Research And Analysis On The Impact Of Land Type Changes On Land Surface Temperature In The Urban Area Of Harbin City

With the development of the times,urban changes,and urban population growth,the land type coverage is closely related to factors,such as urban air quality,climate temperature,ecological indicators,and urbanization process.The change of land use type greatly affects the changes in the ecological environment and also affects the rise and fall of land surface temperature in the core urban area and its surrounding areas.Exploring the changes of urban land types and surface temperatures can better provide planning and decision-making basis for the multi-dimensional development of urban ecology,economy,society,and culture.This article uses the combination of machine learning,remote sensing,and geographic information systems to explore the impact of urban land type changes on surface temperature changes in the main urban area of Harbin,Heilongjiang Province.The main research content of the paper is as follows:(1)In order to solve the problem of lack of missing semantic scene information of land use types in traditional remote sensing image change detection methods,a remote sensing image change detection methods,a remote sensing image change detection model based on visual word bags(BOVW)and support vector machine(SVM)was proposed,by extracting feature point information for clustering operation,a visual dictionary with more complete Semantic information can be obtained,based on which the research on change detection of remote sensing image can be realized.This article takes the urban area of Harbin,Heilongjiang Province as the target research area,and uses Landsat series remote sensing images and the NWPU-RESISC4 public dataset as experimental data.The results show that the remote sensing image change detection model based on BOVW and SVM proposed in this paper can make full use of remote sensing images while retaining the scene Semantic information and land type characteristics of remote sensing images to the maximum extent.(2)In order to solve the problem of land surface temperature error caused by single remote sensing data,this paper proposes a land surface temperature inversion model based on linear weighted fusion.For MODIS data and Landsat data,this model uses MODIS window splitting algorithm and Landsat atmospheric correction algorithm to calculate the land surface specific radiance,brightness temperature and other information,and obtains the inversion results of surface temperature,and the two results are weighted and fused to explore the inversion of surface temperature.(3)In order to fill the gap in the research on the impact of changes in urban land types on surface temperature,three correlation analysis methods were used to study the degree of correlation between buildings,water bodies,vegetation,and bare land types and surface temperature.The numerical results of the analysis reflect the correlation between the impact of land type changes on surface temperature.Through the above methods,it has been proven that land type changes do have a correlated impact on the rise and fall of surface temperature,and the degree of impact of different land types on surface temperature varies.The research results of this article can better help the government understand the relationship between changes in urban land use types and surface temperature,providing important basis for urban scientific planning and urban development.