Code for UAV-Based RGB and Multispectral Mango Leaf Disease Detection

This repository contains the research code used to train, validate, and evaluate the models reported in the associated manuscript on UAV-basedmango leaf disease detection using RGB and multispectral imagery. The repository includes Python scripts for model training, validation, and testing, along with configuration files and dependency information.The code corresponds to the experimental pipeline used to generate the results reported in the manuscript. Data preprocessing was performed using third-party tools and publicly available implementations, which are cited and described in therepository documentation and in the associated research article. No custom preprocessing code was developed beyond these steps. The dataset used in this study is publicly available via a separateZenodo record (https://doi.org/10.5281/zenodo.18424647). This repository contains research code only. Deployment, hardware-specific optimization, and product-level implementations are not included.