Yuxiang Lu

Accurate precipitation nowcasting up to two hours in advance is crucial for weather-sensitive decision-making, such as flood disaster warnings. Although deep learning-based radar echo extrapolation methods show advantages, they still face challenges, particularly in predicting local spatial variations and mitigating performance degradation with increasing prediction time horizon. This study proposes a new multi-scale spatiotemporal prediction LSTM model (MS²T-LSTM) to enhance precipitation nowcasting capabilities by fusing multi-scale spatial features through parallel convolutional structures and improving spatiotemporal reasoning through stacked LSTMs. Notably, multi-scale spatiotemporal feature gate is introduced into LSTM unit to transmit multi-scale spatiotemporal features of radar echo image sequences, mitigating the issue of time-dependent performance degradation. The loss function for model optimization is carefully selected via series of comparative experiments. The experiments show that MS²T-LSTM outperforms benchmark models in predicting precipitation events with different intensities. Specifically, in comparison to PredRNN++, the F1 score, Critical Success Index, Heidke Skill Score, and Probability of Detection for low-intensity precipitation threshold (5mm/h) improved by 2.3%, 3.5%, 13%, and 10.6%. This study demonstrates the effectiveness of integrating multi-scale spatiotemporal features to enhance precipitation forecasting, providing valuable insights for spatiotemporal prediction modeling and optimization.

The rapid growth of Earth observation technologies has resulted in over 2000 operational remote sensing satellites, collectively generating an exabyte-scale volume of data. However, despite the availability of large data-sharing platforms, global remote sensing imagery still faces challenges in seamless access, precise querying, and efficient retrieval. To address these limitations, this study introduces the concept of the “Digital Imagery Object” (DIO) and develops a unified identification framework for satellite remote sensing imagery. The proposed approach establishes a structured identification and parsing system based on core metadata, including data acquisition platforms and imaging timestamps. This enhances the consistency and standardization of multisource imagery encoding, enabling unified identification and interpretation under a common set of rules. The system’s feasibility and effectiveness were demonstrated through the integration and management of diverse global datasets, highlighting its ability to streamline multisource data workflows. By supporting standardized management and one-click parsing, this framework facilitates efficient imagery sharing and lays the foundation for its use as a tradable digital resource on the internet. The study offers a practical solution for addressing current challenges in remote sensing imagery management, paving the way for improved accessibility and interoperability of Earth observation data.

Deep learning technologies have significantly advanced object information extraction from remote sensing data in recent years, achieving broad application across various industrial sectors. However, information loss exists between remote sensing object raster segmentation and geographic information vector mapping, making it challenging to directly apply raster extraction results to vector mapping. This study, taking the automatic extraction of greenhouses based on remote sensing imagery as an example, proposes a vectorization method for remote sensing object segmentation based on frame field. This method bridges the gap between the object pixel segmentation process and the mask vectorization process through the frame field information outputted by the network, resulting in smoother and more regular vector extraction results. To validate the effectiveness of our framework, we introduce the first high-precision greenhouse vector boundary dataset. Extensive experiments demonstrate that our method significantly mitigates the information loss issue prevalent in traditional vectorization processes, achieving a 5.05% improvement in IoU, a 6.06% increase in recall, and a 5.54% reduction in maximum angular error compared to simple vectorization schemes. It outputs more regular greenhouse vector plots, where the precision of the frame field plays a crucial role in the final vectorization quality. This research offers a unique and practical solution, converting remote sensing object segmentation into vector maps.