Magnetic anomaly signal detection method, system, equipment and medium

Abstract

The invention relates to the technical field of magnetic detection, and discloses a magnetic anomaly signal detection method, system and device and a medium. The method comprises the following steps: acquiring a magnetic anomaly field of a target observation area; magnetic anomaly fields of the three magnetic moments in the geomagnetic field direction of the target observation area along standard magnetic dipoles in the x, y and z directions are obtained and serve as two-dimensional scalar magnetic anomaly fields; vectorizing the three two-dimensional scalar magnetic anomaly fields, arranging the vectorized three two-dimensional scalar magnetic anomaly fields in columns to form a template matrix, and performing matrix decomposition on the template matrix to obtain an orthogonal basis matrix; constructing an orthogonal signal subspace of the magnetic dipole according to the orthogonal basis matrix, and projecting a magnetic anomaly field of the target observation area to the orthogonal signal subspace; and performing magnetic target detection according to the energy distribution condition of the magnetic anomaly field of the target observation area in the orthogonal signal subspace. According to the method, magnetic anomaly detection under a low signal-to-noise ratio can be realized without target depth prior information and a large amount of training data.

Description

Technical Field

[0001] This invention relates to the field of magnetic detection technology, and in particular to a method, system, device and medium for detecting magnetic anomaly signals. Background Technology

[0002] Magnetic anomaly detection is a crucial method for near-surface target detection, offering advantages such as rapid detection and passive, uninterrupted operation. However, in practical testing, the magnetic anomaly signals generated by targets are often affected by geomagnetic field fluctuations, surrounding magnetic noise, and interference from the carrier platform, causing these signals to be buried within noise. The ability to detect magnetic anomaly signals at low signal-to-noise ratios can improve the identification distance and accuracy for deeply buried, small ferromagnetic targets, overcoming the depth limitations of traditional magnetic detection methods.

[0003] Existing methods for detecting magnetic anomalies can be broadly categorized into two types: signal processing-based methods and deep learning-based methods. Signal processing-based methods, such as 1D / 2D orthogonal basis filtering (OBF) and minimum entropy detector (MED), require the depth of the magnetic target as prior information to design the filter. If the depth of the designed filter does not match the actual target depth, it can cause severe signal distortion or missed detections.

[0004] Deep learning-based methods for detecting magnetic anomaly signals mainly involve detecting the presence of a target magnetic anomaly signal using supervised training data. Commonly used network architectures include ResNet and CNN. This method does not require prior information about the target depth, but it requires a large amount of labeled magnetic anomaly data during training. The model training consumes a lot of computational resources, and its generalization ability is limited by the diversity of training data. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device and medium for detecting magnetic anomaly signals, which can solve the above-mentioned technical problems.

[0006] An embodiment of the present invention provides a method for detecting magnetic anomaly signals, comprising the following steps: Acquire the magnetic anomaly field of the target observation area; The magnetic anomaly fields of three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the geomagnetic field direction of the target observation area are obtained as two-dimensional scalar magnetic anomaly fields. The three two-dimensional scalar magnetic anomaly fields are vectorized and arranged in columns to form a template matrix. The template matrix is ​​then decomposed to obtain an orthogonal basis matrix. Construct an orthogonal signal subspace for the magnetic dipole based on the orthogonal basis matrix, and project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace; Magnetic target detection is performed based on the energy distribution of the magnetic anomaly field in the target observation area within the orthogonal signal subspace.

[0007] Furthermore, the two-dimensional scalar magnetic anomaly field is a magnetic anomaly field in the direction of the geomagnetic field of an observation plane within the target observation area, consisting of three standard magnetic dipoles with magnetic moments along the x, y, and z directions, respectively. The magnetic moment of the standard magnetic dipole is 1, and its depth is any depth in the underground space. The size of the observation plane is twice the height of the observation plane perpendicular to the standard magnetic dipole.

[0008] Furthermore, the matrix decomposition method of the template matrix is ​​QR decomposition with column pivoting or singular value decomposition (SVD).

[0009] Furthermore, the orthogonal signal subspace of the magnetic dipole is constructed using the following formula: ; The magnetic anomaly field of the target observation area is projected onto the orthogonal signal subspace using the following formula: ; In the formula, It is an orthogonal basis matrix. For orthogonal signal subspaces, For the magnetic anomaly field of the target observation area, This represents the energy distribution of the magnetic anomaly field in the orthogonal signal subspace.

