Urban land utilization change simulation cellular automaton method based on BSVC method

Abstract

The invention relates to an urban land utilization change simulation cellular automaton method based on a BSVC method. The method comprises the following steps: 1) supervising and classifying remote sensing images to obtain an urban land utilization classification map; 2) obtaining urban land utilization change driving factor data, and performing preprocessing; obtaining effective sample points ofthe land utilization map and the driving factor through a random stratified sampling method; 3) training the effective sample points by using a Bayesian space variable coefficient (BSVC) method to obtain a conversion rule of the cellular automaton; 4) obtaining the urban land utilization conversion probability according to the CA conversion rule established by the BSVC; 5) establishing a CA model(CABSVC) based on the BSVC; the urban land use change simulation method based on the CABSVC model has the advantages of being higher in simulation precision and better in urban land use change simulation capacity compared with the prior art, and realizing evaluation through the overall precision (OA) and the figure goodness (FOM).

Description

Technical field [0001] The invention relates to a cellular automata method for simulating urban land use change, in particular to a cellular automata method for simulating urban land use change based on a BSVC method. Background technique [0002] Urban growth reflects the change from the natural surface to urban artificial areas, which has a significant impact on human society and the ecological environment. In the urban growth simulation, some models consider the non-stationarity of space, but the existing model methods cannot fully reflect the spatial heterogeneity of urban dynamics. [0003] Geographical Cellular Automata (CA) is a bottom-up, self-organizing city simulation model, which is used to reconstruct historical patterns of urban land use and predict future scenarios. In order to define conversion rules based on driving factors, CA models usually use biophysical, socio-economic and infrastructure driving factors that reflect urban dynamics. Conversion rules based on d...

AI Technical Summary

Problems solved by technology

Although this method can account for spatial non-stationarity, there are still some characteristics that may make modeling difficult

First, since GWR regression relies on the properties of surrounding samples, dataset outliers may affect regression coefficients through spatial enclave effects

Second, non-constant variance exists widely in the space, which may not satisfy the assumption of normal error

Images