Land Use Change Prediction Model Combining Remote Sensing Images, Meteorological Data and Machine Learning
Year: 2025
Type: conference-paper
Abstract: In the prediction of land use change, traditional models cannot fully utilize the advantages of multi-source remote sensing images, meteorological data and machine learning (ML) algorithms. This paper adopts an ML land use change prediction model based on remote sensing images and meteorological data. The model combines the spatial and temporal characteristics of remote sensing images and meteorological data, and uses ML methods to perform dynamic predictions of land use changes, aiming to provide more accurate and reliable technical support for the monitoring and management of land use changes. First, the land use types in the study area are classified through remote sensing image data, and the spatial distribution information of different land use types is extracted. Then, historical meteorological data (such as temperature, precipitation, wind speed, etc.) are combined with remote sensing image data to construct a multi-source data fusion model to provide more comprehensive environmental background information for prediction. Then, ML algorithms such as SVM and RF are used to train the fused data to build a prediction model for land use change. Finally, the model performance is evaluated by cross-validation, and the results are verified by combining field survey data to ensure the prediction accuracy and stability of the model. As the time window gradually increases from 1 year to 10 years, the accuracy rate steadily increases from 85.0% to 89.9%, the Kappa coefficient increases from 0.77 to 0.87, and the F1 score increases from 0.80 to 0.89. The ML land use change prediction model based on remote sensing images and meteorological data proposed in this study can effectively improve the accuracy and timeliness of land use change prediction. This method provides a new technical means for land resource management and planning.
Source: Procedia Computer Science
Author: Linlin Li
Language: English
Topic: Remote Sensing and Land Use
Subfield: Atmospheric Science
Field: Earth and Planetary Sciences
Domain: Physical Sciences
SDGs: Climate action, Life on land
Open Access status: diamond
OpenAlex ID: W4411175970