Magnetic Anomaly Sensing Method and Related Device Based on Improved RLMD and Support Vector Machine

Abstract

The present invention provides a magnetic anomaly perception method and related device based on improved RLMD and support vector machine, including: acquiring original data s(t), decomposing the original data s(t) by using the RLMD algorithm to obtain a plurality of PF components; calculating the permutation entropy for each of the plurality of PF components respectively, arranging all the permutation entropy values from large to small, screening out the latter n PF components, where n is adaptively determined by the three - quartile, performing linear reconstruction on the PF components to obtain a linearly reconstructed signal, and removing the trend term from the linearly reconstructed signal to obtain the processed static magnetic signal m(t); constructing a corresponding feature vector group [D, K, P]; establishing a training sample data set; inputting the feature vector group corresponding to the signal into the trained support vector machine to determine whether there is a target magnetic anomaly. Applying the technical solution of the present invention can solve the technical problem in the prior art that when performing signal recognition, usually by artificially setting a screening threshold, this method is difficult to fully suppress noise, resulting in low target recognition accuracy.

Description

Technical Field

[0001] The present invention relates to the technical field of airborne magnetic anomaly detection signal processing, and in particular to a magnetic anomaly perception method and related device based on improved RLMD and support vector machine. Background Art

[0002] Targets with certain ferromagnetic properties, such as pipelines and ore deposits, will cause local magnetic field distortion, i.e., magnetic anomaly, under the magnetization of the geomagnetic field. As an exposure source of invisible targets, accurately measuring the target magnetic anomaly is of great significance for the detection and identification of targets. With the continuous improvement of magnetic sensor technology in recent years, magnetic anomaly perception technology has become the bottleneck restricting the detection ability. Magnetic anomaly perception technology faces two main practical problems: First, the intensity of the target magnetic anomaly signal decreases rapidly with the increase of the detection distance, resulting in a very limited detection distance; Second, the target magnetic anomaly signal is easily affected by various factors such as airborne platforms, geological signals, and geomagnetic daily variations, and is usually submerged in background noise, which restricts the further development and application of magnetic anomaly detection. Therefore, how to perceive the weak magnetic anomaly signal of the target from the strong noise background is the key to magnetic anomaly detection technology.

[0003] As an adaptive signal processing method, Robust Local Mean Decomposition (RLMD) has been widely used in fault diagnosis and other aspects. Existing technologies such as CN112557966 A, a method for identifying transformer winding looseness based on local mean decomposition and support vector machine, CN115792523 A, a method for detecting low-voltage DC arc faults, CN 110636053A, a method for detecting network attacks based on local mean decomposition and support vector machine, can achieve the detection and identification of signals. However, when identifying signals in the existing technology, the screening threshold is usually set artificially, and this method is difficult to fully suppress noise, resulting in low target recognition accuracy. Summary of the Invention

[0004] The present invention provides a magnetic anomaly perception method and related device based on improved RLMD and support vector machine, which can solve the technical problem that in the existing technology, when identifying signals, the screening threshold is usually set artificially, and this method is difficult to fully suppress noise, resulting in low target recognition accuracy.

[0005] According to one aspect of the present invention, a magnetic anomaly perception method based on improved RLMD and support vector machine is provided. The magnetic anomaly perception method based on improved RLMD and support vector machine includes: Step 1, using an airborne magnetic survey platform to fly along a preset route to obtain the original data s(t) that may contain the target static magnetic anomaly signal, and decomposing the original data s(t) using the RLMD algorithm to obtain multiple PF components; Step 2, calculating the permutation entropy for each of the obtained multiple PF components, arranging all the permutation entropy values of the PF components from largest to smallest, screening out the latter n PF components, where n is adaptively determined by the three - quartiles of all the permutation entropy values, linearly reconstructing the PF components according to the latter n PF components to obtain a linearly reconstructed signal, and removing the trend term from the linearly reconstructed signal to obtain the processed static magnetic signal m(t); Step 3, calculating the morphological fractal dimension D, kurtosis value K, and peak - to - peak value P of the processed static magnetic signal m(t), and constructing the corresponding feature vector group [D, K, P]; Step 4, establishing a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform, setting a label S for the signal, setting the label S of the signal with target magnetic anomaly to 1, and setting the label S of the signal without target magnetic anomaly to 0, training the support vector machine based on the training sample data set to obtain the optimal magnetic anomaly perception strategy; Step 5, inputting the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judging whether there is a target magnetic anomaly in the original signal according to the output result to realize magnetic anomaly perception.

