DEEP LEARNING-BASED STEREO MATCHING FOR HIGH-RESOLUTION SATELLITE IMAGES: A COMPARATIVE EVALUATION

Utilizing satellite remote sensing data, we established a multi-season rice calendar dataset named ChinaRiceCalendar. It exhibits strong alignment with field observations collected by agricultural meteorological stations across China. ChinaRiceCalendar stands as a reliable dataset for investigating and optimizing the spatiotemporal dynamics of rice phenology in China, particularly in the context of climate and land use changes.

The point cloud of located aftershocks contains the information which can directly reveal the fault geometry and temporal evolution of an earthquake sequence. However, there is a lack of studies using state-of-the-art visual analytics methods to explore the data to discover hidden information about the earthquake fault. We present a novel interactive approach to illustrate 3-D aftershock point clouds, which can help the seismologist to better understand the complex fault system.