计算机工程与应用 ›› 2012, Vol. 48 ›› Issue (20): 227-231.

• 工程与应用 • 上一篇    下一篇

径向基函数网络与GIS/RS融合的UGB预测

张世良1,叶必雄2,肖守中3   

  1. 1.宁德师范学院 计算机系,福建 宁德 352100
    2.中国科学院 地理科学与资源研究所,北京 100101
    3.福建省宁德市国土局,福建 宁德 352100
  • 出版日期:2012-07-11 发布日期:2012-07-10

Prediction UGB based on integration of GIS/RS and radial basis function network

ZHANG Shiliang1, YE Bixiong2, XIAO Shouzhong3   

  1. 1.Department of Computer and Information Engineering, Ningde Normal University, Ningde, Fujian 352100, China
    2.Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing 100101, China
    3.Ningde Land and Resource Bureau, Ningde, Fujian 352100, China
  • Online:2012-07-11 Published:2012-07-10

摘要: 针对城市空间增长特点,探究城市扩展的规律,研究城市增长边界的计算和预测模型,这对于城市的发展规划具有重要的意义。然而,在国内外,关于城市增长边界方面的研究较少,首次提出利用人工神经网络、地理信息系统和遥感相结合的技术建立具有复杂几何形状的城市增长边界模型。通过数值实验结果表明,模型对城市未来增长边界的计算和预测准确度达80%~84%,结果表示直观、真实,能够为当前精明增长模式下的城市用地规划工作提供决策参考。

关键词: 城市边界增长模型, 神经网络, 地理信息系统, 遥感, 城市规划

Abstract: According to the growing characteristics of urban space, it is very great significant for urban development planning to explore the laws of urban expansion and study the calculation and prediction model of the urban growth boundary. Unfortunately, the study of developed models is little able to simulate the urban growth boundary in domestic and foreign country. This paper proposes to build an urban growth boundary model in complex geometry with the combination of artificial neural networks, geographic information systems and remote sensing technology. The model experimental results show that, with the model, the calculation and prediction accuracy of the future urban growth boundaries is up to 80%~84%, the results express intuitive and real and provide a decision-making reference for the urban planning in the current smart growing ways.

Key words: Urban Growth Boundary(UGB), neural netowk, Geographic Information System(GIS), Remote Sensing(RS), urban planning