The Research On Data Fusion Analysis Of Full-Polarimetric Ground Penetrating Radar

Ground Penetrating Radar(GPR)is a non-destructive detection tool suitable for shallow detection based on electromagnetic reflection information.Conventional ground penetrating radars are mostly single-polarization radars,which use the frequency and phase of electromagnetic reflection waves to identify the target,but it can only collect information in one polarization mode.For complex underground targets recognition is sometimes difficult to meet the requirements.In order to obtain more comprehensive and accurate information about underground targets,a fullpolarimetric geochemical ground penetrating radar was developed.The fullpolarimetric ground penetrating radar works through four different antenna combinations,and it can obtain three times more information than the single-polarized ground penetrating radar.The full-polarimetric ground penetrating radar can not only obtain the amplitude,phase,frequency and other information that can be obtained by the traditional ground penetrating radar,but also obtain the polarization attribute information.Therefore,the full-polarimetric ground penetrating radar is better for the recognition of complex targets.ability.However,how to integrate the various information in the four kinds of ground penetrating radar images to obtain final fullpolarimetric ground penetrating radar image requires further research.Data fusion is to extract the data information of the same target from different sensors to obtain an accurate description of the same target by extracting multiple data.In the field of full-polarimetric ground penetrating radar data fusion,only the weighted average fusion method is currently used.This method cannot determine the weights suitable for different polarization methods,and cannot adapt to underground targets with different scattering mechanisms at the same time.Simple average Fusion will conceal the advantages of full polarization,resulting in the fusion of the image results unable to obtain information that highlights the underground target.The main purpose of this research is to find methods(principal component analysis,Laplacian pyramid,multi-scale wavelet transform)to perform full-polarimetric ground penetrating radar data fusion,and strengthen the resolution of the full-polarimetric ground penetrating radar fusion image.The main content of the paper is as follows:1.The acquisition system of the full-polarimetric ground penetrating radar is introduced.Including two kinds of full-polarimetric ground penetrating radar systems,one is a system suitable for laboratory measurement,a full-polarimetric ground penetrating radar system based on Vivaldi antenna;the other is a system suitable for field measurement,based on automatic control of horn antenna full-polarimetric ground penetrating radar system.2.Introduced the theory of full-polarimetric ground penetrating radar data fusion,developed principal component analysis,Laplace pyramid and multi-scale wavelet transform and full-polarimetric ground penetrating radar data fusion method.3.In the laboratory,the three data fusion methods were combined with the target volume data of the three different scattering mechanisms of the full-polarimetric ground penetrating radar,and establish a fusion image selection system with instantaneous amplitude and improved gradient as the core.A data fusion method suitable for targets with different scattering mechanisms,and a data fusion method suitable for targets with unknown scattering mechanisms.4.Detect the ice layer in the wild,and compare the data fusion method suitable for the unknown scattering mechanism target with the traditional weighted average method to achieve a better fusion effect.