Scholarlink

<p>This study conducted a comprehensive analysis of the relationships between environmental factors and urban morphology in small towns of the Qinba Mountains in Southern Shaanxi using Earth observation data and advanced spatial analysis methods. Through the application of U-Net3+ deep learning model and concave hull algorithm, the study successfully extracted and precisely defined the boundaries of 123,992 buildings across 358 small towns. The environmental clustering analysis, employing Self-Organizing Map (SOM) algorithm, identified six distinct environmental types, revealing significant heterogeneity in environmental conditions. The morphological analysis, utilizing ten quantitative indicators, revealed eight distinct urban form clusters, ranging from compact patterns to expansive metropolitan forms. The correlation analysis demonstrated strong relationships between environmental and morphological characteristics, with slope and NDVI showing the strongest correlations with urban development patterns. The Minimum Spanning Tree (MST) analysis further revealed how these patterns manifest spatially, from concentrated distributions in favorable environments to dendritic patterns in challenging terrains. The findings highlight the crucial role of environmental conditions in shaping urban development patterns and provide valuable insights for sustainable urban planning in mountainous regions. The study contributes to urban morphology research by establishing a quantitative framework for analyzing environmental-morphological relationships and demonstrates the effectiveness of combining machine learning with spatial analysis methods in understanding complex urban-environment interactions.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Uncovering urban morphology and environmental interactions of small towns using self organizing maps in Qinba Mountains of Southern Shaanxi China

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2025年11月5日

查看全文
  • Xin Zhao,
  • Zuobin Wu

摘要

This study conducted a comprehensive analysis of the relationships between environmental factors and urban morphology in small towns of the Qinba Mountains in Southern Shaanxi using Earth observation data and advanced spatial analysis methods. Through the application of U-Net3+ deep learning model and concave hull algorithm, the study successfully extracted and precisely defined the boundaries of 123,992 buildings across 358 small towns. The environmental clustering analysis, employing Self-Organizing Map (SOM) algorithm, identified six distinct environmental types, revealing significant heterogeneity in environmental conditions. The morphological analysis, utilizing ten quantitative indicators, revealed eight distinct urban form clusters, ranging from compact patterns to expansive metropolitan forms. The correlation analysis demonstrated strong relationships between environmental and morphological characteristics, with slope and NDVI showing the strongest correlations with urban development patterns. The Minimum Spanning Tree (MST) analysis further revealed how these patterns manifest spatially, from concentrated distributions in favorable environments to dendritic patterns in challenging terrains. The findings highlight the crucial role of environmental conditions in shaping urban development patterns and provide valuable insights for sustainable urban planning in mountainous regions. The study contributes to urban morphology research by establishing a quantitative framework for analyzing environmental-morphological relationships and demonstrates the effectiveness of combining machine learning with spatial analysis methods in understanding complex urban-environment interactions.