HUO Liang, DUAN Yuanjing, ZHU Yi, SHEN Tao, ZHANG Xiaoyong, ZHAI Jialei, FU Jiying. Multi-scale Expression Method for Urban 3D Model Considering Local Features[J]. Geomatics and Information Science of Wuhan University, 2020, 45(8): 1282-1287. DOI: 10.13203/j.whugis20200148
Citation: HUO Liang, DUAN Yuanjing, ZHU Yi, SHEN Tao, ZHANG Xiaoyong, ZHAI Jialei, FU Jiying. Multi-scale Expression Method for Urban 3D Model Considering Local Features[J]. Geomatics and Information Science of Wuhan University, 2020, 45(8): 1282-1287. DOI: 10.13203/j.whugis20200148

Multi-scale Expression Method for Urban 3D Model Considering Local Features

  •   Objectives   In the study of current city 3D model loading methods, large-scale scenes and single models have not been combined, lack of multi-scale constraints, resulting in the lack of some important information. Therefore, how to make full use of multi-scale spatial data for transmission, expression and provide a flexible and good user experience is the current problem that needs to be considered.
      Methods   From the three aspects of multi-scale TIN(triangulated irregular network) modeling, local feature evaluation, and multi-scale expression, research was carried out, and a multi-scale expression method of urban 3D model that took into account local characteristics was proposed, which realized the multi-scale rapid loading of urban 3D model. First, multi-scale TIN modeling is performed on the city’s three-dimensional model data in layers; then, the local feature evaluation model is constructed at the multi-scale level; finally, priority is given to loading and transmitting detailed data with high local features, even cross-level expression to achieve multi-scale expression.
      Results   It can be seen from the realization effect that local areas with high density of linear feature 3D models are preferentially loaded at the macro scale, models with large volume of the bounding box are preferentially loaded at the meso scale, and local areas with strong identification and high importance are preferentially loaded at the micro scale. The frame rate is used as a reference indicator to compare and analyze this method with the traditional method. The frame rate is maintained at 60 frames per second. It can be seen that the multi-scale constraint of this method makes the loading fluency optimized and the user experience improved.
      Conclusions   The model is divided into layers to construct a multi-scale TIN model. Based on spatial cognition, a local feature degree evaluation model is constructed based on the local area data of each layer sequence loading solves the problem that the drawing of local feature data in the progressive transmission method of the current urban 3D model is relatively delayed, and the user’s effective waiting time is not shortened. It realizes the priority loading and transmission of detailed data with high local feature degree, and even cross-level expression to achieve multi-scale expression of the model and optimize the user experience. In the process of multi-scale TIN modeling, issues such as TIN processing time, update efficiency of changing data, data storage and management also need to be considered.
  • loading

Catalog

    /

    DownLoad:  Full-Size Img  PowerPoint
    Return
    Return