Analysis And Simulation Prediction Of Land Use Change In Linxia Prefecture Based On Image Segmentation

The change of land use is the result of the influence of nature,humanity,society and other factors on the spatial relationship.As the most direct reflection of regional development,it objectively records the influence of human beings on the spatial distribution of land.Ethnic minority areas generally have special social,historical and cultural backgrounds and backward economic conditions,and urbanization develops slowly.In the context of the new normal and new development,it is necessary to have a comprehensive understanding of the current regional reality and adjust new ideas to promote the positive and sound development of the region.Therefore,to explore the situation of land use in ethnic minority areas and simulate and predict its future pattern provides a reference for further describing such regional land use problems,which is also of great significance for accelerating the pace of urbanization and improving the quality of coordinated development.This paper uses Landsat imagery as the basic data,combined with object-oriented classification methods,based on the dynamic model of land use change and principal component analysis model,to describe the law and driving factors of land use change in Linxia Hui Autonomous Prefecture,using the CA-Markov model and the basic farmland Scope and religious land scope are used as restrictive factors to simulate and predict the spatial pattern of its land in 2025.The main research content and conclusions of the paper are as follows:Remote sensing image segmentation and classification based on object-oriented method.Multi-scale segmentation algorithm and ESP(Estimation of Scale Parameter)scale evaluation tool were used to determine the optimal segmentation scale parameters such as the segmentation scale,shape factor and compactness factor of remote sensing image.Meanwhile,appropriate classification features are selected and the rule classification of the image is completed in combination with the Assign Class algorithm,and the accuracy is tested.The results show that the optimal segmentation scale parameters of the image are segmentation scale 82,shape factor 0.4,and compactness factor 0.6 respectively.The object-oriented rule classification has a high classification accuracy(the overall accuracy in each year is over 85%,and the Kappa coefficient is over 0.80).(2)Based on the dynam ic model of land use change and principal component analysis model,the characteristics and driving force of land use change were analyzed.Combined with the dynamic model of land use change,the law of land use change was explored from the four aspects of quantity,speed,land class transfer and regional difference.The principal component analysis model was used to analyze the impact of driving factors on land use change in combination with social and economic indicators and physical geographic data.The results showed that the land use in Linxia was unbalanced in the 10 years from 2009 to 2019.The construction land,forestland and unused land increased to different degrees.Among the eight counties and cities,Linxia has the lowest change amount and rate,while other counties and cities are in a relatively rapid development stage,but the development level varies greatly.Factors of urbanization level and related factors of secondary and tertiary industries play a driving role in regional development.Natural factors such as slope,road and river have significant influence on land use pattern.(3)Based on CA-Markov model,land use change simulation and prediction analysis were completed.Based on the Markov transfer matrix as the quantitative simulation basis,the basic farmland and religious land scope were added to the production of the suitability atlas to form the conversion rules.The spatial pattern of land use in Linxia Prefecture in 2019 was simulated and predicted,and the land use classification results in 2019 were used to test its accuracy.The results show that the quantitative accuracy is above 90%and the Kappa coefficient is 0.87,indicating that this model can be applied to the simulation and prediction of land use pattern in Linxia State in 2025.On this basis,the spatial pattern of land use in 2025 is simulated,and the change of land use in Linxia prefecture from 2019 to 2025 is analyzed.