CN-122018025-A - Ground penetrating radar image cavity imaging method suitable for different geological conditions

CN122018025ACN 122018025 ACN122018025 ACN 122018025ACN-122018025-A

Abstract

A ground penetrating radar image cavity imaging method suitable for different geological conditions belongs to the technical field of ground penetrating radar detection. In order to solve the problem that the signal processing technology of the ground penetrating radar cannot be effectively adapted to dynamic changes of medium parameters under different geological conditions, the method comprises the steps of constructing normalized field intensity after background inhibition along with depth dynamic changes, extracting phase consistency structural strength, obtaining depth-medium coupling potential energy field, constructing anisotropic Gaussian kernel function with direction selectivity, calculating thermodynamic upper boundary manifold projection and thermodynamic lower boundary manifold projection, extracting enhanced gradient characteristic field through calculating difference between the thermodynamic upper boundary manifold projection and the thermodynamic lower boundary manifold projection, and establishing multi-scale entropy weight fusion inverse time imaging to obtain a final underground cavity imaging result. The invention ensures that the detection imaging with high signal-to-noise ratio and clear outline can be obtained under different geological conditions.

Inventors

KAN QIAN
LIU XING
MENG ANXIN
CHENG GONG
ZHUANG WEIQUN
LI JUNYUAN

Assignees

深城交科技集团股份有限公司

Dates

Publication Date: 20260512
Application Date: 20260414

Claims (8)

1. The imaging method of the ground penetrating radar image cavity adapting to different geological conditions is characterized by comprising the following steps: S1, underground original data are collected by a ground penetrating radar, and normalized field intensity after background suppression along with depth dynamic change is constructed based on a statistical principle; S2, taking account of exponential amplitude attenuation of electromagnetic waves along with the increase of depth when the electromagnetic waves propagate underground, introducing a phase consistency analysis technology, and extracting phase consistency structural strength; S3, coupling the normalized field intensity after the background suppression obtained in the step S1 with the phase consistency structural strength obtained in the step S2 to obtain a depth-medium coupling potential energy field; s4, constructing an anisotropic Gaussian kernel function with direction selectivity, wherein the anisotropic Gaussian kernel function with direction selectivity is expressed as an elliptic Gaussian distribution with specific length-width ratio and rotation angle in mathematical morphology; S5, calculating thermodynamic upper boundary manifold projection and thermodynamic lower boundary manifold projection based on the depth-medium coupling potential energy field obtained in the step S3 and the anisotropic Gaussian kernel function with direction selectivity obtained in the step S4; S6, extracting an enhanced gradient characteristic field by calculating the difference between thermodynamic upper boundary manifold projection and thermodynamic lower boundary manifold projection; And S7, establishing a multi-scale entropy weight fusion inverse time imaging based on the self-adaptive weighting strategy of the information entropy and the enhanced gradient characteristic field obtained in the step S6 to obtain a final underground cavity imaging result.
2. The method for imaging the ground penetrating radar image cavity adapting to different geological conditions according to claim 1, wherein in step S1, a background model reflecting the electromagnetic characteristics of the current geological horizon is established, and an abnormal response and background fluctuation in an original signal are effectively decoupled by combining a nonlinear mapping function, so that normalized field intensity after background suppression is obtained 。
3. The method for imaging the image cavity of the ground penetrating radar, which is suitable for different geological conditions, according to claim 2, wherein in the step S2, the radar profile is subjected to frequency domain decomposition and reconstruction by constructing a multi-scale filter bank, so as to obtain the phase consistency structural strength, and the expression is as follows: ; Wherein, the For phase consistency structural strength, x is a horizontal measuring line position index, z is a vertical depth position index, and is obtained by original data acquired by a ground penetrating radar; selecting proper quantity for total scale quantity by professional according to the target cavity size range and radar frequency bandwidth through pre-experiment; the weight is expanded for the frequency of the nth scale, and preset by a professional according to the frequency bandwidth or energy distribution of each scale filter; The signal envelope amplitude of the n-th scale is obtained by performing Hilbert transform on the filtered signal; For the phase deviation weighting factor of the nth scale, the instantaneous phase of the nth scale Average phase with scale The deviation between the two is calculated and obtained, ; The noise estimation threshold is determined empirically by an expert.
4. The method for imaging a cavity in a ground penetrating radar image adapted to different geological conditions according to claim 3, wherein a depth compensation mechanism is introduced in step S3, and the method is characterized in that: ; Wherein, the Potential energy field values for depth-medium coupling; The fusion weight coefficient is determined by expert experience; the medium attenuation compensation factor is set by an expert according to the geological type.
5. The method for imaging a cavity in a ground penetrating radar image adapted to different geological conditions according to claim 4, wherein the expression of the anisotropic gaussian kernel function with direction selectivity constructed in step S4 is: ; Wherein, the Is the direction of the scale s S is a scale index, and corresponds to cavity targets with different sizes, and is determined by professionals; the rotation angle of the kernel function is determined and selected by a professional to represent the horizontal, oblique and vertical directions; The row and column indexes in the kernel function local coordinate system are defined in a limited support domain of a convolution kernel and are determined by professionals; 、 the standard deviation of the anisotropic gaussian kernel in the directions of the principal axis and the secondary axis respectively determines the aspect ratio and the spatial resolution of the kernel, and is set by a professional according to s.
6. The method for imaging the image cavity of the ground penetrating radar adapted to different geological conditions according to claim 5, wherein the step S5 combines convolution operation and index mapping, calculates thermodynamic upper boundary manifold projection, and the calculation formula is: ; Wherein, the Is an upper manifold scalar value; d is an integral domain, is a space range covered by a kernel function, and is determined by a professional according to a scale index s; by adopting the mapping transformation of the negative index, the thermodynamic lower boundary manifold projection is calculated, and the calculation formula is as follows: ; Wherein, the Is a lower manifold scalar value.
7. The method for imaging the cavities of the ground penetrating radar image adapting to different geological conditions according to claim 6, wherein the expression of the gradient characteristic field after the enhancement extracted in the step S6 is as follows: ; Wherein, the For an enhanced gradient feature field; 、 second partial derivatives in the x-direction and y-direction, respectively; 、 first partial derivatives in the x-direction and y-direction, respectively.
8. The method for imaging the cavity of the ground penetrating radar image adapting to different geological conditions according to claim 7, wherein in the step S7, the effective information content contained in the image is estimated by calculating the information entropy of the image under each scale, and the final underground cavity imaging result is output by weighted summation and direction maximum projection, and the expression is: ; Wherein, the The final underground cavity imaging result is obtained; For gradient field at s-th scale Is used for the image information entropy of (a), For gradient field at the kth scale Is determined by a professional; is the maximum value of the direction.

