一种基于大模型的城市景观语义分割方法及系统
By using a large-model-based urban landscape semantic segmentation method, the system automatically identifies the categories of environmental elements in street view images and calculates quantitative indicators. This solves the problem of the correlation between urban landscape assessment and geospatial information, achieving efficient and accurate support for urban planning and enhancing the intuitiveness of the analysis results and their decision support capabilities.
CN121600513BActive Publication Date: 2026-07-17URBAN PLANNING & DESIGN INST OF SHENZHEN UPDIS
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- URBAN PLANNING & DESIGN INST OF SHENZHEN UPDIS
- Filing Date
- 2026-01-28
- Publication Date
- 2026-07-17
Smart Images
Figure CN121600513B_ABST
Abstract
本发明涉及智慧城市领域，尤其涉及一种基于大模型的城市景观语义分割方法及系统，获取目标地理区域的街景图像数据；将街景图像数据输入预设的图像语义分割模型中，通过图像语义分割模型对街景图像数据进行解析，识别出每张图像中包含的多种预定义的环境要素类别；基于环境要素类别的识别结果，计算城市景观量化指标；将计算得到的城市景观量化指标的数值，关联至目标地理区域对应的路网空间数据中的对应路段或节点；基于关联后的结果，生成以颜色分级方式表征指标数值高低的道路网络空间可视化图层；根据预设的指标阈值，对可视化图层中符合预警条件的空间路段或节点进行识别并生成提示信息。保证了分析结果的科学性与可比性。
Need to check novelty before this filing date? Find Prior Art