Ahmad Syafik Suraidi, S.1,2, Aimrun, W.2,3, Leifeng, G.4, Samsuzana, A. A.2 and Mui, Y. W.3
Abstract
This review explores how remote sensing and artificial intelligence (AI) are being used to detect crop stress, focusing on recent research from Malaysia and China. The findings show that China is leading the way, with widespread use of advanced AI models like convolutional neural networks (CNNs), Vision Transformers (ViTs), and attention-based systems. These models are supported by large datasets and strong government backing, making it possible to detect crop stress early and accurately across large areas. Malaysia, on the other hand, is still in the early stages. Most studies are limited to small-scale trials using drone imagery and more traditional machine learning models such as Support Vector Machines (SVM) and Random Forest (RF). Despite these limitations, the results have been promising, especially for key crops like oil palm and rice. However, Malaysia faces challenges including a lack of localised data, limited AI infrastructure, and minimal policy support. To move forward, future efforts should focus on developing locally relevant models, using multiple types of data together, and fostering collaboration between researchers, policymakers and farmers. There’s also a need to make sensing technologies more affordable and accessible to those working on the ground. In summary, while China has already laid a strong foundation for AI-powered crop stress detection, Malaysia has the potential to catch up provided that there is strategic investment in research, infrastructure and farmer engagement. With the right support, both countries can strengthen their agricultural resilience and better prepare for climate-related challenges.
Keywords: remote sensing, crop stress detection, AI algorithms, precision agriculture