[0010] Furthermore, the step of performing magnetic target detection based on the energy distribution of the magnetic anomaly field in the target observation area within the orthogonal signal subspace includes: Based on the energy distribution of the magnetic anomaly field in the orthogonal signal subspace of the target observation area, the region in the target observation area whose energy exceeds the preset energy threshold is extracted by the local maximum detection method. Magnetic target detection is performed on regions within the target observation area where the energy exceeds a preset energy threshold.

[0011] Furthermore, the local maximum detection method employs an 8-neighborhood or a non-maximum suppression algorithm with a window larger than 8 neighborhoods.

[0012] Furthermore, the projection of the magnetic anomaly field of the target observation area onto the orthogonal signal subspace includes: Based on the sliding window method, the magnetic anomaly field of the target observation area is projected onto the orthogonal signal subspace.

[0013] Embodiments of the present invention also provide a magnetic anomaly signal detection system, comprising: The first magnetic field acquisition module is used to acquire the magnetic anomaly field of the target observation area; The second magnetic field acquisition module is used to acquire the magnetic anomaly field of three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the geomagnetic field direction of the target observation area, as a two-dimensional scalar magnetic anomaly field. The matrix construction and decomposition module is used to vectorize the three two-dimensional scalar magnetic anomaly fields, arrange them in columns to form a template matrix, and perform matrix decomposition on the template matrix to obtain an orthogonal basis matrix. The first magnetic field projection module is used to construct the orthogonal signal subspace of the magnetic dipole based on the orthogonal basis matrix, and to project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace. The magnetic target detection module is used to detect magnetic targets based on the energy distribution of the magnetic anomaly field in the target observation area in the orthogonal signal subspace.

[0014] Embodiments of the present invention also provide a computer device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the above-described magnetic anomaly signal detection method.

[0015] Embodiments of the present invention also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described magnetic anomaly signal detection method.

[0016] The magnetic anomaly signal detection method provided by this invention has at least the following beneficial effects: This invention approximates the magnetic anomaly field generated by a magnetic target as a magnetic dipole model. Mathematically, the magnetic field signal resides in a linear subspace spanned by magnetic dipole fields in three fundamental directions (x, y, z). Therefore, by projecting the magnetic anomaly field of the target observation area onto an orthogonal signal subspace constructed based on the magnetic anomaly fields of standard magnetic dipoles along the x, y, and z directions in the geomagnetic field direction of the target observation area, signal components conforming to the "magnetic dipole characteristics" can be retained, while interference components in the noise subspace (orthogonal complement space) are eliminated. Finally, the signal components obtained based on this projection are used for magnetic target detection. This process requires no prior information about the target depth; the projection operation relies only on the shape characteristics of the signal rather than the amplitude scale, thus maintaining stable detection statistics under unknown depth conditions and completely eliminating dependence on depth priors. Furthermore, this method is a detection algorithm driven by a physical model, overcoming the "black box" nature of deep learning and requiring no large amounts of training data or computational resources.

[0017] Furthermore, this invention can maintain a high detection probability and a low false alarm rate even in low signal-to-noise ratio or strong interference environments. This is because after orthogonal projection, the signal energy is retained to the maximum extent in the subspace, while the noise (color noise, non-Gaussian noise, non-stationary noise) energy is effectively projected to the orthogonal complementary space, thereby improving the output signal-to-noise ratio and realizing magnetic anomaly detection under low signal-to-noise ratio conditions. Attached Figure Description

[0018] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0019] Figure 1 A flowchart illustrating a magnetic anomaly signal detection method provided by the present invention; Figure 2 A schematic diagram of a magnetic anomaly field distribution generated by a unit magnetic dipole, provided by the present invention; Figure 3 A schematic diagram of a magnetic measurement experiment provided by the present invention; Figure 4 This invention provides a two-dimensional magnetogram of a normalized unit magnetic dipole generating a magnetic anomaly field. Figure 5 A schematic diagram of signal projection energy distribution provided by the present invention; Figure 6 This is a schematic diagram of a magnetic anomaly signal detection result provided by the present invention. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below in conjunction with specific embodiments and corresponding drawings. Obviously, the described embodiments are only a part of the embodiments of this invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0021] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] One embodiment of the present invention relates to a method for detecting magnetic anomaly signals. The specific process of the magnetic anomaly signal detection method in this embodiment can be as follows: Figure 1 As shown, it includes: Step 101: Obtain the magnetic anomaly field of the target observation area.

