Spatio-temporal predictive learning for time-series vegetation index

This study focuses on predicting vegetation indices in South Korea using MODIS sensor products, which provide a temporal resolution of 16 days and a spatial resolution of 500 meters. Data preprocessing, such as outlier removal, was conducted to ensure stability. Leveraging recent advancements in artificial intelligence, the study employed deep learning models for spatio-temporal video prediction. The predicted vegetation indices were compared with the original data using error and similarity matrices to verify accuracy. The results suggest that highly accurate vegetation indices can serve as valuable input for various studies monitoring vegetation changes, offering significant insights into climate change impacts on plant life. Acknowledgments This study was carried out with the support of ´R&D Program for Forest Science Technology (Project No. RS-2024-00404128)´ provided by Korea Forest Service(Korea Forestry Promotion Institute).