CN-121982213-A - Three-dimensional imaging method for underground cable based on terahertz data and X-ray data

CN121982213ACN 121982213 ACN121982213 ACN 121982213ACN-121982213-A

Abstract

The three-dimensional imaging method of the underground cable based on the terahertz data and the X-ray data comprises the steps of solving dielectric constant distribution and material density distribution according to the terahertz data and the X-ray data, establishing an association relation between the terahertz data and the X-ray data, determining projection reconstruction initial parameters, constructing a fitness function according to the projection data and the projection reconstruction initial parameters, reconstructing the dielectric constant distribution and the material density distribution through a quantum genetic optimization algorithm based on the fitness function and the association relation to obtain an initial reconstruction result, performing cross iterative optimization on the dielectric constant distribution and the material density distribution by utilizing a regularization term constructed by the initial reconstruction result until convergence, and performing data fusion and feature optimization after generating an X-ray reconstruction image and a terahertz reconstruction image based on the converged dielectric constant distribution and the material density distribution to obtain a three-dimensional image of the underground cable. Thus, a three-dimensional image of the underground cable with high precision is generated by fusing the optimized bimodal image data.

Inventors

TANG QI
CHEN ZHIPING

Assignees

广东电网有限责任公司佛山供电局

Dates

Publication Date: 20260505
Application Date: 20260127

Claims (10)

1. A three-dimensional imaging method for a subsurface cable based on terahertz data and X-ray data, the method comprising: According to terahertz data and X-ray data for detecting an underground cable, solving dielectric constant distribution and material density distribution, establishing an association relation between the dielectric constant distribution and the material density distribution, and determining projection reconstruction initial parameters, wherein the X-ray data comprises projection data; Constructing an adaptability function according to the projection data and the projection reconstruction initial parameters, and reconstructing the dielectric constant distribution and the material density distribution through a quantum genetic optimization algorithm based on the adaptability function and the association relation to obtain an initial reconstruction result; constructing a regularization term by using the initial reconstruction result, and performing cross iterative optimization on the dielectric constant distribution and the material density distribution by adopting the regularization term until convergence; And generating an X-ray reconstruction image and a terahertz reconstruction image based on the converged dielectric constant distribution and the converged material density distribution, and then carrying out data fusion and feature optimization to obtain a three-dimensional image of the underground cable.
2. The three-dimensional imaging method of a subsurface cable based on terahertz data and X-ray data according to claim 1, wherein the association relation expression of the material density distribution and the dielectric constant distribution is: Wherein, the Representing the density distribution of the material, obtained for solving the X-ray attenuation model, Representing the dielectric constant distribution, for solving the terahertz propagation equation, The dielectric constant of the vacuum is indicated, Indicating the polarization rate of the light, and, Representing the avogalileo constant, Representing the molar mass of the molecule.
3. The method of three-dimensional imaging of a subsurface cable based on terahertz data and X-ray data according to claim 1, wherein the terahertz data includes terahertz original signals and terahertz original images, the X-ray data includes X-ray original signals and X-ray original images, the step of determining projection reconstruction initial parameters includes: Constructing a first time-frequency energy matrix of the terahertz original signal and a second time-frequency energy matrix of the X-ray original signal; Performing tensor decomposition on the first time-frequency energy matrix and the second time-frequency energy matrix to obtain an initial value of registration parameters; And according to the initial value of the registration parameter, nonlinear registration optimization is carried out on the terahertz initial image and the X-ray initial image, so as to obtain the projection reconstruction initial parameter.
4. The method for three-dimensional imaging of a subsurface cable based on terahertz data and X-ray data according to claim 1, wherein the step of reconstructing the dielectric constant distribution and the material density distribution by a quantum genetic optimization algorithm based on the fitness function and the association relation to obtain an initial reconstruction result comprises: initializing each preset quantum chromosome through quantum chromosome coding, and taking each initialized quantum chromosome as a current quantum chromosome of a current iteration round; Calculating an error with the projection data according to a current reconstruction result generated by a current quantum chromosome, and carrying out fitness grading by adopting the fitness function by combining the association relation between the dielectric constant distribution and the material density distribution; If the iteration convergence is not determined according to the fitness score, optimizing the parameter state of the current quantum chromosome, and entering the next iteration until the iteration convergence is determined, and taking the current reconstruction result after the convergence as the initial reconstruction result.
5. The three-dimensional imaging method of a subsurface cable based on terahertz data and X-ray data according to claim 4, wherein the fitness function has an expression of: Wherein, the The fitness score is represented by a score of the fitness, The current reconstruction result is indicated and, The projection operator is represented by a representation of the projection operator, The projection data is represented by a representation of the projection data, Representing the regularization weight factor, And the total variation regular term representing the reconstruction result is used for restraining the smoothness of the image and the edge preservation and combining the association relation between the dielectric constant distribution and the material density distribution.
6. The method for three-dimensional imaging of a subsurface cable based on terahertz data and X-ray data according to claim 1, wherein the expression for cross-iterative optimization of the dielectric constant distribution and the material density distribution is: Wherein, the Represent the first The dielectric constant distribution of the next iteration, Represent the first Dielectric constant distribution for +1 iteration, Represent the first The material density distribution of the next iteration, Represent the first Material density distribution for +1 iteration, 、 The learning rate is indicated as being indicative of the learning rate, Representing the gradient of the current material density profile, Representing the gradient of the current dielectric constant distribution, Representing the gradient of the error term.
7. The method for three-dimensional imaging of an underground cable based on terahertz data and X-ray data according to claim 1, wherein the step of performing data fusion and feature optimization to obtain a three-dimensional image of the underground cable comprises the steps of: performing total variation regularization fusion on the X-ray reconstructed image and the terahertz reconstructed image according to a preset energy functional to obtain a fusion image; and carrying out multi-scale feature optimization on the fusion image by adopting an alternate direction multiplier method to obtain a three-dimensional image of the underground cable.
8. An underground cable three-dimensional imaging device based on terahertz data and X-ray data fusion, which is characterized by comprising: The physical field distribution solving module is used for solving dielectric constant distribution and material density distribution according to terahertz data and X-ray data for detecting the underground cable, establishing an association relationship between the dielectric constant distribution and the material density distribution, and determining projection reconstruction initial parameters, wherein the X-ray data comprises projection data; the initial reconstruction result determining module is used for constructing an adaptability function according to the projection data and the projection reconstruction initial parameters, reconstructing the dielectric constant distribution and the material density distribution through a quantum genetic optimization algorithm based on the adaptability function and the association relation, and obtaining an initial reconstruction result; the physical field distribution optimization module is used for constructing a regularization term by utilizing the initial reconstruction result, and carrying out cross iterative optimization on the dielectric constant distribution and the material density distribution by adopting the regularization term until convergence; And the three-dimensional image determining module is used for generating an X-ray reconstruction image and a terahertz reconstruction image based on the converged dielectric constant distribution and the converged material density distribution, and then carrying out data fusion and feature optimization to obtain a three-dimensional image of the underground cable.
9. A storage medium having stored therein computer readable instructions which, when executed by one or more processors, cause the one or more processors to perform the steps of the method for three-dimensional imaging of a subsurface cable based on terahertz data and X-ray data as set forth in any one of claims 1 to 7.
10. A computer device includes one or more processors and a memory; Stored in the memory are computer readable instructions which, when executed by the one or more processors, perform the steps of the method of three-dimensional imaging of a subsurface cable based on terahertz data and X-ray data as claimed in any one of claims 1 to 7.

