This study presents an application-oriented CFD–FEM integrated workflow for analyzing chamber-side field non-uniformity and kelp-side thermal response during infrared drying. A three-dimensional steady-state CFD model was first established to reconstruct the chamber temperature, airflow, and incident radiation fields under certain operating conditions. Numerical consistency was checked through residual convergence; monitored variables; and global mass balance, for which the net mass imbalance was 0.004077 kg s⁻¹. The reconstructed mid-plane fields were then processed in MATLAB to extract the mean values, extrema, and coefficients of variation, and a composite objective function was used to screen the tested operating conditions in terms of field uniformity, temperature band compliance, and overheating risk. The thermal loads obtained via CFD were subsequently mapped onto a kelp finite element model to simulate the transient surface temperature evolution. Among the tested cases, case01 yielded the lowest composite objective value (J = 0.4535); its mapped kelp response showed a mean surface temperature of 62.23 °C and a maximum temperature of 63.57 °C at the exported time step. The proposed framework is therefore suitable for thermal response assessment and operating condition screening, although determining the full drying behavior still requires coupling of moisture transfer and improved experimental validation.

To overcome the low accuracy of conventional methods for estimating liquid volume and food nutrient content in bowl-type tableware, as well as the tool dependence and time-consuming nature of manual measurements, this study proposes an integrated approach that combines geometric reconstruction with deep learning–based segmentation. After a one-time camera calibration, only a frontal and a top-down image of a bowl are required. The pipeline automatically extracts key geometric information, including rim diameter, base diameter, bowl height, and the inner-wall profile, to complete geometric modeling and capacity computation. The estimated parameters are stored in a reusable bowl database, enabling repeated predictions of liquid volume and food nutrient content at different fill heights. We further propose Bowl Thick Net to predict bowl wall thickness with millimeter-level accuracy. In addition, we developed a Geometry-aware Feature Pyramid Network (GFPN) module and integrated it into an improved Mask R-CNN (Region-based Convolutional Neural Network) framework to enable precise segmentation of bowl contours. By integrating the contour mask with the predicted bowl wall thickness, precise geometric parameters for capacity estimation can be obtained. Liquid volume is then predicted using the geometric relationship of the liquid or food surface, while food nutrient content is estimated by coupling predicted food weight with a nutritional composition database. Experiments demonstrate an arithmetic mean error of −3.03% for bowl capacity estimation, a mean liquid-volume prediction error of 9.24%, and a mean nutrient-content (by weight) prediction error of 11.49% across eight food categories.

To overcome the low accuracy of conventional methods for estimating liquid volume and food nutrient content in bowl-type tableware, as well as the tool dependence and time-consuming nature of manual measurements, this study proposes an integrated approach that combines geometric reconstruction with deep learning–based segmentation. After a one-time camera cali-bration, only a frontal and a top-down image of a bowl are required. The pipeline automatically extracts key geometric information, including rim diameter, base diameter, bowl height, and the inner-wall profile, to complete geometric modeling and capacity computation. The estimated parameters are stored in a reusable bowl database, enabling repeated predictions of liquid vol-ume and food nutrient content at different fill heights. We further propose Bowl Thick Net to predict bowl wall thickness with millimeter-level accuracy. In addition, we developed a Geome-try-aware Feature Pyramid Network (GFPN) module and integrated it into an improved Mask R-CNN framework to enable precise segmentation of bowl contours. By integrating the contour mask with the predicted bowl wall thickness, precise geometric parameters for capacity estima-tion can be obtained. Liquid volume is then predicted using the geometric relationship of the liq-uid or food surface, while food nutrient content is estimated by coupling predicted food weight with a nutritional composition database. Experiments demonstrate an arithmetic mean error of −3.03% for bowl capacity estimation, a mean liquid-volume prediction error of 9.24%, and a mean nutrient-content (by weight) prediction error of 11.49% across eight food categories.