Technology realizes the quantitative and positioning acquisition of soil erosion and land use information, grasps the relationship between the two from space, and provides theoretical reference and scientific basis for local ecological environment construction and soil and water conservation work. This paper uses remote sensing images in my country in 2020 and 2021 as the data source and obtains land use data in four periods respectively. The experimental results show that the land use structure in my country has undergone great changes in 2020, and the land use type has gradually changed from a structure dominated by cultivated land, grassland, and unused land to grassland, forest land, and cultivated land. The other four types of land use area have increased to varying degrees, the unused land has decreased significantly, and the grassland and forest land have increased significantly; there are differences in the changes in the degree of land use in each study period, and the overall level of land use has developed phase by phase toward higher levels. In addition, this paper also studies the clustering algorithm in machine learning and proposes an improved interpolation algorithm for the completion of the original rainfall data. This algorithm can also be applied to the calculation process of rainfall erosion factors, which realizes the automatic calculation system of soil erosion model factors, realizes the real-time calculation and monitoring of soil erosion in the form of calculation tasks, and solves the problem that manual calculation consumes manpower and material resources.
Read full abstract