Fuzzy inference guided cellular automata urban-growth modelling using multi-temporal satellite images

摘要: This paper presents a fuzzy inference guided cellular automata approach. Semantic or linguistic knowledge on urban development is expressed as rules, based which applied to determine the potential for each pixel. A defuzzification process converts required neighbourhood level, taken by initial approximation its transition rules. Such approximations are updated through spatial calibration over townships and temporal with multi-temporal satellite images. Assessment of modelling results three evaluation measures: fitness Type I II errors. The approach model growth city Indianapolis, Indiana period 30 years from 1973 2003. level 100 ±20% 30% average errors can be achieved 80% in urban-growth prediction.