Analysis of Land Use Structure Changes Based on Deep Learning
In order to better study the land use situation and reveal the law of land use development,this article takes Harbin city as the research area. Firstly,land use classification is conducted based on remote sensing images and a land use change dataset is crea-ted. Then,a CNN model is used to simulate the land use structure and evaluate its accuracy. Finally,combined with the Random Forest Regression Prediction (RFR) algorithm,the MCCA model is used to predict the land use structure from 2020 to 2030. The ex-periment shows that this article helps relevant personnel to understand the characteristics of land use structure changes in Harbin,and provides scientific basis for government departments to formulate sustainable land use plans.
NETL
NSTL
万方数据