[0006] Further, in Step 1, decomposing the original data s(t) using the RLMD algorithm to obtain multiple PF components specifically includes: Step 1.1, determining all local extreme points in the original data s(t); Step 1.2, calculating and obtaining the first local mean function m 11 (t) and the first local envelope estimate function a 11 (t) according to multiple local extreme points; Step 1.3, subtracting the first local mean function m 11 (t) from the original data s(t) to obtain the first intermediate function h 11 (t); Step 1.4, dividing the first intermediate function h 11 (t) by the first local envelope estimate function a 11 (t) to obtain the demodulated first demodulated function s 11 (t); Step 1.5, taking the first intermediate function h 11 (t) as the original data, performing the steps from extreme points to subtracting the local mean, repeating Step 1.1 to Step 1.4, and obtaining the second local mean function m 12 (t), the second local envelope estimate function a 12 (t), the second intermediate function h12 (t) and the second demodulated function s 12 (t); Step 1.6, use the second intermediate function h 12 (t) as the original data, perform the steps from the extreme point to subtracting the local mean, repeat Steps 1.1 to 1.4 to obtain the third local mean function m 13 (t), the third local envelope estimate function a 13 (t), the third intermediate function h 13 (t) and the third demodulated function s 13 (t); Step 1.7, and so on, repeat the steps from the extreme point to subtracting the local mean until a pure frequency modulation signal s 1n (t) appears. Calculate the first PF component according to multiple local envelope estimate functions and the pure frequency modulation signal s 1n (t); Step 1.8, use u1(t) = s(t) - PF1(t) as the original signal for the next decomposition, repeat Steps 1.1 to 1.7 to obtain the second PF component; Step 1.9, repeat Steps 1.1 to 1.8 to successively obtain the third PF component, …, the Nth PF component, calculate to obtain u2(t), …, u k (t), until u k (t) stops when there is only one extreme point left or it becomes a monotonic function.

[0007] Further, the local mean m 1i (t) in any local mean function can be calculated according to , and the local envelope estimate value a 11 (t) of any local envelope estimate function can be calculated according to , where n k is the kth extreme point, and n k+1 is the (k + 1)th extreme point.

[0008] Further, the permutation entropy PE of any PF component can be calculated according to , where p j is the probability of the occurrence of permutation j in the PF component, k = 1, 2, …, N, and permutation j is the jth way of the permutation mode of the PF component.

[0009] Further, the processed static magnetic signal m(t) can be calculated according to m(t) = lm(t) - L(t), where lm(t) is the linear reconstruction signal and L(t) is the linear trend term.

[0010] Further, u2(t), …, u k (t) can be calculated according to .

[0011] Further, the morphological fractal dimension D in the feature vector group [D, K, P] = log2N m (δ m ) / (-log2δ m ), where N m (δ m ) represents the number of corresponding non-empty boxes, and the side length of the box is δ m (m = 1, 2,..., M); the kurtosis value in the feature vector group [D, K, P] where x is the amplitude of the processed static magnetic signal m(t), μ is the average value of the processed static magnetic signal m(t), and σ is the standard deviation of the processed static magnetic signal m(t); the peak-to-peak value P in the feature vector group [D, K, P] = max(x) - min(x), where max(x) is the maximum value of the processed static magnetic signal m(t) and min(x) is the minimum value of the processed static magnetic signal m(t).

[0012] According to another aspect of the present invention, a magnetic anomaly perception system based on VMD-1DCNN is provided. The magnetic anomaly perception system based on VMD-1DCNN uses the magnetic anomaly perception method based on the improved RLMD and support vector machine as described above for magnetic anomaly perception. The magnetic anomaly perception system includes: a PF component calculation module, which is used to fly along a preset route by an airborne magnetic survey platform to obtain the original data s(t) that may contain the target static magnetic anomaly signal, and decompose the original data s(t) using the RLMD algorithm to obtain a plurality of PF components; a static magnetic signal acquisition module, which is used to calculate the permutation entropy of each of the obtained plurality of PF components, arrange the permutation entropy values of all PF components from large to small, screen out the latter n PF components, where n is adaptively determined by the three-quartile of all permutation entropy values, linearly reconstruct the PF components according to the latter n PF components to obtain a linearly reconstructed signal, and remove the trend term from the linearly reconstructed signal to obtain the processed static magnetic signal m(t); a feature vector group construction module, which is used to calculate the morphological fractal dimension D, kurtosis value K, and peak-to-peak value P of the processed static magnetic signal m(t), and construct the corresponding feature vector group [D, K, P]; a training sample data set construction module, which is used to establish a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform, set a label S for the signal, set the signal label S with target magnetic anomaly as 1, and set the label S of the signal without target magnetic anomaly as 0, and train the support vector machine based on the training sample data set to obtain the optimal magnetic anomaly perception strategy; a magnetic anomaly perception determination module, which is used to input the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judge whether there is a target magnetic anomaly in the original signal according to the output result to realize magnetic anomaly perception.

[0013] According to yet another aspect of the present invention, a computer device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor. The processor executes the computer program to implement the steps of the magnetic anomaly perception method based on the improved RLMD and support vector machine as described above.

[0014] According to still another aspect of the present invention, a computer-readable storage medium is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, it implements the steps of the magnetic anomaly perception method based on the improved RLMD and support vector machine as described above.

