Research On Tridimensional Green Biomass Of Beijing Study Area Based On WorldView-2 Image

Nowadays, with the rapid development of economy and urbanization, there is more and more environment pollution around us. It has been a necessary question how to evaluate the eco-efficiency level and function of the urban environment in a better way. The traditional assessment index can not meet the demand of evaluation of ecological benefits. So a new index of three-dimensional green biomass comes out for explaining the differences in vegetation spatial structure.Based on terrestrial laser scanning radar system, we scan the cell samples in Beijing, and acquire the complete three-dimensional structure cell information. Then we extract cloud date of trees from it in order to calculate trees volume in form of its triangular grid. On this basis, we acquire interpretation information of the corresponding vegetation in World View-2 remote sensing image. Using SPSS statistical software and geographic information system modeling and spatial analysis tools, we build a model of three-dimensional green biomass. With this model, the amount of three-dimensional green biomass of study area in Beijing was estimated and analyzed. We provide an effective way and method for rapid and accurate estimate the three-dimensional green biomass in large-scale. My main conclusions are as follows:(1) By constructing a closed triangle mesh, and using a method of mesh model, we calculate the trees’ volume. Its model is infinitely close to the shape of the crown. To verify reliability of this model, we compared our model with traditional method.(2)A method of object-oriented classification is used in e Cognition software while extract the vegetable information of World View-2 remote sensing image.The main concern of this study is vegetation, so we choose the 25 division scale in remote sensing by experimental analysis, and the result meets the requirements very well. In order to make the estimated value more accurate, in the end of it,we removed the color roofs and green playground. The classification accuracy is more than 90% after sampling verified.(3)We established three-dimensional green biomass model of our study area based on World View-2 remote sensing, using samples, remote sensing factor and correlation vegetation index. The model accuracy is 83.23%,and is higher than 80% of the international model accuracy criterion. This model has practical value in other area.