truth. The cross-sectional arm provided weak-to-moderate supporting evidence, whereas the longitudinal arm produced strong monotonic radar-to-ground-truth associations and strong leave-one-egg-out prediction results. The best longitudinal models achieved values of 0.899 for air-cell area, 0.918 for air-cell depth, 0.917 for mass-loss percentage, and 0.884 for a nondestructive core score. These results position high-frequency FMCW mmWave radar as a promising solution for egg quality monitoring in storage environments such as farms, food-processing facilities, and retail backrooms.
Onta Shahriar (Thu,) studied this question.