Vertical Distribution Feature Extraction of Virtual Simulation Greening System by UAV-A Case Study of High-Density Cities

To fill the lack of overall planning and unreasonable distribution of urban vertical greening, a method for extracting vertical distribution characteristics of high-density urban greening systems was proposed in this study. A remote sensing image acquisition platform was built with unmanned aerial vehicles (UAVs) to collect greening system images. A convolutional neural network (CNN) architecture was established to remove initial image noise through residual learning. The tasseled hat transformation algorithm was used to rotate the image structure axis to enhance the image and highlight the image details. According to the similarity and discontinuity of the spectrum, a data-driven algorithm was used to establish a homogeneity standard, and the image was divided into different areas to ensure that the same target was in the same area. The vertical distribution characteristics of the greening system were extracted from three aspects: spectrum, shape, and texture. The experimental results showed that in the target area, the vertical distribution characteristics of the commercial area and the residential area were similar, but the green area of the residential area was larger, and the plant species in the commercial area were more abundant. According to the extracted features, the greening system can be a reference for further study.