The accurate classification of banana ripeness is essential for optimising agricultural practices and enhancing food industry processes. This study investigates the classification of banana ripeness using Machine Learning (ML) and Deep Learning (DL) techniques. The dataset consisted of 1565 high-resolution images of bananas captured over a 20-day ripening period using a Canon EOS 90D camera under controlled lighting and background conditions. High-resolution images of bananas at different ripeness stages were classified into ‘unripe’, ‘ripe’, and ‘overripe’ categories. The training set consisted of 1398 images (89.33%), and the validation set consisted of 167 images (10.67%), allowing for robust model evaluation. Various ML models, including Decision Tree, Random Forest, KNN, SVM, CNN, and VGG models, were trained and evaluated for ripeness classification. Among these, DL models, particularly CNN and VGG, outperformed traditional ML algorithms, with the CNN and VGG achieving accuracy rates of 90.42% and 89.22%, respectively. These rates surpassed those of Decision Trees (71.86%), Random Forests (85.63%), KNNs (86.83%), and SVMs (89.22%). The study points out the importance of dataset quality, model selection, and preprocessing techniques in achieving accurate ripeness classification. Practical applications of these results include optimised harvesting practices, enhanced post-harvest handling, improved consumer experience, streamlined supply chain logistics, and automation in sorting systems. These results confirm the feasibility of using deep learning for the automated classification of ripening stages, with implications for reducing postharvest losses and improving supply chain logistics. These findings have significant implications for stakeholders in the banana industry, from farmers to consumers, and pave the way for the development of innovative solutions for banana ripeness classification.

This study set out to determine the physical and mechanical properties of edible films made from native starch extracted from non-exportable Ecuadorian bananas (Musa sapientum L.). The following treatments were applied: 0.5% starch - 0.25% glucose (T1), 0.5% starch - 0.5% glucose (T2), 1% starch - 0.25% glucose (T3), and 1% starch - 0.5% glucose (T4). Variables such as physicochemical properties, amylose content, maximum gelation temperature, and the morphology of native starch granules were assessed. Variables in the film, such as thickness, opacity, film solubility, and tensile strength, were also measured. The fruit used in this study was composed of 88.55% starch (dry basis, 15.35% amylose). The starch granules had a dominant ovoid morphology with an average diameter of 25 - 30 μm. The peak gelatinization temperature was 73.80 °C. The maximum value for viscosity was 3,836 cP at 86.8 °C. Films that presented good malleability, thickness, transparency, tensile strength, and low solubility were obtained from T4, which turned out to be the best treatment.

Los hidróxidos dobles laminares (LDH, por sus siglas en inglés), también denominados como Hidrotalcitas (HTs), son una clase de materiales arcillosos bidimensionales conocidos por su estructura laminar única, composición versátil y propiedades de intercambio iónico. En este trabajo se reporta la determinación de la relación estequiométrica de arcillas tipo HTs sintetizadas en el laboratorio por el método de co-precipitación. El contenido de metales di y tri-valente que conforman las estructuras de las arcillas fue cuantificado por espectroscopia de absorción atómica por llama. Se sintetizaron dos tipos de HTs, Ni (HT-Ni) y Mg (HT-Mg); donde de acuerdo con los cálculos realizados la formula estequiométrica para la primera arcilla es Ni0,82Al0,18(OH)2(NO3)0,18 y Mg0,80Al0,20(OH)2(NO3)0,20 para la segunda, resultados que concuerdan perfectamente con la literatura. Las arcillas sintetizadas fueron caracterizadas por difracción de rayos X, espectroscopia FT-IR y microscopia electrónica de barrido. Los patrones de difracción de rayos X confirmaron la formación de HTs de tamaño de partícula nanométrico, donde la HT-Mg sintetizada posee mejores propiedades cristalinas.

"EEl presente trabajo tuvo como objetivo fundamental evaluar el potencial tecnológico del almidón de banano como ingrediente alternativo en la elaboración de pastas alimenticias, analizando sus propiedades físico-químicas, funcionales y su impacto en la calidad del producto. Se estudiaron las variedades Cavendish, Filipino, Valery y Orito, a quienes se les extrajo el almidón mediante el método húmedo (Bello - Pérez y col. 2002). La apariencia microscópica del gránulo de almidón fue determinada mediante microscopía electrónica de barrido JSM.6300 Scanning."

The base for this research work was set in the study of the interaction between legume flours and tuber starch blends to be used for sweet biscuits formulations and the successive evaluation of quality parameters. As main inputs the following were used: legume flours, manioc starch, margarine, sugar, eggs, and baking soda. Basic utensils: plastic and aluminum trays, biscuits shapers, cutters, other biscuit items. Analytical balance, mixer, and oven. Methods: Establishment of experimental units, analysis of manioc starch and legume flours, moisture analysis, ash content determination, rheological analysis of blends, gluten analysis in the blend of starch and legumes flours, dough formulation and biscuit making, textural a color analysis, sensory analysis of biscuits, statistical analysis. The results of moisture and ash analyzes in the inputs for blends were within the limits established of 14,5%. Ash content found was 2.3% in starch, 6.3% in soybean meal, and 2.9% in chickpea flour. Gluten content was around 10 mg/g in average in the 3 samples. Concerning rheological characterization, water absorption for the 3 blends was relatively low. Values resulted for the torque were within 1.1 Nm; time required to reach this value did not exceed two minutes of kneading stability. The 3 blends denoted low amylasic activity, as well as low starch damage. Gel stability results were due to the high amylasic activity occurred in manioc starch. The retrogradation intensity predicted that the 3 blends would have a long shelf life since retrogradation processes were low for the 3 cases. Hardness presented in samples was due to protein supply provided by soybean meal. According to the sensory analysis average results for texture, flavor, and color attribute, differences were found. The ANOVA for sensory evaluation in biscuits showed differences regarding texture, flavor, and color attributes in biscuits. The moisture values obtained for each biscuit sample were similar to each other and did not exceed 3%. Tuber starch and legume flour incorporation in formulation to produce gluten-free biscuits contributes to the development of a flavorsome product and provides an alternative for people with gluten intolerance. The results obtained may be used as the basis for the production of gluten-free biscuits.

El objetivo de la presente investigación fue, definir un modelo predictivo para la medición del estado de madurez de Musáceas sp., cultivadas al suroeste del Ecuador, usando algoritmos de Machine Learning. La metodología utilizada en la presente investigación, se dividió en dos etapas: la primera etapa, consistió en obtener los datos de las características fisicoquímicas, a través de los grados Brix y acidez titulable (AOAC,2009); la segunda etapa, mediante el uso de algoritmos de Machine Learning, se procesaron dos sets de datos, tanto para el entrenamiento, como para el Test. El presente trabajo llegó a la conclusión de que, se puede predecir el estado inmaduro, maduro o sobre maduro del banano, a través de un modelo predictivo, con una exactitud del 91 %.