A Review of Crop Yield Prediction Models based on Crop Phenology Using Satellite Imagery and Environmental Data

Accurately predicting crop yields is crucial for ensuring food security and promoting agriculture. In this review, we delve into the integration of satellite imagery and environmental data to develop models that can precisely predict crop yields based on their phenology. Our focus lies specifically on analyzing vegetation indices such as NDVI derived from satellite data, which offer insights into stages of crop growth. To demonstrate the applications of these models, we examine real-world case studies involving rice yield prediction in Pakistan and sunflower crop yield in Hungary, which are studied from other review articles. Our research methodology involves an analysis of existing literature at the intersection of sensing crop phenology, predictive analytics, and yield modeling. We also integrate recent advancements in ecological risk assessment, field delineation methods, and neural network classification of satellite images to enhance the comprehensiveness of our review. The research highlights phases of plant growth, including the stages of vegetative growth, flowering, and ripening, as significant factors for accurately predicting crop yields. The paper concludes by discussing the benefits and practical uses of models driven by satellite data to assist in making decisions in agriculture. Moreover, it recognizes the constraints and suggests areas for further investigation to improve the resilience and usefulness of these models in diverse agricultural settings and different weather conditions. To sum up, this review adds to the changing field of precision agriculture by providing an examination of models used to predict crop yields. It offers insights for policymakers, farmers, and stakeholders who are dedicated to advancing food security efforts.