Description

Three-dimensional imaging method for underground cable based on terahertz data and X-ray data Technical Field The application relates to the technical field of underground cable detection, in particular to a three-dimensional imaging method of an underground cable based on terahertz data and X-ray data. Background Underground cables are an important component of urban infrastructure, and their operating state is directly related to the stability of the power system and the safety of urban operation. In order to realize efficient detection and fault diagnosis of underground cables, image imaging technology is widely introduced, and through visualization of underground structures, operation and maintenance efficiency and intelligent level are improved. At present, a multi-mode data fusion technology, such as fusion of geological radar images and infrared thermal imaging, is often adopted in an underground cable imaging system so as to acquire more comprehensive cable embedded information and abnormal characteristics. With the development of deep learning, the image fusion quality and the degree of automation are continuously improved, and the method has become the research focus in the field. However, existing fusion methods still have certain limitations. On one hand, complementary characteristics among the multi-mode data are not fully mined and utilized, so that the fusion result information expression capability is limited, and on the other hand, the resolution of the fusion image is low, details are unclear, the processing speed is low, and the actual requirements of underground cable detection on high precision and high real-time performance are difficult to meet. Disclosure of Invention The present application aims to solve at least one of the above technical drawbacks, and in particular, the technical drawback of poor image fusion effect of multi-modal data in the prior art. In a first aspect, the application provides a three-dimensional imaging method of a subsurface cable based on terahertz data and X-ray data, the method comprising: According to terahertz data and X-ray data for detecting the underground cable, solving dielectric constant distribution and material density distribution, establishing an association relationship between the dielectric constant distribution and the material density distribution, and determining projection reconstruction initial parameters, wherein the X-ray data comprises projection data; Constructing an fitness function according to the projection data and the projection reconstruction initial parameters, reconstructing dielectric constant distribution and material density distribution through a quantum genetic optimization algorithm based on the fitness function and the association relation, and obtaining an initial reconstruction result; Constructing a regularization term by using an initial reconstruction result, and performing cross iterative optimization on dielectric constant distribution and material density distribution by adopting the regularization term until convergence; And generating an X-ray reconstructed image and a terahertz reconstructed image based on the converged dielectric constant distribution and the converged material density distribution, and then carrying out data fusion and feature optimization to obtain a three-dimensional image of the underground cable. In one embodiment, the relationship between the material density distribution and the dielectric constant distribution is expressed as: Wherein, the Representing the density distribution of the material, obtained by solving the X-ray attenuation model,Representing the dielectric constant distribution, in order to solve the terahertz propagation equation,The dielectric constant of the vacuum is indicated,Indicating the polarization rate of the light, and,Representing the avogalileo constant,Representing the molar mass of the molecule. In one embodiment, the terahertz data comprises terahertz original signals and terahertz initial images, the X-ray data comprises X-ray original signals and X-ray initial images, and the step of determining projection reconstruction initial parameters comprises the following steps: Constructing a first time-frequency energy matrix of the terahertz original signal and a second time-frequency energy matrix of the X-ray original signal; Tensor decomposition is carried out on the first time-frequency energy matrix and the second time-frequency energy matrix, and a registration parameter initial value is obtained; And according to the initial value of the registration parameter, performing nonlinear registration optimization on the terahertz initial image and the X-ray initial image to obtain the projection reconstruction initial parameter. In one embodiment, reconstructing the dielectric constant distribution and the material density distribution by a quantum genetic optimization algorithm based on the fitness function and the association relation to obtain an ini