[0015] Applying the technical solution of the present invention, a magnetic anomaly perception method based on improved RLMD and support vector machine is provided. In this method, RLMD can decompose non-linear and non-stationary signals into a series of product functions (PFs), and each PF contains physical information with a specific frequency. Reasonable reconstruction of the PFs processed by RLMD can effectively improve the signal-to-noise ratio of the data, which is beneficial to subsequent feature extraction. Magnetic anomaly perception can be regarded as a binary classification problem, and the support vector machine is a binary classification method with good performance and small sample requirements. The idea of the support vector machine is based on the principle of structural minimization to find an optimal classification hyperplane that can accurately divide the data. Extract features from the signal after PF reconstruction and input them into the support vector machine to output a binary judgment on whether the original signal has a target, so as to realize the perception of the target magnetic anomaly signal. In addition, when screening PF components in the present invention, the permutation entropy is used as an index. The core idea is that the target magnetic anomaly signal and background noise have different physical characteristics, and the permutation entropy reflects this feature. By calculating the permutation entropy of all PF components and sorting them from large to small, according to the three-quantiles, a group of PF components with the smallest permutation entropy is determined for reconstruction. This method has obvious self-adaptability, fully considering the differences between the target magnetic anomaly signal and background noise at the physical level, without the need for artificial setting of screening thresholds, achieving the purpose of fully suppressing background noise. Therefore, compared with the prior art, the magnetic anomaly perception method based on improved RLMD and support vector machine provided by the present invention reduces the manual intervention in feature extraction, can improve the magnetic anomaly perception ability in complex environments, fully considers the differences between the target magnetic anomaly signal and background noise at the physical level, can effectively improve the accuracy of magnetic anomaly detection in the prior art, and effectively solves the technical problem in the prior art that when performing signal recognition, the screening threshold is usually set artificially, and it is difficult to fully suppress noise, resulting in low target recognition accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings included are used to provide a further understanding of the embodiments of the present invention, which form a part of the specification, are used to illustrate the embodiments of the present invention, and are used to explain the principles of the present invention together with the text description. Obviously, the accompanying drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0017] Figure 1 The flowchart of the magnetic anomaly perception method based on improved RLMD and support vector machine provided according to a specific embodiment of the present invention is shown. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments may be combined with each other. The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way restricts the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0019] It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0020] Unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions, and numerical values set forth in these embodiments do not limit the scope of the present invention. At the same time, it should be understood that, for the sake of convenience of description, the dimensions of the various parts shown in the drawings are not drawn in actual proportional relationships. Technologies, methods, and devices known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the technologies, methods, and devices should be regarded as part of the authorization specification. In all the examples shown and discussed here, any specific value should be construed as merely exemplary and not as a limitation. Therefore, other examples of the exemplary embodiments may have different values. It should be noted that like reference numerals and letters indicate similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further discussed in subsequent drawings.

[0021] As Figure 1As shown, a magnetic anomaly perception method based on improved RLMD and support vector machine is provided according to a specific embodiment of the present invention. The magnetic anomaly perception method based on improved RLMD and support vector machine includes: Step 1, using an airborne magnetic survey platform to fly along a preset route to obtain original data s(t) that may contain target static magnetic anomaly signals, and decomposing the original data s(t) using the RLMD algorithm to obtain multiple PF components; Step 2, calculating the permutation entropy for each of the obtained multiple PF components, arranging all the permutation entropy values of the PF components from large to small, screening out the last n PF components, where n is adaptively determined by the three - quartiles of all the permutation entropy values, linearly reconstructing the PF components according to the last n PF components to obtain a linearly reconstructed signal, and removing the trend term from the linearly reconstructed signal to obtain a processed static magnetic signal m(t); Step 3, calculating the morphological fractal dimension D, kurtosis value K, and peak - to - peak value P of the processed static magnetic signal m(t), and constructing a corresponding feature vector group [D, K, P]; Step 4, establishing a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform, setting a label S for the signal, setting the label S of the signal with target magnetic anomaly to 1, and setting the label S of the signal without target magnetic anomaly to 0, training the support vector machine based on the training sample data set to obtain an optimal magnetic anomaly perception strategy; Step 5, inputting the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judging whether there is a target magnetic anomaly in the original signal according to the output result to achieve magnetic anomaly perception.

[0022] Using this configuration method, a magnetic anomaly perception method based on improved RLMD and support vector machine is provided. In this method, RLMD can decompose non-linear and non-stationary signals into a series of product functions (PFs), and each PF contains physical information of a specific frequency. Reasonable reconstruction of the PFs processed by RLMD can effectively improve the signal-to-noise ratio of data, which is beneficial to subsequent feature extraction. Magnetic anomaly perception can be regarded as a binary classification problem, and the support vector machine is a binary classification method with good performance and small demand for samples. The idea of the support vector machine is based on the principle of structural minimization to find an optimal classification hyperplane that can accurately divide the data. Extract features from the signal after PF reconstruction and input them into the support vector machine, and output a binary judgment on whether the original signal has a target, so as to realize the perception of the target magnetic anomaly signal. In addition, when screening the PF components in the present invention, the permutation entropy is used as an index. The core idea is that the target magnetic anomaly signal and background noise have different physical characteristics, and the permutation entropy reflects this characteristic. By calculating the permutation entropy of all PF components and sorting them from large to small, according to the three-quartile, a group of PF components with the smallest permutation entropy is determined for reconstruction. This method has obvious self-adaptability, fully considers the differences between the target magnetic anomaly signal and background noise at the physical level, does not require artificial setting of screening thresholds, and achieves the purpose of fully suppressing background noise. Therefore, compared with the prior art, the magnetic anomaly perception method based on improved RLMD and support vector machine provided by the present invention reduces the manual intervention in feature extraction, can improve the magnetic anomaly perception ability in complex environments, fully considers the differences between the target magnetic anomaly signal and background noise at the physical level, can effectively improve the accuracy of magnetic anomaly detection in the prior art, and effectively solves the technical problem in the prior art that when performing signal recognition, usually by artificially setting screening thresholds, this method is difficult to fully suppress noise, resulting in low target recognition accuracy.

