The study evaluates the feasibility of classifying Arabica and Robusta coffee beans using convolutional neural networks (CNNs). A custom CNN model (CNN_Coffee_Classifier) was developed and its performance was compared to that of three state-of-the-art architectures: MobileNet, ResNet50, and ResNet101. The models were trained both from scratch and using transfer learning on a dataset comprising 495 standardized images. The aforementioned dataset was derived from both public repositories and original photographs. Given the limited number of images in the training set, the models were trained using 10-fold cross-validation to ensure robust results. The custom CNN demonstrated an average accuracy of 75% (with 95% confidence interval of [0.68, 0.83]), while models employing transfer learning - particularly ResNet101 - exhibited superior performance, achieving up to 95% accuracy (with 95% confidence interval of [0.92, 0.96]). The findings substantiate the hypothesis that CNNs, particularly when integrated with transfer learning, provide a robust methodology for automated coffee variety classification, even when operating with a modest dataset.
As the beer market diversifies, novel types of beers are appearing. An interesting possibility is the use of functional plant materials, which may improve physicochemical properties of beers and contribute to creating a beverage with intriguing sensory qualities. The aim of this study was to evaluate the contribution of selected plant additives to the physicochemical and sensory profile of beer. The study was conducted on brewers’ wort with aniseed fruit, rosehip, cistus, moringa oleifera, Jerusalem artichoke, and ashwagandha at two doses (0.1% and 0.2%). A beer without any additions was used as a control sample. The weight changes of beers during fermentation were analyzed. Alcohol content, real extract, free amino nitrogen, titratable acidity, color, pH, and sensory analysis were performed on the produced beers. The highest alcohol content was obtained in beers with Jerusalem artichoke (0.2%), while the lowest was found in a beverage with ashwagandha (0.2%). Beers with evaluated ingredients were characterized by a lower titratable acidity, compared to the control sample. The sensory analysis revealed that the beers with Jerusalem artichoke (0.2%) and moringa leaves (0.2%) found the greatest approval among panelists.
Sausages, one of the oldest forms of meat processing, rely heavily on the quality of casings to achieve desired texture, appearance, and stuffing efficiency. This study aimed to improve the hydration process of natural pork casings for Calabrian sausage production by employing phosphoric acid as a substitute for trisodium phosphate dodecahydrate. Key quality control tools, including the Ishikawa diagram, 5W1H tool, and PDCA cycle, were utilized to identify and address production inefficiencies. The implemented modifications led to significant advancements: stuffing yield increased from 0.87 to 1.0 kg/m over seven months, surpassing the recommended standard of 0.91 kg/m, and casing consumption decreased by 10.39%. These improvements not only enhanced operational efficiency but also reduced casing costs by 10.37%. The findings confirm the viability of phosphoric acid as an effective, cost-efficient alternative in the hydration process, paving the way for improved sausage production standards.
This study analysed the effect of applying gaseous ozone for 15 and 30 minutes on mechanical and chemical properties of six Japanese quince variety. Among the tested varieties, Gold Calif demonstrated superior mechanical durability, as it required the highest levels of force and energy to puncture and underwent the most substantial deformation. Extended storage of quince fruits for 60 days resulted in a marked decline in vitamin C levels, with an average reduction of 19.8%. Notably, the Rasa variety retained the greatest concentration of vitamin C, measuring 120.29 mg·100 g⁻¹. Exposure to gaseous ozone significantly boosted the total phenolic compound content; a 30-minute treatment led to a 6.2% increase when compared to untreated, fresh samples. After two months of storage, phenolic levels dropped by 6.5% relative to the initial fresh fruit values. Additionally, Japanese quince fruits exposed to ozone for 30 minutes exhibited enhanced antioxidant properties, reflected by a 5.8% rise in DPPH activity and a 5.5% increase in FRAP values over non-treated fruits. On the basis of the study, gaseous ozone can significantly influence chemical composition, increasing the mechanical resistance and health-promoting value of fruits.
Recent research has focused on enhancing the overall quality of gluten-free products. This study evaluated the physicochemical properties of biofortified yellow maize flour (BMF) supplemented with Brachystegia eurycoma flour (BEF) at varied amount (2.5, 5, 7.5 and 10 %), and sensory qualities of their gluten-free cookies. Sodium carboxyl methylcellulose (SCMC) served as reference hydrocolloid while refined wheat flour cookies served as reference cookies. The addition of BEF significantly (p < 0.05) decreases the functional properties (bulk density, water absorption capacity, soluble sugar and amylose content) of BMF. BMF blends has higher (p < 0.05) pasting properties than the 100% BMF except the pasting temperature. Among all the flour samples, 100% BMF has the highest level of crude protein (12.78±0.08 %) and fiber (2.94 ± 0.12 %) but the lowest total carbohydrate (74.76 ± 0.14 %). The cookies made from 100% RWF have the lowest value for appearance (7.2 ± 1.02) and color (7.2 ± 0.75) and 2.5%BEF-97.5% BMF cookies have the highest over acceptability value (8.40 ± 1.20). Hence, the physicochemical properties of BMF as well as their cookies sensory qualities can be enhanced by supplementing with BEF.
