Flood routing process flooding range measurement method based on deep learning

Abstract

The invention discloses a flood routing process flooding range measurement method based on deep learning. The method comprises the steps that firstly, a camera is arranged to be used for collecting video data of a whole test river channel; calibrating the camera through a camera checkerboard calibration method to correct the perspective distortion effect; extracting image data at different time points from the video data; constructing a flood test flooding range sample library; carrying out primary labeling on the sample through a Label labeling tool; and finally, adopting an MASK R-CNN imageinstance segmentation algorithm to realize automatic segmentation and recognition of a flooding range, and obtaining the submerging range change of the whole test riverway by splicing the recognitionpictures. The method has the advantages of being low in economic cost, high in intelligent degree, high in efficiency, high in precision, high in applicability and the like, and therefore the method can be used for extracting the obtained flooding range data in the flood routing test.

Description

technical field [0001] The invention belongs to the technical field of data monitoring, and relates to a method for measuring the submerged range of a flood evolution process based on deep learning. Background technique [0002] As a natural disaster, flood has a high probability of occurrence, a wide range, and great harm, which may cause heavy casualties and property losses. Numerical models are increasingly used to simulate flood propagation more and more widely. However, the verification of these models is mainly based on the comparison of different model predictions, or the water depth data of individual survey points from field surveys and experiments, and lacks data on the spatial variation of inundation range during flood evolution. [0003] In the traditional flood evolution process, the water level sensors and current velocity meters arranged in the river channel are mainly used to monitor the data of individual measuring points. This method can only monitor the ...

AI Technical Summary

Problems solved by technology

[0006] Aiming at the deficiencies of the prior art, the purpose of the present invention is to provide a method for measuring the submerged range of the flood evolution process based on deep learning, which solves the problem that the existing measurement methods in the prior art have only individual measuring points and no spatial range change measurement data , and the manual data processing method is time-consuming and labor-intensive. This solution can easily extract the water surface video data of the entire test river, and automatically identify the change of the submerged range during the flood evolution process based on camera perspective distortion correction and deep learning technology.

Images