Method for removing background of magnetic background

Abstract

The invention discloses a background removing method for a magnetic background. The method comprises the following steps of: firstly, acquiring magnetic background data of a detected area by using an unmanned aeromagnetic detection system before a target shooting task is started, and storing the magnetic background data into a database after aeromagnetic compensation and magnetic daily variation correction; after the target shooting task is completed, the magnetic detection data of the detected area are collected again, aeromagnetic compensation, magnetic daily variation correction, magnetic background estimation data calculation of each sampling point, magnetic background data removal of each sampling point and other processing are carried out, magnetic background background removal is finally achieved, and the magnetic detection data with the magnetic background removed are obtained. According to the method, the interference of the magnetic background signal can be effectively filtered out in a complex magnetic environment, and the magnetic abnormal signal of the detected target is highlighted, so that the detection accuracy of the unexploded ordnance in the target range is improved, false detection and missed detection are avoided, and safe and accurate unexploded ordnance detection is realized.

Description

Technical Field

[0001] The invention belongs to the technical field of unmanned detection of unexploded bombs, and in particular relates to a method for removing magnetic background. Background Art

[0002] At present, the use of drones equipped with magnetic detection payloads to perform unexploded bomb detection tasks is an effective means of detecting ferromagnetic unexploded bombs. It has gradually replaced manual detection of unexploded bombs and plays an important role in the detection of unexploded bombs after firing at the shooting range.

[0003] Due to years of target practice, the shooting range has a large number of shrapnel, large geomagnetic interference, and an extremely complex magnetic environment. Unexploded bombs are randomly scattered in the shooting range area. The interference caused by the complex geomagnetic environment of the shooting range can easily drown out the weak magnetic anomaly signals of the unexploded bombs, making it difficult to accurately locate underground unexploded bombs, and greatly increasing the false alarm rate of detection and identification. In addition, the targets of the shooting range may be facilities such as pavements and foundations with buried ferromagnetic steel bars. These facilities will generate magnetic interference, affecting the extraction of weak magnetic anomaly signals of unexploded bombs, resulting in the leakage of unexploded bombs.

[0004] In the field of image processing technology, the so-called background subtraction refers to taking the first frame of the image captured by the camera as the background, and subtracting the background frame from each subsequent frame. After subtraction, what remains is the extra part of the feature object (the object to be detected). However, the subtracted part will also affect the grayscale value of the feature object, and generally, it is necessary to set a relevant threshold for judgment. This method processes different image frames taken by the same camera. The time difference between the image frames is small, the coordinate system is the same, and the lighting environment is similar. It is essentially different from the magnetic background removal process and cannot be applied to this technical field.

[0005] In terms of traditional aeromagnetic background field processing, different experts and scholars have adopted a variety of strategies to remove the influence of the background field. Many experts and scholars have proposed different filtering methods, such as using Butterworth filters and spectral features to remove the influence of the background field, or using wavelet transform methods to deal with background field problems. Although these methods have their own characteristics, in practical applications, the background field is often treated as an unknown item. In the sub-field of unexploded bomb detection, since the magnetic anomaly signal is very weak, the direct use of filtering methods can easily weaken the effective magnetic anomaly signal to an unrecognizable level and cause missed detection. Therefore, the technical route of using filtering methods to eliminate magnetic background field interference is not applicable to the field of unexploded bomb detection technology. Summary of the invention

[0006] The technical problem solved by the present invention is to overcome the deficiencies of the prior art and provide a magnetic background elimination method, which can effectively filter out the interference of complex magnetic environments, improve the accuracy of unexploded bomb detection in complex magnetic environments, avoid false detection and missed detection, and realize safe and accurate unexploded bomb detection.

[0007] The present invention includes the following technical solutions: A method for removing magnetic background, comprising:

[0008] Before the target shooting mission begins, the original magnetic background data of the measured area is collected, and after the aeromagnetic compensation and magnetic diurnal variation correction are completed, it is stored in the magnetic background database;

[0009] After the target shooting task is completed, the original magnetic detection data of the measured area is collected. According to the data in the magnetic background database before the start of the target shooting task, the magnetic background of the original magnetic detection data of the measured area collected after the target shooting task is eliminated to obtain the magnetic detection data after the magnetic background is eliminated.

