A Computational Framework for Crop Yield Estimation and Phenological Monitoring

Accurate crop yield estimation is crucial for the agricultural industry, as it enables effective planning, resource management, and market forecasting. This study explores the application of machine learning techniques for yield estimation and phenological monitoring in fruit production, focusing on crops from Chile. To achieve this, a comprehensive dataset was compiled, including satellite imagery, climate data, high-resolution images of fruit trees, and corresponding yield records collected from multiple farms in the central valley of Chile. The dataset was meticulously preprocessed to eliminate noise and ensure consistency across diverse sources. Vegetation indices and climate data were integrated as contextual information to enhance the predictive power of the models. Various machine learning algorithms, including random forest and gradient boosting regressors, were trained and evaluated using cross-validation and performance metrics such as mean absolute error, root mean square error, and the coefficient of determination. The results demonstrate the effectiveness of the proposed approach in accurately estimating fruit yield. The inclusion of contextual information significantly improved the models’ accuracy. Practical examples from the Chilean central valley illustrate the adaptability of the developed methodology to different fruit crops. This study highlights the potential of machine learning techniques to transform yield estimation and phenological monitoring in fruit production, providing farmers with valuable insights to optimize resource allocation and enhance productivity.