A recognition method for building cluster pattern supported by Fourier shape descriptors
LIU Pengcheng1,2， LIN Wenhui1,2， XU Xiaofeng1,2
1.Key Laboratory for Geographical Process Analysis & Simulation， Hubei Province， Wuhan 430079， China；2.College of Urban and Environmental Science， Central China Normal University， Wuhan 430079， China
Abstract：Using Fourier shape descriptors， similarity analysis was performed on the spatial distribution patterns of buildings. First， the equivalent polygons of the building complex are constructed in the order of the polygon's mass point-center angle. Then， the Fourier spectrum technique is used to extract the shape descriptor of the equivalent polygon as the feature vector of the spatial distribution pattern of the building cluster. Finally， the similarity of distribution patterns between buildings is measured by the Euclidean distance of shape descriptors. An artificial comparison experiment was conducted on a certain number of buildings， and the experimental results were basically consistent with the artificial cognition.
2018, Vol. 52 