[0010] Before the target shooting task begins, the original magnetic background data of the measured area is collected, and after the magnetic compensation processing and the magnetic diurnal variation correction processing are completed, the data is stored in the magnetic background database, including:

[0011] Before the target shooting mission begins, the UAV is equipped with a magnetic detection payload to collect the original magnetic background data of the measured area and obtain the original magnetic background data.

[0012] Carry out aeromagnetic compensation processing on the collected raw magnetic background data;

[0013] The original magnetic background data after aeromagnetic compensation is corrected for magnetic diurnal variation.

[0014] The magnetic background data after aeromagnetic compensation and magnetic diurnal variation correction are stored in the magnetic background database.

[0015] The UAV adopts a multi-rotor UAV, and the magnetic detection payload includes a fluxgate probe capable of collecting three-component magnetic field data and at least one scalar probe capable of collecting total magnetic field data, so that the magnetic detection payload can synchronously collect a set of three-component magnetic field data and at least one set of total magnetic field data, and synchronize with the RTK longitude and latitude coordinates, plane coordinates, flight altitude, angle between the UAV nose direction and true north, and time reference information of the UAV to form a data frame, that is, magnetic background raw data.

[0016] The raw magnetic background data of the measured area is obtained by autonomous detection by an unmanned aerial vehicle equipped with a magnetic detection payload according to the planned trajectory; the specific trajectory requirements of the unmanned aerial vehicle are: a "bow"-shaped reciprocating straight trajectory within the measured area, and the trajectory covers the measured area; the flight altitude of the unmanned aerial vehicle is set to 1m~2m, the flight speed is 1m / s~3m / s, and the route spacing is 1m~2m; the unmanned aerial vehicle is equipped with a magnetic detection payload according to the above parameters to complete the collection of raw magnetic background data.

[0017] The aeromagnetic compensation processing includes: using an unmanned aerial vehicle equipped with a magnetic detection payload to fly to a distance of 40m to 100m from the ground, away from the interference of the complex magnetic field on the surface, hovering in the air to complete maneuvers of pitch ±10°, yaw ±360°, and roll ±10°, and collecting three-component magnetic field data and total magnetic field data; then obtaining the attitude information matrix of the TL model; using the aeromagnetic compensation algorithm to solve and obtain 18 compensation coefficients of the TL model; finally, substituting the 18 compensation coefficients and the original magnetic background data into the TL model to complete the aeromagnetic compensation processing of the original magnetic background data.

[0018] The magnetic background raw data after aeromagnetic compensation is subjected to magnetic diurnal variation correction processing, including:

[0019] Differentiation and processing are performed according to the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload, as follows: in the case where the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload is one, a scalar probe capable of collecting total magnetic field data is synchronously deployed on the ground, and the same time and space reference is adopted as the scalar probe carried by the UAV, and data are collected simultaneously, and the total magnetic field data collected by the scalar probe carried by the UAV is subtracted from the total magnetic field data collected by the scalar probe on the ground, so as to realize correction of diurnal magnetic variation; in the case where the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload is greater than one, correction of diurnal magnetic variation is realized by performing horizontal or vertical gradient calculation on the total magnetic field data synchronously collected by multiple scalar probes.

[0020] The step of performing magnetic background removal on the collected data to obtain magnetic detection data after the magnetic background is removed includes:

[0021] After the target shooting mission is completed, the same acquisition method as before the target shooting mission is adopted to obtain the raw data of magnetic detection, and then the aeromagnetic compensation processing and magnetic diurnal variation correction processing are carried out on it in turn;

[0022] The original magnetic background data after magnetic compensation and magnetic diurnal variation processing before the target shooting mission is taken out from the magnetic background database;

[0023] The processed magnetic background raw data and magnetic detection raw data in the measured area are divided according to the route, and classified into magnetic background raw data and magnetic detection raw data of route 1 to route N;

[0024] According to the processed magnetic background raw data taken out from the magnetic background database, the magnetic background estimation data corresponding to the m spatially discrete sampling points of route n are calculated, where n=1, 2, …, N; the spatially discrete sampling points m of each route n may be the same or different; that is, at the spatially discrete sampling points [(x1′, y1′, z1′), (x′2, y′2, z′2), … (x′ m ,y′ m ,z′m )] corresponding to the estimated magnetic background data [M(x1′,y1′,z1′),M(x2′,y′2,z′2),…M(x′ m ,y′ m ,z′ m )];

