Research On Extracting Digital Drainage Network Of Watershed Based On DEM And Remote Sensing

The digital drainage network of watershed is quite indispensable to the hydrological simulation, so extracting digital drainage network and physical descriptors of watershed based on DEM is the precondition of research on building distributed hydrological model and hydrological simulation.After analyzing the defects of extracting digital drainage network based on DEM, this paper makes up these defects by importing the Remote sensing data. Then this paper explicates the method of extracting digital drainage network of watershed based on DEM and remote sensing, and performs an experiment upon Qinhuai River basin using this method. The result of this experiment indicates that this method can effectively overcome and solve the shortcoming of parallel water ways and the main channel which departs its natural position. Not only in the flat but also in mountainous area, this method can extracting digital drainage network of watershed which accords with the natural drainage network.The primary contents and conclusions of this paper are as follows:1. By classification in remote sensing image based on decision tree algorithm, the water information which is needed to correct extracting digital drainage network based on DEM can be extracted. The method improves the degree of automation in obtaining correction information, and makes the correction information easier to update. With mathematical morphology, the main channel and large lakes are filtered from the water information, and the result indicates that this method can availably remove the useless impurity from the water information.2. This paper extracts the medial axis of river by using Delaunay Triangulations, then deletes the excrescent embranchments. The result indicates that the medial axis of river which is extracted using this method can preserve the connectedness and topological structure of the main channel and large lakes, and that the method has the feature of high operating efficiency.3. After several pretreatments, the digital river and lake network is extracted from remote sensing image, and is used to correct the flow direction of the river and lake in the flat. The digital drainage network of watershed which is finally extracted with this method indicates that the parallel water ways are deleted, and the main channel which departs its natural position is corrected, so the digital drainage network accords with the fact. Finally this paper corrects the large lakes which are missing, and reinforces the accurate locality of the two-dimensional lake so that it can provides data support for advancing the precision of hydrodynamic calculation.