[0023] Step 102: Obtain the magnetic anomaly fields of the three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the direction of the geomagnetic field of the target observation area, as two-dimensional scalar magnetic anomaly fields.

[0024] Step 103: Vectorize the three two-dimensional scalar magnetic anomaly fields and arrange them in columns to form a template matrix. Then, perform matrix decomposition on the template matrix to obtain an orthogonal basis matrix.

[0025] Step 104: Construct the orthogonal signal subspace of the magnetic dipole based on the orthogonal basis matrix, and project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace.

[0026] Step 105: Magnetic target detection is performed based on the energy distribution of the magnetic anomaly field in the orthogonal signal subspace of the target observation area.

[0027] The implementation details of the magnetic anomaly signal detection method in this embodiment are described below. The following content is only for the convenience of understanding and is not necessary for implementing this solution.

[0028] First, regarding the magnetic anomaly field generated by near-surface targets, when the distance between the magnetic field sensor and the target is greater than three times the target's maximum size, the magnetic anomaly field generated by the target can be approximated as a point-like magnetic dipole model. In a Cartesian coordinate system, the magnetic moment of a single magnetic dipole is... The position of the magnetic dipole is Observation point The magnetic field vector generated by the magnetic dipole The calculation is as follows:

[0029] (1); In the formula, The permeability of free space, The distance vector between the observation point and the magnetic dipole. Distance vector The modulus, It is a unit distance vector.

[0030] For scalar magnetic field sensors, the measurement results It is the projection of the magnetic anomaly field generated by the target onto the direction of the Earth's magnetic field, where the unit vector direction of the Earth's magnetic field is... ,get: (2); Next, the projection operator of the magnetic dipole field is constructed. According to formula (1), it can be expanded into the following form:

[0031] (3); In the formula, The magnetic moment along the x-direction The magnetic dipole field produced by a unit magnetic dipole. The magnetic moment is along the y-direction The magnetic dipole field produced by a unit magnetic dipole. The magnetic moment is along the z-direction The magnetic dipole field produced by a unit magnetic dipole. , , It is a unit distance vector Three components in a rectangular coordinate system.

[0032] Combining formulas (2) and (3), the magnetic anomaly field measured at the observation plane can be summarized as follows: (4); Therefore, the magnetic anomaly field at the observation plane It can be represented as the field of three standard dipole fields in the direction of the geomagnetic field. , and The weighted superposition form, with each component corresponding to the three directional components of the dipole moment, reflects the linear superposition relationship between the magnetic anomaly field and the dipole source. The template matrix A is constructed by vectorizing the projections of the three unit dipoles onto the Earth's magnetic field. (5); According to formulas (3)-(4), the magnetic field generated by a magnetic dipole along any direction can be regarded as a linear combination of the magnetic moment along the three standard dipole fields of x, y, and z.

[0033] Regarding the orthogonality of the magnetic moment along the three standard dipole fields x, y, and z: (6); The diagonal integral on a sphere of radius centered at the magnetic dipole satisfies ,therefore: (7); Similarly, we can obtain: (8); (9); Therefore, the in-spherical area (angular direction) or the area on a sphere with a fixed radius r is considered. During inner product, the three dipole fields are mutually orthogonal. Equivalently, they are angular (spherical) harmonic bases. Vector fields are therefore mutually orthogonal. However, in actual tests, the region where the magnetic field is observed is generally in a two-dimensional plane z=C, rather than in a spherical space. In this case, the three standard dipole fields x, y, and z are not strictly orthogonal, but approximately orthogonal.

[0034] Figure 2The results show the field distribution of three standard dipole fields (x, y, z) located at the origin when z = 2m. (a) represents the magnetic anomaly field generated by a unit magnetic dipole with magnetic moment along the x-direction; (b) represents the magnetic anomaly field generated by a unit magnetic dipole with magnetic moment along the y-direction; and (c) represents the magnetic anomaly field generated by a unit magnetic dipole with magnetic moment along the z-direction. The magnetic inclination of the Earth's magnetic field is... Magnetic declination is The observation plane ranges as follows: x: -2 to 2m, y: -2 to 2m. The normalized correlation coefficient is used... Determining their orthogonality, we obtain: , , This indicates that on the observed two-dimensional horizontal plane, the cross-correlation coefficients of the three standard dipole fields x, y, and z are all less than 0.1, and are approximately orthogonal. The template subspace can be orthogonalized through QR decomposition.

[0035] , (10); In the formula, Let these be three orthogonal bases of the template subspace. An upper triangular matrix is ​​not used for projection.