[0025] According to the above method, the magnetic detection data of route 1 to route N are calculated as the corresponding magnetic background estimation data at m spatially discrete sampling points;

[0026] For [(x1′,y1′,z1′),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )], the processed raw magnetic detection data is subtracted from the corresponding magnetic background estimation data, that is, the magnetic detection data C(x′) of each sampling point after the magnetic background is removed is obtained. m ,y′ m ,z′ m ) = M′(x′ m ,y′ m ,z′ m )-M(x′ m ,y′ m ,z′ m ).

[0027] The processed magnetic background raw data and magnetic detection raw data in the measured area are divided into magnetic background data and magnetic detection raw data of route 1 to route N according to the route, including:

[0028] The known UAV route planning information and route error are used for classification. Specifically, the UAV route is set to fly along the due north direction, the route accuracy is <0.5m, the number of planned routes is n, n=1,2,…,N, and the horizontal coordinates of each waypoint on route n are x no ; Each spatially discrete sampling point in the magnetic background raw data and magnetic detection raw data has a unique horizontal coordinate x. When x is at (x no -0.5,x no +0.5), the sampling point is classified as a point on route n.

[0029] The calculation of magnetic background estimation data corresponding to m spatially discrete sampling points of route n includes:

[0030] Assume that the processed raw magnetic background data of route n is represented by spatially discrete sampling points [(x1, y1, z1), (x2, y2, z2), … (x n ,y n ,z n)] and its corresponding magnetic field values ​​[M(x1,y1,z1),M(x2,y2,z2),…M(x n ,y n ,z n )]; Assume that the processed raw magnetic detection data of route n is represented by spatially discrete sampling points [(x1′, y1′, z1′), (x′2, y′2, z′2), … (x′ m ,y′ m ,z′ m )] and its corresponding magnetic field values ​​[M′(x1′,y1′,z1′),M′(x′2,y′2,z′2),…M′(x′ m ,y′ m ,z′ m )]; using the spatial interpolation algorithm, we can obtain the function f(x′ m ,y′ m ,z′ m ), [(x1′,y1′,z1′), (x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )] Substitute into the function f(x′ m ,y′ m ,z′ m ), we get the discrete sampling points in space [(x1′,y1′,z1′),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )], that is, [M(x1′,y1′,z1′),M(x2′,y′2,z′2),…M(x′ m ,y′ m ,z′ m )].

[0031] The spatial interpolation algorithm adopts the Kriging interpolation method, that is, the best linear unbiased prediction value of the unknown point is predicted by calculating the weighted average of each known spatial sampling point.

[0032] Compared with the prior art, the present invention has the following advantages:

[0033] (1) This method can effectively filter out the interference of magnetic background signals and highlight magnetic anomaly signals in complex magnetic environments, thereby improving the accuracy of unexploded bomb detection in the target range, avoiding false detection and missed detection, and achieving safe and accurate unexploded bomb detection.

[0034] (2) When the drone flies along the trajectory, there are random errors in both the horizontal and height directions. This will result in the discrete point positions of the magnetic background data and magnetic detection data being different, although the trajectory, flight altitude, flight speed, and flight distance set for the two drone flights before and after the target shooting are the same. Figure 3 As shown in the figure, it is impossible to directly subtract the two data. This method solves this problem by using a method of spatial interpolation after data preprocessing, making it feasible to remove magnetic background in the field of unmanned detection of unexploded bombs.

[0035] (3) The original magnetic background data and the original magnetic detection data in the measured area are divided according to the route, and are classified into the magnetic background data and the original magnetic detection data of route 1 to route N. This method can effectively reduce the complexity of the calculation of the magnetic background estimation data of each sampling point and improve the overall calculation efficiency of the algorithm. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a flowchart of this method.

[0037] Figure 2 This is a schematic diagram of UAV flight trajectory planning.

[0038] Figure 3 It is a schematic diagram of the discrete point locations of magnetic background data and magnetic detection data.

[0039] Figure 4 The magnetic background data after magnetic compensation and magnetic diurnal variation correction is drawn into a plane magnetic map.

