Pest and disease outbreaks remain a major constraint on agricultural productivity, making reliable crop monitoring a priority for precision agriculture. In this context, remote sensing and artificial intelligence (AI) have emerged as complementary tools for improving the detection and monitoring of phytosanitary problems across spatial scales. This systematic review synthesizes recent advances in the integration of remote sensing and AI for crop pest and disease monitoring, with emphasis on sensing platforms, spectral information, AI approaches, and barriers to field adoption. Following PRISMA guidelines, literature was retrieved from Scopus and Web of Science, focusing on peer-reviewed studies published between 2019 and early September 2025. A total of 147 studies met the inclusion criteria for qualitative analysis. The reviewed literature shows increased use of UAV and satellite observations, supported by RGB, multispectral, hyperspectral, thermal, and proximal sensing configurations. Among AI methods, convolutional neural networks remain the dominant approach, while machine learning models such as random forest and support vector machines continue to be widely applied, particularly in spectral datasets. Across the reviewed studies, spectral indices and hyperspectral features are recurrent inputs for early stress detection, whereas recent developments increasingly incorporate segmentation, object detection, transformers, and multimodal data fusion. Despite these advances, large-scale implementation remains limited by data availability, cost, variability across field conditions, and insufficient external validation. Overall, the evidence indicates that the integration of remote sensing and AI is advancing toward scalable and operational crop health monitoring systems, although greater emphasis on robustness, transferability, and field deployment is still needed.
Valdiviezo et al. (Mon,) studied this question.