Ground clutter identification method and device, electronic equipment and storage medium

Abstract

The embodiment of the invention discloses a ground clutter identification method and device, electronic equipment and a storage medium, and the method comprises the steps: extracting a characteristic parameter set (the standard deviation PDPstd of a differential phase, the standard deviation ZDRstd of a differential reflectivity and the vertical gradient DBZg of the reflectivity of an adjacent elevation angle) from a distance library of radar echoes; and inputting the characteristic parameter set into a pre-trained neural network model according to the ratio PR of the coherent power to the incoherent power and the texture structure TDBZ of the reflectivity factor so as to obtain a ground clutter identification result which is output by the neural network model and corresponds to the distance library. The neural network model performs automatic learning according to the sample, the degree of dependence on the experience and level of people is small, and thus the accuracy of ground clutter recognition is improved. Moreover, the characteristic parameters adopted by the scheme of the invention are characteristic quantities with obvious differences under the ground object and rainfall conditions so that the neural network model can be a shallow neural network model, and the machine learning efficiency is improved.

Description

technical field [0001] The present application relates to the technical field of signal processing, and more specifically, relates to a method and device for identifying ground object clutter, electronic equipment, and a storage medium. Background technique [0002] For weather radar, ground clutter is a kind of clutter, which seriously pollutes meteorological data and will cause major interference to weather forecast and early warning. Therefore, before the application of meteorological data, ground clutter identification and processing are required for meteorological data. [0003] At present, in the weather radar business system, the automatic identification of ground clutter is based on fuzzy logic technology. This method normalizes the feature quantities of each distance library through membership functions, and then weights the normalized values ​​to obtain Fuzzy output value, compare the fuzzy output value with the threshold to judge whether the echo of the distance l...

AI Technical Summary

Problems solved by technology

[0004] However, the acquisition of membership functions and the determination of weights are largely dependent on people's experience and level, and the subjectivity is too high, resulting in insufficient ability to distinguish some weather objects with similar characteristics.

Images