车辆轨迹与遥感影像多层次融合的城市级路网生成研究

城市路网是发展智慧城市的基础与支撑信息。考虑到在生成城市路网时，车辆轨迹与遥感影像具有覆盖广、更新快、成本低、互补性强等突出优点，本项目设计交叉口优先的城市路网构建策略，在多个层次上分析与融合两者关于路网的高判别性互补特征，将复杂的可导航城市路网生成问题分解为三项逐层递进的研究内容：1）发展轨迹-影像协作的道路交叉口多模集成识别方法,2）设计交叉口信息引导的路段中心线分类提取方法,3）提出多元特征融合的路网导航信息计算方法。最终形成车辆轨迹与遥感影像融合的城市道路及其导航信息的高现势、低成本、广覆盖获取技术，实现城市级可导航路网的分而治之生成,不仅可以极大提高城市路网识别的完备性与精确性，而且能够有效降低城市路网构建的复杂度，从而为集成车辆轨迹与遥感影像的可导航路网生成提供新的研究思路与方法，在智能交通、位置服务等领域具有重要的学术与应用价值。

Urban road network is the foundation and supporting information for the development of smart city. Considering the outstanding advantages of vehicle trajectories and remote sensing images in generating urban road network, such as wide coverage, fast updating, low cost and strong complementarity, this project designs an intersection-first strategy to construct urban road network. By analyzing and integrating high discrimination and complementary characteristics of the two sources on describing road from multiple levels, the problem of complex navigable urban road network generation is divided into three progressive research issues: 1) develop the multi-mode integrated method for the recognition of road intersection based on trajectory-image cooperation, 2) design the hierarchical extraction method of road segment guided by intersection information, and 3) propose the calculation method of road network navigation information based on multi-feature fusion. Finally, based on the fusion of vehicle trajectory and remote sensing image, the technology of up-to-date, low cost and wide coverage for the acquisition of urban road and navigation information is formed, and the generation of city-level navigable road network is realized. It can not only greatly improve the completeness and accuracy of the construction of urban road network, but also effectively reduce the complexity of the construction of urban road network. Thus, it provides new research ideas and methods for navigable road network generation that integrates vehicle trajectories and remote sensing images, which has important academic and application value in the fields of intelligent transportation and location-based service.