Federated Learning (FL) has emerged as a privacy-preserving alternative to centralized machine learning, enabling collaborative model training without sharing raw data. This is particularly valuable in smart agriculture, where farmers generate sensitive, distributed data that…