Rule-based rapid fine-scale city impervious surface extraction method

Abstract

The invention discloses a rule-based rapid fine-scale city impervious surface extraction method, and belongs to the technical field of city planning information application. The method comprises the following steps of: preprocessing an original high-resolution multispectral image to obtain a fused image; establishing an extracted decision-making tree according to principles that the easiest is taken the first and surface feature difference between nodes is obvious; differentiating a road part and a non-road part on an image pixel layer; segmenting the non-road part; carrying feature description on different surface features, constructing a classification rule set according to a classification decision-making tree, and obtaining a fine-scale impervious surface distribution result of a cityarea; and carrying out precision verification on the classification result through evaluation after classification and a Kappa index method. The method is good in classification effect, and is capableof effectively overcoming the spiced salt phenomenon in remote sensing image information extraction, well decreasing the phenomena of same spectrum with different objects and same object with different spectrums, and improving the efficiency and correctness of obtaining city impervious surfaces.

Description

technical field [0001] The invention belongs to the technical field of urban planning informatization application, and in particular relates to a rule-based rapid extraction method for fine-scale urban impermeable surfaces. Background technique [0002] The types and distribution of urban impervious surfaces are important concerns in urban green space system planning and drainage and waterlogging prevention planning. Urban planning and design also have high requirements for the fineness of the spatial distribution of impervious surfaces. Existing impervious surface estimation methods often use the properties of construction land parcels for estimation, and construction land parcels are generally bounded by urban roads. The specific calculation method is as follows: within the scope of statistical estimation, the land use nature is the area of ​​"park green space" and "protective green space". The existing data acquisition methods have the following deficiencies. First, the ...

AI Technical Summary

Problems solved by technology

At present, there is no rule set for extracting fine-scale urban impervious surfaces. It is necessary to establish a set of universal fine-scale urban underlying surface extraction rule sets to improve the timeliness, accuracy, and accuracy of obtaining this data in the field of urban planning. economy

Images