多源数据支持的特大城市高时空分辨率人口分布估计与人口流动感知

特大城市建设为湖北省经济社会发展重要增长极，其人口密集，辐射广、环境复杂等显著特征，使精细化城市管理与服务困难，亟需高时空分辨率人口分布与人口流动信息提供精准科学依据和保障。针对城市人口研究信息片面、精度不足、应用固化瓶颈，如何系统科学研究精准、长效的人口分布估计与人口流动分析方法，构建场景关联、快速感知人口流动的应用方法，是依循以人为本、精准服务的需求牵引，实现湖北省经济社会协调持续发展关键科学问题。本项目一体化语义描述并按需构建多尺度人口分析单元，动态关联并融合多源数据；定量表达人口分布时空耦合规律，精准估计人口分布；构建层次清晰、多元关联、主题牵引的人口流动网络，揭示复杂人口流动时空模式与演化规律；语义理解城市场景，快速感知人口流动密集趋势，形成城市场景驱动、从人口分析单元到人口流动网络、从核心方法到应用框架一体化城市人口研究方法，为智能精准城市管理与信息服务提供科学方法和技术支撑。

The construction of megacities is an important way for the growth of economic and social development in Hubei Province. The significant megacity features of population agglomeration, high influence, and complex environment make the fine city management and services difficult. The distribution and mobility of the population with high spatiotemporal resolution can provide an accurate scientific basis and support. The issues of urban population research are incomplete information, insufficient accuracy, and inflexible application. Implementing the systematic and scientific research on the precise and long-term effective population distribution estimation and population mobility analysis methods, and constructing application methods for scene association and rapid perception of population mobility are the keys to address these issues, satisfying the requirement of people-oriented and precise services, achieving the coordinated and sustainable economic and social development of Hubei Province. In this project, we research the integrated semantic description, build population analysis units with demand-oriented and multi-scale, associate and fuse multi-source data dynamically. Then, we quantify the spatiotemporal coupling patterns of population distribution and estimate population distribution accurately. Further, we construct a population mobility network with multi-levels, multi-associations, and theme-oriented to reveal the spatiotemporal patterns and the laws in the evolution of complex population mobility. The last step is to semantically understand urban scenes, quickly perceive the changing trend of population mobility density. In this project, we develop an integrated research method for urban population that is driven by urban scene, from population analysis units to a population mobility network, and from core methods to the application framework. This research will support scientific methods and technical for intelligent and precise city management and information services.