최신 공간정보 기술

Recently, geospatial technologies in Korea have been rapidly advancing through the integration of high-resolution satellite imagery, drone platforms, spatial data analytics, and artificial intelligence (AI). In response to the growing complexity of climate change, urbanization, and disaster risks, geospatial technologies have emerged as essential tools in both public and industrial sectors, particularly in land cover classification, spatio-temporal prediction, disaster monitoring, and object detection. This special issue provides a comprehensive review of recent research articles, examining the current trends and technological advancements in applied geospatial technologies. In particular, the issue focuses on strategies for addressing class imbalance using loss functions, approaches to cope with spatio-temporal variability in kriging, the effectiveness of multi-modal data integration, and the practical implementation of deep learning using KOMPSAT imagery. Key findings highlight the importance of optimized loss function design for high-precision classification, modeling strategies based on spatio-temporal autocorrelation structures, limitations in resolution and expressiveness of auxiliary data, and the complementary potential of change detection and object-based classification. This review offers critical insights into the effectiveness and scalability of current geospatial technologies and provides foundational guidance for future policy and system design.