Physical classification is the procedure adopted by the rice unloading, delivery, storage, and processing units for the commercial characterization of the quality of the grains. This step occurs mostly by the conventional method, which demands more time and specialized labor, and the results are subjective since the evaluation is visual. In order to make the operation faster, more accurate, and less dependent, non-destructive technologies and computational intelligence can be applied to characterize grain quality. Therefore, this study aimed to characterize and predict the quality of whole, processed rice grains, as well as classify any defects present. This was achieved by sampling from the upper and lower points of four silo dryers with capacities of up to 40,000 sacks. The grain samples had moisture contents of 16%, 17%, 18%, and 19% and were subjected to drying-aeration until reaching 12% moisture content (w.b.). Near-infrared spectroscopy technology and Machine Learning algorithm models (Artificial Neural Networks, decision tree algorithms Quinlan’s algorithm, Random Tree, REPTree, and Random Forest) were employed for this purpose. By analyzing Pearson’s correlation statistics, a strong negative correlation (R² = 0.98) was found between moisture content and the yield of whole grains. Conversely, a strong positive correlation (R² = 0.97) was observed between moisture content and classified physical defects across the various characterized physicochemical constituents. These findings indicate the effectiveness of near-infrared spectroscopy technology. The Random Tree model (RandT) successfully predicted the grain quality outcomes and is therefore recommended as the model of choice, obtained Pearson’s correlation coefficient (r = 0.96), mean absolute error (MAE = 0.017), and coefficient of determination (R² = 0.92). The results obtained here reveal that the combination of near-infrared spectroscopy technology and Machine Learning algorithm models is an excellent non-destructive alternative to manual physical classification for characterizing the physicochemical quality of whole and defective rice grains.

Several studies have been conducted for developing new dryers and improving existing ones to obtain a higher yield and maintain the quality of agricultural products. Thus, the objective of this study was to model and validate a new concept of a silo–dryer–aerator with a drying chamber and four storage compartmented cells in a fixed and thick layer with the application of high and low air-drying temperatures. To fit the experimental data to a mathematical model, nonlinear regression analysis was performed. To assess the fluid from the heated and cooling air in the silo–dryer–aerator, computational fluid dynamics was used. The experimental prototype of the fixed-bed dryer enabled a uniform distribution of the heated air in the drying cells. There was a uniform distribution of the air in the drying cells of the silo–dryer–aerator. The drying system achieved high operational performance and quality for soybeans. The different mathematical models could predict and adjust to the drying conditions in a fixed and thick layer. It was concluded that the drying model developed and validated is suitable for the new concept of a silo–dryer–aerator with a drying chamber and segregated storage to be applied to soybean drying. Graphical Abstract

This study aimed to evaluate the physiological performance of white clover seed lots of different tegument colours, besides to analyses the electrical conductivity test methodology with different seeds number, water volume and soaking periods. The experiment was developed at the Seed Didactic Laboratory in the Agronomy College “Eliseu Maciel” at the Federal University of Pelotas, in Pelotas-RS, Brazil. White clover seeds were manually separate, composing four lots of different coloured seeds: yellow, orange, brown and mixed. White clover seeds of yellow colour tend to present greater physiological potential through its germination and vigour. The electrical conductivity test was not efficient on identifying different vigour levels in white clover seed lots.

The quality of rice depends on physical, nutritional, and sensory attributes. However, in industrial practice, quality is predominantly based on physical characteristics evaluated by the conventional method for categorizing commercial atches. In this context, the present study aimed to characterize the physical quality and proximate composition and to classify commercial batches of polished white rice using machine learning (ML) algorithms based on spectral data. Individual samples (healthy grains and physical defects) and samples from commercial batches (Type 1 to Type 5 and Off-Type) were analyzed and prepared in accordance with current legislation. Spectral data were obtained using NIR and hyperspectral measurements covering the VIS/NIR/SWIR regions, and proximate composition was determined for moisture (MOI), starch (ST), protein (PRO), lipids (LIP), fiber (FIB), and ash (ASH). Multivariate analyses and ML classification models were applied to evaluate differences among grain types and commercial categories and to assess the discriminatory capacity of spectral information. The results showed that including physicochemical attributes to evaluate the quality of commercial batches simplifies the commercial categories currently used. For spectral behavior, batches classified as Type 1 and Type 2 showed low reflectance in the NIR and SWIR regions, suggesting greater interaction of radiant energy with compounds associated with nutritional and sensory quality. The MLP, LGBM, CAT, XGB and RF models performed best for the classification of commercial white polished rice batches, with metrics above 95%. The SWIR region, especially the 2173 nm spectral point, demonstrated high discriminatory power. In conclusion, the application of machine learning models based on VIS/NIR/SWIR spectroscopy proved highly efficient for classifying commercial batches of polished white rice, integrating physical and physicochemical attributes of the grains.