Banana (Musa spp.) crops face severe yield and economic losses due to foliar diseases such as Moko disease and Black Sigatoka. In Ecuador, Moko outbreaks have increasingly devastated banana plantations, threatening one of the country’s most important export commodities and putting significant pressure on local producers and the national economy. Traditional field inspection methods are labor-intensive, subjective, and often ineffective for timely disease detection and containment. In this study, we propose an improved deep learning-based segmentation approach using YOLOv8 architectures to automatically detect and segment Moko and Black Sigatoka infections from unmanned aerial vehicle (UAV) imagery. Multiple YOLOv8 configurations were systematically analyzed and compared, including variations in backbone depth, model size, and hyperparameter tuning, to identify the most robust setup for field conditions. The final optimized configuration achieved a mean precision of 79.6%, recall of 80.3%, mAP@0.5 of 84.9%, and mAP@0.5:0.95 of 62.9%. The experimental results demonstrate that the improved YOLOv8 segmentation model significantly outperforms previous classification-based methods, offering precise instance-level localization of disease symptoms. This study provides a solid foundation for developing UAV-based automated monitoring pipelines, contributing to more efficient, objective, and scalable disease management strategies.

Banana (Musa spp.) crops face severe yield and economic losses due to foliar diseases such as Moko disease and Black Sigatoka. In Ecuador, Moko outbreaks have increasingly devastated banana plantations, threatening one of the country’s most important export commodities and putting significant pressure on local producers and the national economy. Traditional field inspection methods are labor-intensive, subjective, and often ineffective for timely disease detection and containment. In this study, we propose an improved deep learning-based segmentation approach using YOLOv8 architectures to automatically detect and segment Moko and Black Sigatoka infections from unmanned aerial vehicle (UAV) imagery. Multiple YOLOv8 configurations were systematically analyzed and compared, including variations in backbone depth, model size, and hyperparameter tuning, to identify the most robust setup for field conditions. The final optimized configuration achieved a mean precision of 79.6%, recall of 80.3%, mAP@0.5 of 84.9%, and mAP@0.5:0.95 of 62.9%. The experimental results demonstrate that the improved YOLOv8 segmentation model significantly outperforms previous classification-based methods, offering precise instance-level localization of disease symptoms. This study provides a solid foundation for developing UAV-based automated monitoring pipelines, contributing to more efficient, objective, and scalable disease management strategies.

La demanda constante y en aumento de alimentos como el maíz obliga a los productores a hacer uso de fertilizantes químicos, lo cual provoca degradación del suelo, contaminación por metales pesados, repercusión en las comunidades microbianas y pone en alto riesgo la salud humana. Las microalgas pueden ser una alternativa potencial como bioestimulantes y biofertilizantes, por lo cual, el objetivo de este trabajo fue determinar el efecto agronómico y productivo de la biofertilización a base de las microalgas Chaetoceros gracilis y Chlorella vulgaris en el cultivo de maíz (Zea mays L.) en Pueblo Viejo, Ecuador. Se utilizó un Diseño en Bloques Completamente al azar (DBCA), para determinar la eficiencia de los tratamientos se utilizó la Prueba de Tukey al 95% de probabilidad. Se estudiaron cuatro dosis de biofertilizantes a base de microalgas (5, 10, 15 y 20 mg planta^-1), tanto para Chaetoceros gracilis como para Chlorella vulgaris, además de usar un testigo experimental; se utilizaron 900 plantas de maíz por todo el experimento. Los resultados indicaron que, el tratamiento T4, donde se utilizó 20 mg planta^-1 de Chaetoceros gracilis como biofertilizante, presentó los mejores resultados en altura de planta (220.1 cm), longitud de mazorca (18.17 ± 2.02 cm), número de hileras por mazorca (17.88 ± 1.36), número de granos por mazorca (652.16 ± 58.23), peso de la mazorca con tusa (295.03 ± 8.91 g) y el mayor rendimiento (9,144.33 kg ha^-1). El uso de Chaetoceros gracilis en dosis de 20 mg planta^-1 permitió mejorar las variables agronómicas y productivas del cultivo.

Accurate banana yield prediction is essential for optimizing agricultural management and ensuring food security in tropical regions, yet traditional estimation methods remain labor-intensive and error prone. This study developed a predictive model for banana yield in Buena Fé, Ecuador, using Random Forest integrated with phenological data, soil properties, spectral technology, and UAV imagery. Data were collected from a 75.2 ha banana farm divided into 26 lots, combining multispectral drone imagery, soil physicochemical analyses, and banana agronomic measurements (height, diameter, bunch weight). A rigorous variable selection process identified six key predictors: NDVI, plant height, plant diameter, soil nitrogen, porosity, and slope. Three machine learning algorithms were compared through 5-fold cross-validation with systematic hyperparameter optimization. Random Forest demonstrated superior performance, with R² = 0.956 and RMSE=1164.9 kg ha⁻¹, representing only CV = 2.79% of mean production. NDVI emerged as the most influential predictor (importance = 0.212), followed by slope (0.184) and plant structural variables. Local sensitivity analysis revealed distinct response patterns between low- and high-production scenarios, with plant diameter showing the greatest impact (+74.9 boxes ha⁻¹) under limiting conditions, while NDVI dominated (−140.4 boxes ha⁻¹) under optimal conditions. The model provides a robust tool for precision agriculture applications in tropical banana production systems.

Accurate banana yield prediction is essential for optimizing agricultural management and ensuring food security in tropical regions, yet traditional estimation methods remain labor-intensive and error - prone. This study developed a predictive model for banana yield in Buena Fé, Ecuador, using Random Forest integrated with phenological data, soil properties, spectral technology, and UAV imagery. Data were collected from a 75.2 ha banana farm divided into 26 lots, combining multispectral drone imagery, soil physicochemical analyses, and banana agronomic measurements (height, diameter, bunch weight). A rigorous variable selection process identified six key predictors: NDVI, plant height, plant diameter, soil nitrogen, porosity, and slope. Three machine learning algorithms were compared through 5-fold cross-validation with systematic hyperparameter optimization. Random Forest demonstrated superior performance with R²=0.956 and RMSE=1164.9 kg ha⁻¹, representing only 2.79% of mean production. NDVI emerged as the most influential predictor (importance=0.212), followed by slope (0.184) and plant structural variables. Local sensitivity analysis revealed distinct response patterns between low and high production scenarios, with plant diameter showing greatest impact (+74.9 boxes ha-1) under limiting conditions, while NDVI dominated (-140.4 boxes ha-1) under optimal conditions. The model provides a robust tool for precision agriculture applications in tropical banana production systems.