Monitoring and managing drought stress in wheat using artificial intelligence: A review of recent advances

Wheat (Triticum aestivum), as of the most important crop in global food security, is severely affected by drought stress a problem exacerbated by climate change and decreasing water resources. Traditional methods for drought assessment, due to limitations such as operational scale, low accuracy, and insufficient speed, fail to meet the demands of modern agriculture. In contrast, advanced technologies such as artificial intelligence, remote sensing, and the Internet of Things have enabled precise, real-time, and comprehensive analysis of environmental conditions in agricultural fields. Models including deep neural networks, such as Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM), alongside machine learning algorithms utilizing spectral, thermal, and RGB image data, have revolutionized the detection and prediction of drought stress. Tools such as drones, IoT-connected sensors, and satellite imagery collect extensive and diverse datasets, providing timely and effective decision-making opportunities for farmers. Furthermore, explainable artificial intelligence models, by clarifying the decision-making processes of algorithms, have enhanced trust in these systems. Although challenges such as data heterogeneity and climatic variability persist, innovative approaches like transfer learning and data fusion from multiple sources offer suitable solutions for achieving sustainable agricultural development. This review summarizes the applications of artificial intelligence in managing wheat crop under drought stress and outlines a novel perspective for more precise and efficient agriculture.

کلیدواژه ها:

Drought stress ، Artificial intelligence (AI) ، Remote sensing ، Internet of things (IoT) ، Smart irrigation