[0023] To realize magnetic anomaly perception based on improved RLMD and support vector machine, first, it is necessary to use an airborne magnetic survey platform to fly along a preset route to obtain the original data s(t) that may contain the target static magnetic anomaly signal, and use the RLMD algorithm to decompose the original data s(t) to obtain multiple PF components.

[0024] As a specific embodiment of the present invention, in step one, the original data s(t) is a level flight data containing about 3,000 sampling points. Using the RLMD algorithm to decompose the original data s(t) to obtain multiple PF components specifically includes:

[0025] Step 1.1, determine all local extreme points n in the original data s(t) k (k = 1, 2,...), including the maximum and minimum values;

[0026] Step 1.2, calculate and obtain the first local mean function m 11 (t) and the first local envelope estimation value function a 11 (t). The local mean m 1i (t) in any local mean function can be obtained according to calculation. After obtaining multiple local means, the multiple local means are jointly fitted to form the first local mean function; the local envelope estimation value a 1i (t) in any envelope estimation function can be obtained according to calculation, where n k is the k-th extreme point, and n k+1 is the (k + 1)-th extreme point.

[0027] Step 1.3, subtract the first local mean function m 11 (t) from the original data s(t) to obtain the first intermediate function h 11 (t);

[0028] Step 1.4, divide the first intermediate function h 11 (t) by the first local envelope estimation value function a 11 (t) to obtain the demodulated first demodulated function s 11 (t);

[0029] Step 1.5, take the first intermediate function h 11 (t) as the original data, execute the steps from the extreme points to subtracting the local mean, repeat Steps 1.1 to 1.4, and obtain the second local mean function m 12 (t), the second local envelope estimation value function a 12 (t), the second intermediate function h 12 (t) and the second demodulated function s 12 (t). Specifically, take the first intermediate function h 11 (t) as the original data, determine all local extreme points in the original data, calculate and obtain the second local mean function and the second local envelope estimation value function according to multiple local extreme points, subtract the second local mean function from the original data (i.e., the first intermediate function h 11 (t)) to obtain the second intermediate function h 12 (t), and divide the second intermediate function h 12 (t) by the second local envelope estimation value function a 12 (t) to obtain the demodulated second demodulated function s 12 (t);

[0030] Step 1.6, take the second intermediate function h 12(t) As the original data, perform the steps from the extreme points to subtracting the local mean, repeat Steps 1.1 to 1.4 to obtain the third local mean function m 13 (t), the third local envelope estimation value function a 13 (t), the third intermediate function h 13 (t), and the third demodulated function s 13 (t). Specifically, take the second intermediate function h 12 (t) as the original data, determine all local extreme points in the original data, calculate and obtain the third local mean function and the third local envelope estimation value function according to multiple local extreme points, subtract the third local mean function from the original data (i.e., the second intermediate function h 12 (t)) to obtain the third intermediate function h 13 (t), divide the third intermediate function h 13 (t) by the third local envelope estimation value function a 13 (t) to obtain the demodulated third demodulated function s 13 (t);

[0031] Step 1.7, and so on, repeat the steps from the extreme points to subtracting the local mean until a pure frequency modulation signal s 1n (t) appears. Calculate and obtain the first PF component according to multiple local envelope estimation value functions and the pure frequency modulation signal s 1n (t). In the present invention, s 1n (t) = h 1n (t) / a 1n (t), n = 1, 2, 3, …, the envelope function of the first PF component Multiply the pure frequency modulation signal by the envelope function to obtain the first PF component PF1(t) = s 1n (t)a1(t);

[0032] Step 1.8, take u1(t) = s(t) - PF1(t) as the original signal for the next decomposition, repeat Steps 1.1 to 1.7 to obtain the second PF component;

[0033] Step 1.9, and so on, repeat Steps 1.1 to 1.8, sequentially obtain the third PF component, …, the Nth PF component, calculate and obtain u2(t), …, u k (t), until u k (t) stops when there is only one extreme point left or it is a monotonic function. Among them, u2(t), …, u k (t) can be calculated according to Calculate and obtain.

[0034] Specifically, take u2(t) = u1(t) - PF2(t) as the original signal for the next decomposition, repeat steps 1.1 to 1.7 to obtain the third PF component; take u3(t) = u2(t) - PF3(t) as the original signal for the next decomposition, repeat steps 1.1 to 1.7 to obtain the fourth PF component; take u4(t) = u3(t) - PF4(t) as the original signal for the next decomposition, repeat steps 1.1 to 1.7 to obtain the fourth PF component; and so on until u k (t) stops when there is only one extreme point left or it becomes a monotonic function. Through the above loop, the signal

[0035] Furthermore, after obtaining multiple PF components, the permutation entropy can be calculated for each of the obtained PF components respectively. Arrange the permutation entropy values of all PF components from largest to smallest, and select the last n PF components, where n is adaptively determined by the three - quartiles of all permutation entropy values. Perform linear reconstruction on the PF components according to the last n PF components to obtain a linearly reconstructed signal, and remove the trend term from the linearly reconstructed signal to obtain the processed magnetostatic signal m(t).

[0036] The permutation entropy PE of any PF component can be obtained according to where p j is the probability of the occurrence of permutation j in the PF component, k = 1, 2, …, N, permutation j is the j - th way of the permutation mode of the PF component, and it is a well - known parameter.

