A Novel Spatial Interpolation Method Based on the Integrated RBF Neural Network

With the development of the GIS and computer technology, it is necessary to use sampling data for spatial analysis in their researches. However, the emergence of spatial data interpolation algorithms makes it possible for us to get continuous spatial data that satisfy certain accurate requirement. These traditional methods are suffering some limitations, such as strong subjectivity, numerous assumptions, poor adaptive variation, etc. This paper presents a noval approach by using the establishment of integrated radial basis function (RBF) neural network for improving the spatial interpolation performance. Based on the experimental results on the lead element content in the soil in Kunming and its extensional areas through spatial interpolation by RBF network, it can be concluded that the proposed approach provide the more detail spatial distribution than that by using the traditional methods.