Multi-spectral UAV Data Analysis for Wheat Yield Prediction: A Deep Learning Approach
Abstract
Through the use of deep learning models and a comprehensive dataset of multispectral time series data collected by unmanned aerial vehicles (UAVs), this thesis aims to predict the grain yield of a wheat field. To find the best candidate for this prediction task, a variety of machine learning and deep learning models are investigated, including Gradient Boosting Regressor (GBR), Deep Neural Networks (DNNs), and Bidirectional Long Short-Term Memory (BiLSTM) networks.