[0037] Since the larger the PE value, the higher the proportion of noise in the modal component and the more random it is. To fully suppress the background noise, arrange the PE values of all modal components from largest to smallest, and select the last n components, where n is adaptively determined by the three - quartiles of all PE values, to obtain the linearly reconstructed signal Fit to obtain the linear trend term L(t) of the linearly reconstructed signal lm(t).

[0038] The purpose of detrending is to remove the linear trend term of the reconstructed data. Its core is to fit the least - squares line segment of the original data and then subtract it from the original data, so that the mean value of the processed data approaches 0. The calculation process can be described as: estimate the parameters k, b of the linear function and make them satisfy Then k, b are calculated as: where x i ∈(t1, t2, …, t n ), y i ∈(lm(t1), lm(t2), … lm(t n), pinv is a subfunction for finding the generalized inverse of a matrix, which is implemented by singular value decomposition of the matrix. For example, if the inverse of matrix X is required, first perform singular value decomposition on it: [U, S, V] = SVD(X), where X = U * S * V T , and U * U T = I, V * U T = I, then the generalized inverse X -1 of matrix X is: X -1 = pinv(X) = V * S -1 * U T . Therefore, L(t) = k * lm(t) + b.

[0039] Subtract the linear trend term L(t) from the linearly reconstructed signal lm(t) to obtain the processed magnetostatic signal m(t), that is, m(t) = lm(t) - L(t), where lm(t) is the linearly reconstructed signal and L(t) is the linear trend term.

[0040] Further, after obtaining the processed magnetostatic signal m(t), construct a feature vector group for the processed magnetostatic signal m(t). Specifically, calculate the morphological fractal dimension D, kurtosis value K, and peak-to-peak value P of the processed magnetostatic signal m(t), and construct the corresponding feature vector group [D, K, P].

[0041] In particular, the method for constructing a feature vector group for the processed magnetostatic signal m(t) is as follows:

[0042] Calculate the morphological fractal dimension D of m(t), and the morphological fractal dimension D in the feature vector group [D, K, P] = log2N m (δ m ) / (-log2δ m ), where N m (δ m ) represents the number of non-empty boxes corresponding to it, and the side length of the box is δ m (m = 1, 2,..., M);

[0043] Calculate the kurtosis value K of m(t), and the kurtosis value in the feature vector group [D, K, P] where x is the amplitude of the processed magnetostatic signal m(t), μ is the average value of the processed magnetostatic signal m(t), and σ is the standard deviation of the processed magnetostatic signal m(t);

[0044] Calculate the peak-to-peak value P of m(t), and the peak-to-peak value P in the feature vector group [D, K, P] = max(x) - min(x), where max(x) is the maximum value of the processed magnetostatic signal m(t) and min(x) is the minimum value of the processed magnetostatic signal m(t).

[0045] The corresponding eigenvector group of m(t) is [D, K, P].

[0046] Furthermore, after constructing the corresponding eigenvector group [D, K, P], a training sample data set can be established, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform. Set the label S for the signal. The signal label S with target magnetic anomaly is set to 1, and the signal label S without target magnetic anomaly is set to 0. Train the support vector machine based on the training sample data set to obtain the optimal magnetic anomaly perception strategy. In particular, the sample data set is generated by the Monte Carlo method, basically covering various situations where the target to be detected may appear.

[0047] After establishing the training sample data set, the eigenvector group corresponding to the processed static magnetic signal m(t) can be input into the trained support vector machine, and it can be judged whether there is a target magnetic anomaly in the original signal according to the output result, so as to realize magnetic anomaly perception.

[0048] According to another aspect of the present invention, a magnetic anomaly perception system based on VMD-1DCNN is provided. The magnetic anomaly perception system based on VMD-1DCNN uses the magnetic anomaly perception method based on improved RLMD and support vector machine as described above for magnetic anomaly perception. The magnetic anomaly perception system includes a PF component calculation module, a static magnetic signal acquisition module, a feature vector group construction module, a training sample data set construction module, and a magnetic anomaly perception determination module. The PF component calculation module is used to fly along a preset route using an airborne magnetic survey platform to obtain the original data s(t) that may contain the target static magnetic anomaly signal, and decompose the original data s(t) using the RLMD algorithm to obtain multiple PF components. The static magnetic signal acquisition module is used to calculate the permutation entropy for each of the obtained multiple PF components, arrange all the permutation entropy values of the PF components from large to small, and select the latter n PF components, where n is adaptively determined by the three-quartiles of all the permutation entropy values. The PF components are linearly reconstructed according to the latter n PF components to obtain a linearly reconstructed signal, and the trend term is removed from the linearly reconstructed signal to obtain the processed static magnetic signal m(t). The feature vector group construction module is used to calculate the morphological fractal dimension D, kurtosis value K, and peak-to-peak value P of the processed static magnetic signal m(t), and construct the corresponding feature vector group [D, K, P]. The training sample data set construction module is used to establish a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform, set a label S for the signal, set the label S of the signal with target magnetic anomaly to 1, and set the label S of the signal without target magnetic anomaly to 0. The support vector machine is trained based on the training sample data set to obtain the optimal magnetic anomaly perception strategy. The magnetic anomaly perception determination module is used to input the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judge whether there is a target magnetic anomaly in the original signal according to the output result, so as to realize magnetic anomaly perception.