[0036] This embodiment utilizes the subspace of the constructed magnetic dipole field to project the observed magnetic field data into this subspace, and detects magnetic anomaly signals based on the projection energy. The specific process is as follows:

[0037] (1) Use a scalar magnetic field sensor to collect spatial magnetic field data in the horizontal test area, perform low-pass filtering and background trend removal preprocessing on the raw magnetic field data to obtain the magnetic anomaly field of the observation area. (2) Constructing the orthogonal signal subspace Q of the magnetic dipole involves: calculating the two-dimensional scalar magnetic anomaly field generated by the unit magnetic dipole along the x, y, and z directions on a small observation plane. , and ,Will , and After vectorization, the vectors are arranged column-wise to form a template matrix A. Matrix decomposition is then performed on template matrix A to extract the orthogonal basis matrix Q, thus obtaining the orthogonal signal subspace of the magnetic dipoles. ; Among them, there are three unit magnetic dipoles along the x, y, and z directions respectively, and ideal (standard) magnetic dipoles with a magnetic moment of 1. The depth information of the unit magnetic dipoles can be set at any depth in the underground space (default is -1m). The size of the observation plane is twice the vertical height between the observation plane and the unit magnetic dipole, ensuring that the magnetic anomaly field energy coverage in the observation area exceeds 90% of the magnetic anomaly field generated by the magnetic dipole.

[0038] Furthermore, QR decomposition with column pivoting or singular value decomposition (SVD) is used to ensure that the obtained orthogonal basis Q values ​​are stable and have full column rank.

[0039] (3) It can be viewed as the projection matrix of the orthogonal signal subspace, and the projection matrix of the noise subspace is... Where I is the identity matrix; (4) Project the observed magnetic anomaly field vector obtained in step (1) onto the orthogonal signal subspace. The projected signal distribution is obtained; In the process of projecting the observed magnetic anomaly field vector into the signal subspace, a sliding window method is required because the observation interval of the template subspace is smaller than the entire magnetic measurement interval.

[0040] (5) Construct a statistical detection quantity based on the projection energy of the signal subspace, calculate the projection energy at each position in the two-dimensional horizontal measurement area, and form a projection energy distribution map; extract significant energy centers through local maximum detection, set an energy threshold, and make a judgment on areas exceeding the threshold, marking them as potential magnetic anomaly signal areas, thereby realizing the detection of magnetic anomaly targets.

[0041] The local maximum detection method employs a non-maximum suppression algorithm with an 8-neighborhood or larger window to accurately locate the energy center of potential magnetic anomaly targets. Methods for setting the energy threshold include, but are not limited to: the Otsu adaptive threshold method, the constant false alarm rate (CFAR) criterion, or the k-times standard deviation method based on noise energy statistics in target-free areas.

[0042] As shown below: (14); In the formula, To count the number of tests, This is the energy threshold.

[0043] As can be seen, this invention constructs a magnetic anomaly signal detection method based on subspace projection, which can achieve magnetic anomaly signal detection under low signal-to-noise ratio: the magnetic dipole field generated by the target is represented as the field of the standard dipole field along the x, y, and z directions of the geomagnetic field. , and A template matrix is ​​constructed using a weighted superposition method, and an orthogonal subspace of the magnetic anomaly signal is obtained through QR decomposition. The observed magnetic field data is projected onto this subspace, the projected energy of the survey area is estimated, and significant response centers are extracted from the local maximum detection map. A threshold is set, and regions exceeding the threshold are considered potential magnetic anomaly signal regions. This method can reliably detect weak anomalies in targets under strong noise backgrounds without requiring prior information on target depth, which is significant for improving target detection depth, increasing detection probability, and reducing false alarm rate in magnetic surveys.

[0044] In a specific embodiment, the magnetic anomaly signal detection method of the present invention is implemented through the following process: A UAV magnetic measurement system based on an optically pumped magnetic field sensor collects scalar magnetic field data in space. The sampling rate of the magnetic field data is 160Hz, the UAV's flight altitude is 1.5m, and the UAV's flight trajectory is shown in Figure N. After low-pass filtering (cutoff frequency of 1Hz) and removing the trend term, the magnetic field distribution data of the measurement area is obtained as follows: Figure 3 As shown, (a) is the flight trajectory of the UAV magnetic measurement system, and (b) is the magnetic anomaly field.

