Construction method and device of reality semantic structure model

Abstract

The present invention provides a method and device for constructing a real-scene semantic structure model, which relates to the technical field of three-dimensional modeling and obtains an input data package of a specified building, wherein the input data package includes the semantic structure model of the specified building, the specified building The real-scene image and the positioning information corresponding to the real-scene image; the geometric information of the specified building included in the semantic structure model is extracted; according to the geometric information and the positioning information corresponding to the real-scene image, the real-scene image is mapped to the corresponding position of the semantic structure model to generate The real scene semantic structure model effectively alleviates the technical problems of low efficiency of manual intervention modeling, large errors, inability to automatically attach attributes, imperfect texture mapping mechanism in the technical process of single modeling, and storage.

Description

technical field [0001] The invention relates to the technical field of three-dimensional modeling, in particular to a method and device for constructing a real-scene semantic structure model. Background technique [0002] With the continuous advancement of urban informatization and the promotion of geographic information technology and smart cities, 3D models and their applications are becoming more and more extensive. At present, the production and application modes of 3D models are mainly figure 1 First, to obtain different forms of data sources; second, to establish a 3D model; third, to attach attributes to the model as needed; fourth, to store and organize data; analyze. [0003] The existing urban 3D model creation and application schemes are mainly to create continuous Mesh (wireless mesh network) or individualized 3D models through automatic, semi-automatic or manual methods, and through attribute connection, data storage, data application and other links To meet ...

AI Technical Summary

Problems solved by technology

There are still many problems in these existing technical processes: first, detailed structures such as window structures need to be generated by manual intervention; second, whether it is a continuous mesh model obtained by oblique photogrammetry modeling or a single The model results of the volumetric modeling method are pure geometric data of the model, and attribute linking needs to be performed through secondary processing. These attribute information are not recorded on the actual model. If users want to obtain this information, they can only access it in a third-party browser. Or the platform manually measures or hooks up the application attributes; third, the texture mapping mechanism in the technical process of single modeling is not perfect; fourth, the modeling tools are diverse and the formats are diverse, and the data conversion between different data and applications becomes one Important issues; Fifth, the large amount of model data has obstacles and bottlenecks in storage, management, browsing, and application

Images