[0049] By applying this configuration method, a magnetic anomaly perception system based on improved RLMD and support vector machine is provided. In this system, RLMD can decompose non-linear and non-stationary signals into a series of product functions (PFs), and each PF contains physical information of a specific frequency. Reasonable reconstruction of the PFs processed by RLMD can effectively improve the signal-to-noise ratio of the data, which is beneficial to subsequent feature extraction. Magnetic anomaly perception can be regarded as a binary classification problem, and the support vector machine is a binary classification method with good performance and small sample requirements. The idea of the support vector machine is based on the principle of structural minimization to find an optimal classification hyperplane that can accurately divide the data. The features of the signal after PF reconstruction are extracted and input into the support vector machine, and a binary judgment of whether the original signal has a target is output, so as to realize the perception of the target magnetic anomaly signal. In addition, when screening the PF components in the present invention, the permutation entropy is used as an index, and its core idea is that the target magnetic anomaly signal and background noise have different physical characteristics, and the permutation entropy reflects this feature. By calculating the permutation entropy of all PF components and sorting them from large to small, according to the three quantiles, a group of PF components with the smallest permutation entropy is determined for reconstruction. This system has obvious self-adaptability, fully considers the physical differences between the target magnetic anomaly signal and background noise, and does not require artificial setting of screening thresholds, achieving the purpose of fully suppressing background noise.

[0050] According to another aspect of the present invention, there is provided a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the magnetic anomaly perception method based on improved RLMD and support vector machine as described above.

[0051] According to still another aspect of the present invention, there is provided a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the magnetic anomaly perception method based on improved RLMD and support vector machine as described above.

[0052] For a further understanding of the present invention, the following combines Figure 1 to elaborate in detail on the magnetic anomaly perception method based on improved RLMD and support vector machine provided by the present invention.

[0053] As Figure 1 shown, around the people's livelihood and military needs such as mineral resource exploration, sunken ship archaeology, and underwater target detection, the present invention proposes a magnetic anomaly perception method based on improved RLMD and support vector machine, which can effectively improve the accuracy of magnetic anomaly detection in the prior art to achieve effective detection of weak magnetic anomaly signals. Its complete process is as Figure 1 shown, including the following steps:

[0054] Step S01: Conduct a flight test, collect the raw data that may contain the target magnetic anomaly, and intercept the level flight segment for subsequent processing;

[0055] Step S02: Decompose the raw data s(t) using RLMD to obtain several PF components. Specifically, for any raw data s(t), the method of decomposing it using RLMD is as follows:

[0056] Use spline interpolation to process all the maximum and minimum values of s(t) respectively, construct the upper and lower envelope functions, and obtain the local mean function m 11 (t) and the local envelope function a 11 (t);

[0057] Separate m 11 (t) from s(t) to obtain h 11 (t), divide h 11 (t) by the envelope function a 11 (t) to obtain the demodulated s 11 (t), and repeat the above steps until a pure frequency modulation signal s 1n (t) appears, where s 1n (t) = h 1n (t) / a 1n (t), n = 1, 2, 3, ….

[0058] The envelope signal of the first PF component

[0059] Multiply the pure frequency modulation signal by the envelope function to obtain the first PF component PF1(t) = s 1n (t)a1(t);

[0060] Take u1(t) = s(t) - PF1(t) as the initial signal for the next cycle until u k (t) has only one extreme point or is a monotonic function and then stop.

[0061]

[0062] Through the above cycle, the raw data s(t) is decomposed into k PF components and a residual function, which can be expressed as: Performing RLMD decomposition on the raw data can not only suppress the mode mixing of other adaptive decomposition algorithms, but also has high computational efficiency, obtaining physical information with more obvious local characteristics.

[0063] Step S03: Calculate the permutation entropy of the PF components, adaptively screen the PF components according to the three - quartiles of the permutation entropy, reconstruct the screened components and remove the linear trend to obtain the reconstructed static magnetic data signal m(t). Specifically, the method for screening the PF components is as follows:

[0064] Calculate the permutation entropy PE of each PF component:

[0065]

[0066] where p j represents the probability that permutation j appears in the PF component.

[0067] Arrange the PE values of all PF components from largest to smallest, and screen out the last n PF components, where n is adaptively determined by the three - quartiles of the PE value, to obtain the linear reconstruction signal

[0068] Fit and subtract the linear trend term of lm(t) to obtain the processed static magnetic data signal m(t).

[0069] When screening the PF components, the index of permutation entropy is adopted. Its core idea is that the target magnetic anomaly signal and background noise have different physical characteristics, and the permutation entropy reflects this feature. By calculating the permutation entropy of all PF components and sorting them from largest to smallest, according to the three - quartiles, a group of PF components with the smallest permutation entropy is determined for signal reconstruction. This method has obvious self - adaptability, fully considering the differences between the target magnetic anomaly signal and background noise at the physical level, without the need for artificial setting of screening thresholds, achieving the purpose of suppressing background noise.