[0045] As can be seen from the figure, due to the presence of the drone interference magnetic field, the preprocessed data is a superposition of the drone interference magnetic field, the target magnetic anomaly field, and geomagnetic noise, and the target magnetic anomaly signal cannot be delineated from the magnetic anomaly field map.

[0046] An orthogonal subspace was constructed, with the observation height 1.5m above the ground, the observation area being 5m x 5m, the geomagnetic inclination at 52.3°, and the geomagnetic declination at -8.5°. o The unit magnetic dipole is 1m deep and located at the center of the observation area. Estimate the magnetic anomaly fields generated by the unit magnetic dipole along the x, y, and z directions, respectively. , and ,like Figure 4 As shown, (a) represents the magnetic anomaly field of a unit magnetic dipole with magnetic moment along the x-direction, (b) represents the magnetic anomaly field of a unit magnetic dipole with magnetic moment along the y-direction, and (c) represents the magnetic anomaly field of a unit magnetic dipole with magnetic moment along the z-direction. A template matrix A is constructed based on the magnetic anomaly fields generated by three unit magnetic dipoles, and an orthogonal subspace is obtained through QR decomposition. The energy distribution of the observed magnetic anomaly field after projection into the orthogonal subspace is shown below. Figure 5 As shown. With a threshold set to 0.09, six anomaly centers can be identified using the local maximum detection method. Therefore, the magnetic anomaly region can be determined as follows: Figure 5 Anomalies 1-6 were identified as the main anomalies; other areas were pure noise. The test results were consistent with the actual test area. Anomalies 1, 2, and 3 in the test area consisted of three iron pipes, while anomalies 4, 5, and 6 consisted of small iron fragments. From... Figure 6As can be seen from the first sub-plot, magnetic anomaly signals 3, 4, 5, and 6 are completely submerged in noise and cannot be directly detected from the magnetogram. (Comparison) Figure 6 The second sub-figure shows the processing results of the magnetic anomaly detection algorithm based on subspace projection. The energy of the magnetic anomaly centers 3, 4, 5, and 6 is higher than the threshold, so they can be effectively detected, verifying that the algorithm can detect magnetic anomaly signals under low signal-to-noise ratio.

[0047] The steps of the various methods described above are only for clarity. In practice, they can be combined into one step or some steps can be split into multiple steps. As long as they include the same logical relationship, they are all within the protection scope of this invention.

[0048] Another embodiment of the present invention relates to a magnetic anomaly signal detection system. The implementation details of this magnetic anomaly signal detection system are described below. The following details are provided for ease of understanding and are not essential for implementing this solution. The magnetic anomaly signal detection system of this embodiment includes: The first magnetic field acquisition module is used to acquire the magnetic anomaly field of the target observation area; The second magnetic field acquisition module is used to acquire the magnetic anomaly field of three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the geomagnetic field direction of the target observation area, as a two-dimensional scalar magnetic anomaly field. The matrix construction and decomposition module is used to vectorize the three two-dimensional scalar magnetic anomaly fields, arrange them in columns to form a template matrix, and perform matrix decomposition on the template matrix to obtain an orthogonal basis matrix. The first magnetic field projection module is used to construct the orthogonal signal subspace of the magnetic dipole based on the orthogonal basis matrix, and to project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace. The magnetic target detection module is used to detect magnetic targets based on the energy distribution of the magnetic anomaly field in the target observation area in the orthogonal signal subspace.

[0049] It is not difficult to see that this embodiment is a system embodiment corresponding to the above method embodiments, and this embodiment can be implemented in conjunction with the above method embodiments. The relevant technical details and technical effects mentioned in the above embodiments are still valid in this embodiment, and will not be repeated here to reduce repetition. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the above embodiments.

[0050] It is worth mentioning that all modules involved in this embodiment are logical modules. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by this invention; however, this does not mean that other units are absent from this embodiment.

[0051] Another embodiment of the present invention relates to a computer device, comprising: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the magnetic anomaly signal detection method of the above embodiments.

[0052] The memory and processor are connected via a bus, which can include any number of interconnecting buses and bridges, connecting various circuits of one or more processors and memories. The bus can also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and will not be described further herein. The bus interface provides an interface between the bus and the transceiver. The transceiver can be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by the processor is transmitted over the wireless medium via an antenna, which further receives data and transmits it to the processor.

[0053] The processor manages the bus and general processing, and also provides various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory is used to store data used by the processor during operation.

[0054] Another embodiment of the present invention relates to a computer-readable storage medium storing a computer program. When executed by a processor, the computer program implements the method embodiments described above.

