Open Geosciences (Apr 2025)
Research on community characteristics of vegetation restoration in hilly power engineering based on multi temporal remote sensing technology
Abstract
In order to accurately identify and evaluate the interference caused to vegetation environment during the construction process of current power transmission and transformation projects in hilly areas, especially in response to the obvious shortcomings of traditional unmanned aerial vehicles and manual detection methods, this study analyzes the composition of vegetation disturbance environment of power transmission and transformation projects in hilly areas from two aspects of natural environmental factors and human environmental factors. Taking the ±800 kV UHVDC transmission project in a certain area as an example, the multi temporal remote sensing data of the project area are collected by Landsat satellite. Combining the deep learning model with the multi model hierarchical classification algorithm, the remote sensing prediction map based on deep learning is used as the classification base map, and the normalized difference vegetation index threshold is used to classify the uncovered area. Combined with the non-building and water area in the classification base map, the candidate disturbance area composed of the two is obtained. The disturbance area is identified in the candidate disturbance area according to the size of the average reflectivity. In the disturbance area, the vegetation restoration community is analyzed according to the difference of normalized difference vegetation index in remote sensing images of different time phases. The experimental results show that the difference between the intelligently identified disturbance area and the engineering design disturbance area can be controlled within 2%, and the vegetation restoration at different time points in the disturbance area can be effectively analyzed.
Keywords
