Climate-Aware Plant Phenotyping and Crop Suitability Prediction Using CNN + XGBoost

Sustainable agriculture relies mostly on accurate crop yield predictions, based on various environmental factors, climatic conditions and plant phenotype. Usually, traditional methods struggle to integrate visual and structured data effectively. Artificial Intelligence and Machine Learning for plant phenotyping has become an emerging research area for producing high yield and sustainable agriculture. This research presents a hybrid model that combines Convolutional Neural Networks (CNN) a deep-learning model used for plant image feature extraction and XGBoost a machine-learning model used for structured climate and environment data analysis. This approach utilizes structured data to perform crop recommendation and fertilizer recommendation. On the next phase, Images are used to perform phenotype analysis and crop yield prediction. Our experimental results demonstrate improved yield forecasting, paving the way for advanced decision-support systems in agriculture.