Method for determining water collection area threshold value during digital river network extraction

Abstract

The invention discloses a method for determining a water collection area threshold value during digital river network extraction, and the method comprises the steps: carrying out the recognition of awatershed water body through watershed remote sensing image data, and carrying out the rasterization of a water body recognition result; calculating a confluence accumulation value of each grid unit in the drainage basin by utilizing the watershed DEM data, and generating a confluence accumulation grid; extracting the closest possible river channel in the river basin according to the minimum threshold value of the confluence accumulation value; calculating the river channel reliability within a certain range one by one by taking the grid unit on the closest possible river channel as a center according to the rasterized water body recognition result; and according to the rasterized river channel reliability, establishing a relation curve of a threshold value and a river channel reliabilitymean value, and determining a critical threshold value point which is a water collection area threshold value. The method has the advantages of being stable and reliable in data source, high in calculation efficiency, objective and reasonable in result and the like, and facilitates determination of a basin water collection area threshold value and extraction of a digital river network.

Description

technical field [0001] The invention relates to the technical field of hydrology, in particular to a method for determining the catchment area threshold value during digital river network extraction. Background technique [0002] The characteristics of the river network reflect the comprehensive hydrological characteristics of the watershed. With the development of technologies such as remote sensing, geographic information, and digital watersheds, the distributed hydrological model based on the grid digital elevation model (DEM, Digital Elevation Model) has become an important model for its full consideration of the spatial variation of rainfall and underlying surface conditions. Trends in watershed hydrological modeling. The accurate extraction of the river network provides guarantee for the model calculation. In order to further promote the development of the hydrological model of the watershed, it is necessary to further study the determination method of the catchment a...

AI Technical Summary

Problems solved by technology

Theoretically speaking, the water body identification result of remote sensing image data is a digital reflection of the real water body, but because the model algorithm for water area extraction based on remote sensing image research cannot identify any ground object, nor can it perfectly distinguish water bodies from other land objects, and because The remote sensing image data itself has errors due to cloud cover images, or there are small water bodies in the basin itself. The water body identification results are discontinuous in the main stream and scattered in the tributaries. It is impossible to directly extract the basin water system from the remote sensing image water body identification results.

Images