Ground penetrating radar signal denoising method based on mixed Fourier-wavelet analysis

Abstract

The invention provides a ground penetrating radar signal denoising method based on mixed Fourier-wavelet analysis. Through carrying out multi-scale wavelet decomposition on a ground penetrating radar echo signal, a threshold function is used for updating a wavelet coefficient, the advantage of high resolution in the frequency domain of Fourier transform is inherited, defects that the Fourier transform is not applied to a non-stationary signal, and frequency feature information of a local time segment can not be extracted are overcome, advantages of wavelet transform localization and automatic focusing are made full use of, the defect that the wavelet transform has poor effects on removing low-frequency noise is remedied, particular features of the ground penetrating radar are fully considered, a specific formula in each step during the using process is given, parameters such as a sampling rate, a central frequency and a bandwidth can be directly substituted for denoising processing as for different ground penetrating radars in actual use, and the operation is simple and convenient.

Description

technical field [0001] The present invention relates to the technical field of radar, and more particularly relates to a ground-penetrating radar signal noise reduction method based on hybrid Fourier-wavelet analysis. Background technique [0002] As a new type of geophysical detection instrument, ground penetrating radar has the advantages of high detection efficiency, simple operation, rapid sampling, non-damage detection, high precision, high detection resolution, and strong anti-interference ability. effective means of detection. In ground penetrating radar detection, in order to obtain more reflected wave features, a wide frequency band is usually used for recording, so it is inevitable to record various interference noises while recording various effective waves. Often, unprocessed sections cannot be interpreted geologically. Therefore, it is necessary to denoise the received GPR echo signals to improve data quality and provide clear and reliable GPR signals and prof...

AI Technical Summary

Problems solved by technology

However, this method has two defects: one is that the window cannot change with the frequency, and the other is that it lacks a discrete orthonormal basis.

However, in the actual signal processing of ground penetrating radar, wavelet analysis is only good for removing higher frequency components in the signal, and cannot effectively remove low frequency noise in the signal

[0006] In the process of realizing the present invention, the applicant found that the above prior art has the following technical defects: the commonly used Fourier transform cannot describe the time-frequency local characteristics of the signal, and due to the attenuation caused by the underground medium, the ground-penetrating radar signal is a typical non- Therefore, the Fourier transform cannot extract the frequency feature information of the local time period, and the denoising effect is not good; although the traditional discrete wavelet transform threshold denoising method has good time-frequency characteristics, it is only suitable for signal removal in most cases. The high-frequency components in the signal, but the suppression effect on the inherent low-frequency noise in the signal is poor; the above two methods need to select appropriate parameters based on experience to achieve good results during operation, which is not simple enough in practical engineering applications

Images