[0040] Figure 5 The magnetic detection data after magnetic compensation processing and magnetic diurnal variation correction processing are drawn into a plane magnetic map.

[0041] Figure 6 It is to draw a plane magnetic map based on the magnetic detection data after removing the magnetic background. DETAILED DESCRIPTION

[0042] In order to obtain a better recognition rate and positioning accuracy of unexploded bombs in a shooting range environment with a complex magnetic background, it is necessary to remove or weaken the influence of the complex magnetic background. This method firstly requires the use of an unmanned aeromagnetic detection system to collect magnetic background data of the measured area before the shooting task begins, and then store it in the database after aeromagnetic compensation and magnetic diurnal correction; after the shooting task is completed, the magnetic detection data of the measured area is collected again, and aeromagnetic compensation, magnetic diurnal correction, magnetic background estimation data calculation, and magnetic background data removal of each sampling point are performed, and finally the magnetic background is eliminated, and the magnetic detection data after the magnetic background is eliminated is obtained, and then subsequent processing is performed.

[0043] The common implementation modes of the present invention will be described in further detail below with reference to the accompanying drawings.

[0044] like Figure 1 As shown, the present invention provides a method for removing magnetic background, comprising:

[0045] Before the target shooting mission begins, the original magnetic background data of the measured area is collected, and after the aeromagnetic compensation and magnetic diurnal variation correction are completed, it is stored in the magnetic background database;

[0046] After the target shooting task is completed, the original magnetic detection data of the measured area is collected. According to the data in the magnetic background database before the start of the target shooting task, the magnetic background of the original magnetic detection data of the measured area collected after the target shooting task is eliminated to obtain the magnetic detection data after the magnetic background is eliminated.

[0047] Before the target shooting task begins, the original magnetic background data of the measured area is collected, and after the magnetic compensation processing and the magnetic diurnal variation correction processing are completed, the data is stored in the magnetic background database, including:

[0048] Before the target shooting mission begins, the UAV is equipped with a magnetic detection payload to collect the original magnetic background data of the measured area and obtain the original magnetic background data.

[0049] Carry out aeromagnetic compensation processing on the collected raw magnetic background data;

[0050] The original magnetic background data after aeromagnetic compensation is corrected for magnetic diurnal variation.

[0051] The magnetic background data after aeromagnetic compensation and magnetic diurnal correction are stored in the magnetic background database. At this time, the magnetic background data after magnetic compensation and magnetic diurnal correction are plotted into a plane magnetic map, such as Figure 4 shown.

[0052] The UAV adopts a multi-rotor UAV, and the magnetic detection payload includes a fluxgate probe capable of collecting three-component magnetic field data and at least one scalar probe capable of collecting total magnetic field data, so that the magnetic detection payload can synchronously collect a set of three-component magnetic field data and at least one set of total magnetic field data, and synchronize with the RTK longitude and latitude coordinates, plane coordinates, flight altitude, angle between the UAV nose direction and true north, and time reference information of the UAV to form a data frame, that is, magnetic background raw data.

[0053] The original magnetic background data of the measured area is obtained by autonomous detection by the UAV equipped with magnetic detection payload according to the planned trajectory; the specific trajectory requirements of the UAV are: a "bow"-shaped reciprocating straight trajectory line in the measured area, and the trajectory line covers the measured area; the flight altitude of the UAV is set to 1m~2m, the flight speed is 1m / s~3m / s, and the route spacing is 1m~2m; the UAV is equipped with a magnetic detection payload according to the above parameters to complete the collection of the original magnetic background data. The flight track diagram is shown in Figure 2shown.

[0054] The aeromagnetic compensation processing includes: using an unmanned aerial vehicle equipped with a magnetic detection payload to fly to a distance of 40m to 100m from the ground, away from the interference of the complex magnetic field on the surface, hovering in the air to complete maneuvers of pitch ±10°, yaw ±360°, and roll ±10°, and collecting three-component magnetic field data and total magnetic field data; then obtaining the attitude information matrix of the TL model; using the aeromagnetic compensation algorithm to solve and obtain 18 compensation coefficients of the TL model; finally, substituting the 18 compensation coefficients and the original magnetic background data into the TL model to complete the aeromagnetic compensation processing of the original magnetic background data.