[0070] Step S04: Construct the feature vector group [D, K, P] for the processed data m(t). Specifically, each element in the feature vector group is obtained in the following way:

[0071] The morphological fractal dimension D of m(t)=log2N m (δ m ) / (-log2δ m ), where N m (δ m ) represents the number of non - empty boxes corresponding to it, and the side length of the box is δ m (m = 1,2,…,M);

[0072] The kurtosis value of m(t) where x is the amplitude of the processed static magnetic signal m(t), μ is the average value of the processed static magnetic signal m(t), and σ is the standard deviation of the processed static magnetic signal m(t);

[0073] The peak-to-peak value P of m(t) is P = max(x) - min(x), where max(x) is the maximum value of the processed static magnetic signal m(t), and min(x) is the minimum value of the processed static magnetic signal m(t).

[0074] In step S05, a training sample data set is established, including various situations such as the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the platform. Labels S are set for the signals. The signal label S for the presence of target magnetic anomalies is set to 1, and the S for the absence of target magnetic anomaly signals is set to 0. The sample data set is generated by the Monte Carlo method, basically covering all kinds of situations where the target to be detected may appear. The support vector machine is trained based on the training sample data set to obtain the optimal magnetic anomaly perception strategy.

[0075] In step S06, the feature vector group [D, K, P] corresponding to m(t) is input into the support vector machine, and a judgment on whether the target magnetic anomaly signal is contained in m(t) is output, realizing the perception and classification of magnetic anomalies. The support vector machine has obvious advantages in solving small sample, non-linear and high-dimensional pattern recognition problems. Its mathematical model is simple and clear, and it is widely used in scenarios such as pattern recognition. The algorithm of the support vector machine can be reduced to solving a constrained QP problem. Its main idea is to map the input vector to a high-dimensional feature space through a non-linear mapping selected by a certain implementation, and construct an optimal hyperplane in this space. In this example, the input vector is the feature vector group [D, K, P] corresponding to m(t), and the output includes two situations, that is, m(t) contains the target magnetic anomaly signal and m(t) does not contain the target magnetic anomaly signal. Thus, the target magnetic anomaly perception is completed.

[0076] In summary, the present invention provides a magnetic anomaly perception method based on improved RLMD and support vector machine. This method reduces the manual intervention in feature extraction, can improve the magnetic anomaly perception ability in complex environments, fully considers the physical differences between target magnetic anomaly signals and background noise, can effectively improve the accuracy of magnetic anomaly detection in the prior art, and effectively solves the technical problem in the prior art that when performing signal recognition, usually by artificially setting a screening threshold, this method is difficult to fully suppress noise, resulting in low target recognition accuracy.

[0077] For ease of description, spatial relative terms such as "above", "over", "on the upper surface", "upper" and the like may be used herein to describe the spatial positional relationship of one device or feature to other devices or features as shown in the figures. It should be understood that the spatial relative terms are intended to encompass different orientations in use or operation in addition to the orientation depicted in the figures. For example, if the device in the figure is inverted, the device described as "above" or "over" other devices or structures will then be positioned "below" or "beneath" the other devices or structures. Thus, the exemplary term "above" can include both the orientations of "above" and "below". The device may also be positioned in other different ways (rotated 90 degrees or in other orientations), and the corresponding explanations for the spatial relative descriptions used herein will be made accordingly.

[0078] In addition, it should be noted that the use of terms such as "first" and "second" to define components is only for the convenience of distinguishing the corresponding components. Without additional statements, these terms have no special meanings, and thus should not be construed as limiting the scope of protection of the present invention.

[0079] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principle of the present invention shall be included within the scope of protection of the present invention.

Claims

1. An improved RLMD and support vector machine-based magnetic anomaly perception method, characterized in that The magnetic anomaly perception method based on improved RLMD and support vector machine includes: Step 1: Use an airborne magnetic survey platform to fly along a preset route to obtain the original data s(t) that may contain the target static magnetic anomaly signal. Decompose the original data s(t) using the RLMD algorithm to obtain multiple PF components; Step 2: Calculate the permutation entropy for each of the obtained multiple PF components respectively. Arrange all the permutation entropy values of the PF components from largest to smallest, and select the last n PF components, where n is adaptively determined by the three - quartiles of all the permutation entropy values. Perform linear reconstruction on the PF components according to the last n PF components to obtain a linearly reconstructed signal, and remove the trend term from the linearly reconstructed signal to obtain the processed static magnetic signal m(t); Step 3: Calculate the morphological fractal dimension D, kurtosis value K, and peak - to - peak value P of the processed static magnetic signal m(t), and construct the corresponding feature vector group [D, K, P]; Step 4: Establish a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of the detection platform. Set a label S for the signal. The label S of the signal with target magnetic anomaly is set to 1, and the label S of the signal without target magnetic anomaly is set to 0. Train the support vector machine based on the training sample data set to obtain the optimal magnetic anomaly perception strategy; Step 5: Input the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judge whether there is a target magnetic anomaly in the original signal according to the output result to achieve magnetic anomaly perception.