[0055] That is, those skilled in the art will understand that all or part of the steps in the methods of the above embodiments can be implemented by a program instructing related hardware. This program is stored in a storage medium and includes several instructions to cause a device (which may be a microcontroller, chip, etc.) or processor to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0056] Those skilled in the art will understand that the above embodiments are specific embodiments for implementing the present invention, and in practical applications, various changes can be made to them in form and detail without departing from the spirit and scope of the present invention.

Claims

1. A method for detecting magnetic anomaly signals, characterized in that, The method includes: Acquire the magnetic anomaly field of the target observation area; The magnetic anomaly fields of three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the geomagnetic field direction of the target observation area are obtained as two-dimensional scalar magnetic anomaly fields. The three two-dimensional scalar magnetic anomaly fields are vectorized and arranged in columns to form a template matrix. The template matrix is ​​then decomposed to obtain an orthogonal basis matrix. Construct an orthogonal signal subspace for the magnetic dipole based on the orthogonal basis matrix, and project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace; Magnetic target detection is performed based on the energy distribution of the magnetic anomaly field in the target observation area within the orthogonal signal subspace.

2. The magnetic anomaly signal detection method according to claim 1, characterized in that, The two-dimensional scalar magnetic anomaly field is the magnetic anomaly field in the direction of the geomagnetic field of an observation plane within the target observation area, consisting of three standard magnetic dipoles with magnetic moments along the x, y, and z directions, respectively. The magnetic moment of the standard magnetic dipole is 1, and its depth is any depth in the underground space. The size of the observation plane is twice the height of the observation plane perpendicular to the standard magnetic dipole.

3. The method for detecting magnetic anomaly signals according to claim 1, characterized in that, The matrix decomposition method for the template matrix is ​​either QR decomposition with column pivoting or Singular Value Decomposition (SVD).

4. The method for detecting magnetic anomaly signals according to claim 1, characterized in that, The orthogonal signal subspace of a magnetic dipole can be constructed using the following formula: ； The magnetic anomaly field of the target observation area is projected onto the orthogonal signal subspace using the following formula: ； In the formula, It is an orthogonal basis matrix. For orthogonal signal subspaces, For the magnetic anomaly field of the target observation area, This represents the energy distribution of the magnetic anomaly field in the orthogonal signal subspace.

5. The method for detecting magnetic anomaly signals according to claim 1, characterized in that, The magnetic target detection based on the energy distribution of the magnetic anomaly field in the target observation area within the orthogonal signal subspace includes: Based on the energy distribution of the magnetic anomaly field in the orthogonal signal subspace of the target observation area, the region in the target observation area whose energy exceeds the preset energy threshold is extracted by the local maximum detection method. Magnetic target detection is performed on regions within the target observation area where the energy exceeds a preset energy threshold.

6. The method for detecting magnetic anomaly signals according to claim 5, characterized in that, The local maximum detection method employs an 8-neighborhood or a non-maximum suppression algorithm with a window larger than 8 neighborhoods.

7. The method for detecting magnetic anomaly signals according to claim 2, characterized in that, The projection of the magnetic anomaly field of the target observation area onto the orthogonal signal subspace includes: Based on the sliding window method, the magnetic anomaly field of the target observation area is projected onto the orthogonal signal subspace.

8. A magnetic anomaly signal detection system, characterized in that, The system includes: The first magnetic field acquisition module is used to acquire the magnetic anomaly field of the target observation area; The second magnetic field acquisition module is used to acquire the magnetic anomaly field of three standard magnetic dipoles with magnetic moments along the x, y, and z directions in the geomagnetic field direction of the target observation area, as a two-dimensional scalar magnetic anomaly field. The matrix construction and decomposition module is used to vectorize the three two-dimensional scalar magnetic anomaly fields, arrange them in columns to form a template matrix, and perform matrix decomposition on the template matrix to obtain an orthogonal basis matrix. The first magnetic field projection module is used to construct the orthogonal signal subspace of the magnetic dipole based on the orthogonal basis matrix, and to project the magnetic anomaly field of the target observation area onto the orthogonal signal subspace. The magnetic target detection module is used to detect magnetic targets based on the energy distribution of the magnetic anomaly field signal subspace in the target observation area.

9. A computer device, characterized in that, include: At least one processor; And a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to perform the magnetic anomaly signal detection method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the magnetic anomaly signal detection method as described in any one of claims 1 to 7.

Images