[0055] The magnetic background raw data after aeromagnetic compensation is subjected to magnetic diurnal variation correction processing, including:

[0056] Differentiation and processing are performed based on the number of scalar probes in the magnetic detection payload that can collect total magnetic field data, as follows: In the case where the number of scalar probes in the magnetic detection payload that can collect total magnetic field data is 1, a scalar probe that can collect total magnetic field data is synchronously deployed on the ground, using the same time and space reference as the scalar probe carried by the UAV, and collecting data at the same time. The total magnetic field data collected by the scalar probe carried by the UAV is subtracted from the total magnetic field data collected by the scalar probe on the ground to achieve correction of magnetic diurnal variation; In the case where the number of scalar probes in the magnetic detection payload that can collect total magnetic field data is greater than 1, correction of magnetic diurnal variation is achieved by calculating the horizontal or vertical gradient of the total magnetic field data synchronously collected by multiple scalar probes. At this time, the magnetic detection data is plotted into a plane magnetic map, such as Figure 5 As shown in the figure, it can be seen that due to the complex magnetic background, the magnetic anomaly signal of a certain ferromagnetic unexploded bomb has been completely submerged by the noise, and the traditional method cannot analyze it, resulting in the phenomenon of missed detection.

[0057] The step of performing magnetic background removal on the collected data to obtain magnetic detection data after the magnetic background is removed includes:

[0058] After the target shooting mission is completed, the same acquisition method as before the target shooting mission is adopted to obtain the raw data of magnetic detection, and then the aeromagnetic compensation processing and magnetic diurnal variation correction processing are carried out on it in turn;

[0059] The original magnetic background data after magnetic compensation and magnetic diurnal variation processing before the target shooting mission is taken out from the magnetic background database;

[0060] The processed magnetic background raw data and magnetic detection raw data in the measured area are divided according to the route, and classified into magnetic background raw data and magnetic detection raw data of route 1 to route N;

[0061] According to the processed magnetic background raw data taken out from the magnetic background database, the magnetic background estimation data corresponding to the m spatially discrete sampling points of route n are calculated, where n=1, 2, …, N; the spatially discrete sampling points m of each route n may be the same or different; that is, at the spatially discrete sampling points [(x1′, y1′, z1′), (x′2, y′2, z′2), … (x′ m ,y′ m ,z′ m )] corresponding to the estimated magnetic background data [M(x1′,y1′,z1′),M(x2′,y′2,z′2),…M(x′ m ,y′ m ,z′ m )];

[0062] According to the above method, the magnetic detection data of route 1 to route N are calculated as the corresponding magnetic background estimation data at m spatially discrete sampling points;

[0063] For [(x1′,y1′,z1′),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )], the processed raw magnetic detection data is subtracted from the corresponding magnetic background estimation data, that is, the magnetic detection data C(x′) of each sampling point after the magnetic background is removed is obtained. m ,y′ m ,z′ m ) = M′(x′ m ,y′ m ,z′ m )-M(x′ m ,y′ m ,z′ m ). At this time, the magnetic detection data after removing the magnetic background is drawn into a plane magnetic map, such as Figure 6 As shown, after the magnetic background is removed, the magnetic anomaly signal of a ferromagnetic unexploded bomb has been highlighted and will not be missed. The conventional identification and positioning algorithm of the unexploded bomb can be used to identify and locate the unexploded bomb. In this embodiment, after identification and positioning, the depth of the unexploded bomb is 3.6m underground, and the plane positioning accuracy is 0.4m, which also verifies that the system detection depth is better than 3m and the plane positioning accuracy is better than 1m in the complex magnetic background environment of this embodiment.

[0064] The processed magnetic background raw data and magnetic detection raw data in the measured area are divided into magnetic background data and magnetic detection raw data of route 1 to route N according to the route, including:

[0065] The known UAV route planning information and route error are used for classification. Specifically, the UAV route is set to fly along the due north direction, the route accuracy is <0.5m, the number of planned routes is n, n=1,2,…,N, and the horizontal coordinates of each waypoint on route n are x no ; Each spatially discrete sampling point in the magnetic background raw data and magnetic detection raw data has a unique horizontal coordinate x. When x is at (x no -0.5,x no +0.5), the sampling point is classified as a point on route n.