2. The magnetic anomaly perception method based on improved RLMD and support vector machine according to claim 1, characterized in that In the said Step 1, decomposing the original data s(t) using the RLMD algorithm to obtain multiple PF components specifically includes: Step 1.1: Determine all local extreme points in the original data s(t); Step 1.2, calculate and obtain the first local mean function m 11 (t) and the first local envelope estimate function a 11 (t); Step 1.3, subtract the first local mean function m 11 (t) from the original data s(t) to obtain a first intermediate function h 11 (t); Step 1.4, divide the first intermediate function h 11 (t) by the first local envelope estimation function a 11 (t) to obtain the demodulated first demodulated function s 11 (t); Step 1.5, take the first intermediate function h 11 (t) as the original data, perform the steps from the extreme point to subtracting the local mean, repeat the steps from Step 1.1 to Step 1.4 to obtain the second local mean function m 12 (t), the second local envelope estimation value function a 12 (t), the second intermediate function h 12 (t), and the second demodulated function s 12 (t); Step 1.6, take the second intermediate function h 12 (t) as the original data, perform the steps from the extreme point to subtracting the local mean, repeat the steps from Step 1.1 to Step 1.4, and obtain the third local mean function m 13 (t), the third local envelope estimation value function a 13 (t), the third intermediate function h 13 (t), and the third demodulated function s 13 (t); Step 1.7, and so on. Repeat the steps from the extreme point to subtracting the local mean until a pure frequency modulation signal s 1n (t) is obtained. Calculate the first PF component PF1(t) according to multiple local envelope estimation value functions and the pure frequency modulation signal s 1n (t); Step 1.8: Take u1(t) = s(t) - PF1(t) as the original signal for the next decomposition, and repeat the said Step 1.1 to the said Step 1.7 to obtain the second PF component; Step 1.9, repeat the said Step 1.1 to the said Step 1.8, sequentially obtain the third PF component, …, the Nth PF component, calculate and obtain u2(t), …, u k (t), until u k (t) has only one extreme point or becomes a monotonic function and then stop. u2(t), …, u k (t) is obtained according to calculation.

3. The magnetic anomaly perception method based on improved RLMD and support vector machine according to claim 2, characterized in that, The local mean m in any of the local mean functions 1i (t) can be obtained according to The local envelope estimate function a of any of the local envelope estimate value functions 1i (t) can be obtained according to Calculation, where n k Is the k-th extreme point, n k+1 Is the (k + 1)-th extreme point.

4. The magnetic anomaly perception method based on improved RLMD and support vector machine according to claim 3, wherein The permutation entropy PE of any of the PF components can be obtained according to calculation, where p j is the probability that the permutation j appears in the PF component, k = 1, 2,..., N, and the permutation j is the j-th way of the PF component permutation method.

5. The magnetic anomaly perception method based on improved RLMD and support vector machine according to claim 4, wherein The processed magnetostatic signal m(t) can be obtained by calculating m(t) = lm(t) - L(t), where lm(t) is the linearly reconstructed signal and L(t) is the linear trend term.

6. The magnetic anomaly perception method based on improved RLMD and support vector machine according to claim 1, characterized in that The morphological fractal dimension D in the feature vector group [D, K, P] = log2N m (δ m ) / (-log2δ m ), where N m (δ m ) represents the number of non-empty boxes corresponding to δ m , and the side length of the box is δ m , m = 1, 2,..., M; The kurtosis value in the feature vector group [D, K, P] where x is the amplitude of the processed static magnetic signal m(t), μ is the average value of the processed static magnetic signal m(t), and σ is the standard deviation of the processed static magnetic signal m(t); the peak-to-peak value P in the feature vector group [D, K, P] = max(x) - min(x), where max(x) is the maximum value of the processed static magnetic signal m(t) and min(x) is the minimum value of the processed static magnetic signal m(t).

7. A magnetic anomaly perception system based on improved RLMD and support vector machine, characterized in that, The magnetic anomaly perception system based on improved RLMD and support vector machine uses the magnetic anomaly perception method based on improved RLMD and support vector machine as described in any one of claims 1 to 6 for magnetic anomaly perception. The magnetic anomaly perception system includes: A PF component calculation module, which is used to use an airborne magnetic survey platform to fly along a preset route to obtain the original data s(t) that may contain the target static magnetic anomaly signal, and decompose the original data s(t) using the RLMD algorithm to obtain multiple PF components; A static magnetic signal acquisition module, which is used to calculate the permutation entropy for each of the obtained multiple PF components respectively, arrange all the permutation entropy values of the PF components from largest to smallest, select the last n PF components, where n is adaptively determined by the three - quartiles of all the permutation entropy values, perform linear reconstruction on the PF components according to the last n PF components to obtain a linearly reconstructed signal, and remove the trend term from the linearly reconstructed signal to obtain the processed static magnetic signal m(t); Feature vector group construction module, which is used to calculate the morphological fractal dimension D, kurtosis value K and peak-to-peak value P of the processed static magnetic signal m(t), and construct the corresponding feature vector group [D, K, P]; Training sample data set construction module, which is used to establish a training sample data set, including various situations of the presence or absence of ferromagnetic targets, different distances of ferromagnetic targets, and different speeds of detection platforms, set a label S for the signal, set the signal label S with target magnetic anomaly to 1, and set the label S of the signal without target magnetic anomaly to 0, and train the support vector machine based on the training sample data set to obtain the optimal magnetic anomaly perception strategy; Magnetic anomaly perception determination module, which is used to input the feature vector group corresponding to the processed static magnetic signal m(t) into the trained support vector machine, and judge whether there is a target magnetic anomaly in the original signal according to the output result to realize magnetic anomaly perception.

8. A computer device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the steps of the magnetic anomaly perception method based on the improved RLMD and support vector machine according to any one of claims 1 to 6.

9. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it realizes the steps of the magnetic anomaly perception method based on the improved RLMD and support vector machine according to any one of claims 1 to 6.

Images