Description

Ground penetrating radar image cavity imaging method suitable for different geological conditions Technical Field The invention belongs to the technical field of ground penetrating radar detection, and particularly relates to a ground penetrating radar image cavity imaging method suitable for different geological conditions. Background Although the ground penetrating radar is used as a mainstream nondestructive testing means for detecting underground hidden engineering, the imaging quality is easily controlled by the complexity and the variability of different geological conditions in a detection area. In an actual engineering scene, due to uneven water content distribution of an underground medium, grading difference of roadbed materials and staggered change of stratum layer arrangement, echo signals received by a radar often show nonlinear attenuation characteristics and strong background clutter interference. The existing conventional data processing method is mostly prone to adopting a globally unified amplitude threshold value cut-off or linear frequency domain filtering algorithm, and the processing logic lacking self-adaption is poor in effect when facing different geological environments, so that background noise is easily misjudged to be abnormal in a shallow layer strong reflection area, and a cavity target is easily missed due to weak signals in a deep layer high attenuation area. Therefore, the conventional means that rely on amplitude intensity or phase continuity alone has been difficult to meet the engineering requirements for achieving accurate positioning and morphological reconstruction of underground cavities in a diverse geological background. Disclosure of Invention The invention aims to solve the problem that the ground penetrating radar signal processing technology cannot effectively adapt to dynamic changes of medium parameters under different geological conditions, and provides a ground penetrating radar image cavity imaging method adapting to different geological conditions. In order to achieve the above purpose, the present invention is realized by the following technical scheme: a ground penetrating radar image cavity imaging method adapting to different geological conditions comprises the following steps: S1, underground original data are collected by a ground penetrating radar, and normalized field intensity after background suppression along with depth dynamic change is constructed based on a statistical principle; S2, taking account of exponential amplitude attenuation of electromagnetic waves along with the increase of depth when the electromagnetic waves propagate underground, introducing a phase consistency analysis technology, and extracting phase consistency structural strength; S3, coupling the normalized field intensity after the background suppression obtained in the step S1 with the phase consistency structural strength obtained in the step S2 to obtain a depth-medium coupling potential energy field; s4, constructing an anisotropic Gaussian kernel function with direction selectivity, wherein the anisotropic Gaussian kernel function with direction selectivity is expressed as an elliptic Gaussian distribution with specific length-width ratio and rotation angle in mathematical morphology; S5, calculating thermodynamic upper boundary manifold projection and thermodynamic lower boundary manifold projection based on the depth-medium coupling potential energy field obtained in the step S3 and the anisotropic Gaussian kernel function with direction selectivity obtained in the step S4; S6, extracting an enhanced gradient characteristic field by calculating the difference between thermodynamic upper boundary manifold projection and thermodynamic lower boundary manifold projection; And S7, establishing a multi-scale entropy weight fusion inverse time imaging based on the self-adaptive weighting strategy of the information entropy and the enhanced gradient characteristic field obtained in the step S6 to obtain a final underground cavity imaging result. Further, in step S1, a background model reflecting the electromagnetic characteristics of the current geological horizon is established, and the abnormal response and background fluctuation in the original signal are effectively decoupled by combining the nonlinear mapping function, so as to obtain the normalized field intensity after the background suppression。 Further, step S2 performs frequency domain decomposition and reconstruction on the radar profile by constructing a multi-scale filter bank to obtain phase consistency structural strength, where the expression is: Wherein, the For phase consistency structural strength, x is a horizontal measuring line position index, z is a vertical depth position index, and is obtained by original data acquired by a ground penetrating radar; selecting proper quantity for total scale quantity by professional according to the target cavity size range and radar frequency bandwidth through pr