A Material Classification Method for Urban Buildings by Fusion of Multi-source Satellite Remote Sensing Data

Abstract

A method for classifying urban building materials by fusing multi-source satellite remote sensing data. The steps include: S1: preprocessing hyperspectral images and high-resolution images; S2: performing independent component analysis on the preprocessed hyperspectral images to obtain independent Component, the independent component and the building material are matched and analyzed to obtain the independent component image of the building material; S3: Morphological transformation is performed on the independent component image of the building material to obtain various building material feature images; S4: The preprocessed Multi-scale segmentation of high-resolution images to obtain building boundary images; S5: Decision-making fusion of building material feature images and building boundary images to obtain building material classification results. The present invention fuses hyperspectral images and high-resolution images To realize the material classification of buildings, improve the accuracy of classification results and avoid problems such as missing geometric shapes and incomplete boundaries.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, and more specifically relates to a method for classifying urban building materials by fusing multi-source satellite remote sensing data. Background technique [0002] Urban buildings are an important part of the city and a stable space for human habitation and daily activities. Rapid and accurate acquisition of urban building material information plays an important role in urban surveys, urban planning, urban management, disaster assessment and natural disaster compensation. With the continuous development of social economy and increasing human activities, the material information of urban buildings has attracted more and more attention. The traditional method of material classification of urban buildings is generally manual field survey, which wastes a lot of manpower and material resources and is inefficient. [0003] With the rapid development of technologies such as ...

AI Technical Summary

Problems solved by technology

However, due to the prevalence of individual differences in urban buildings, different spectra of the same object and the same spectrum of different objects, the spectrum of a certain point on the surface of the building is not representative.

In addition, the spatial resolution of hyperspectral remote sensing images is generally low, and the material classification of buildings is only based on hyperspectral images. The classification results have problems such as lack of spatial geometry of buildings and incomplete boundaries.

Images