[0066] The calculation of magnetic background estimation data corresponding to m spatially discrete sampling points of route n includes:

[0067] Assume that the processed raw magnetic background data of route n is represented by spatially discrete sampling points [(x1, y1, z1), (x2, y2, z2), … (x n ,y n ,z n )] and its corresponding magnetic field values ​​[M(x1,y1,z1),M(x2,y2,z2),…M(x n ,y n ,z n )]; Assume that the processed raw magnetic detection data of route n is represented by spatially discrete sampling points [(x1′, y1′, z1′), (x′2, y′2, z′2), … (x′ m ,y′ m ,z′ m )] and its corresponding magnetic field values ​​[M′(x1′,y1′,z1′),M′(x′2,y′2,z′2),…M′(x′ m ,y′ m ,z′ m )]; using the spatial interpolation algorithm, we can obtain the function f(x′ m ,y′ m ,z′ m ), [(x1′,y1′,z1′), (x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )] Substitute into the function f(x′ m ,y′ m ,z′ m ), we get the discrete sampling points in space [(x1′,y1′,z1′),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m)], that is, [M(x1′,y1′,z1′),M(x2′,y′2,z′2),…M(x′ m ,y′ m ,z′ m )].

[0068] The spatial interpolation algorithm adopts the Kriging interpolation method, that is, the best linear unbiased prediction value of the unknown point is predicted by calculating the weighted average of each known spatial sampling point.

[0069] The contents not described in detail in the specification of the present invention belong to the common knowledge of the professionals in this field.

Claims

1. A method for removing magnetic background, characterized in that: include: Before the target shooting mission begins, the original magnetic background data of the measured area is collected, and after the aeromagnetic compensation and magnetic diurnal variation correction are completed, it is stored in the magnetic background database; After the target shooting task is completed, the original magnetic detection data of the measured area is collected. According to the data in the magnetic background database before the start of the target shooting task, the magnetic background of the original magnetic detection data of the measured area collected after the target shooting task is eliminated to obtain the magnetic detection data after the magnetic background is eliminated.

2. A magnetic background removal method according to claim 1, characterized in that: Before the target shooting task begins, the original magnetic background data of the measured area is collected, and after the magnetic compensation processing and the magnetic diurnal variation correction processing are completed, the data is stored in the magnetic background database, including: Before the target shooting mission begins, the UAV is equipped with a magnetic detection payload to collect the original magnetic background data of the measured area and obtain the original magnetic background data. Carry out aeromagnetic compensation processing on the collected raw magnetic background data; The original magnetic background data after aeromagnetic compensation is corrected for magnetic diurnal variation. The magnetic background data after aeromagnetic compensation and magnetic diurnal variation correction are stored in the magnetic background database.

3. A magnetic background removal method according to claim 2, characterized in that: The UAV adopts a multi-rotor UAV, and the magnetic detection payload includes a fluxgate probe capable of collecting three-component magnetic field data and at least one scalar probe capable of collecting total magnetic field data, so that the magnetic detection payload can synchronously collect a set of three-component magnetic field data and at least one set of total magnetic field data, and synchronize with the RTK longitude and latitude coordinates, plane coordinates, flight altitude, angle between the UAV nose direction and true north, and time reference information of the UAV to form a data frame, that is, magnetic background raw data.

4. A magnetic background removal method according to claim 2, characterized in that: The raw magnetic background data of the measured area is obtained by autonomous detection by an unmanned aerial vehicle equipped with a magnetic detection payload according to the planned trajectory; the trajectory requirements of the unmanned aerial vehicle are as follows: a "bow"-shaped reciprocating straight trajectory within the measured area, and the trajectory covers the measured area; the flight altitude of the unmanned aerial vehicle is set to 1m~2m, the flight speed is 1m / s~3m / s, and the route spacing is 1m~2m; the unmanned aerial vehicle is equipped with a magnetic detection payload according to the above parameters to complete the collection of raw magnetic background data.

5. The method for removing magnetic background according to claim 1, characterized in that: The aeromagnetic compensation processing includes: using an unmanned aerial vehicle equipped with a magnetic detection payload to fly to a distance of 40m to 100m from the ground, away from the interference of the complex magnetic field on the surface, hovering in the air to complete maneuvers of pitch ±10°, yaw ±360°, and roll ±10°, and collecting three-component magnetic field data and total magnetic field data; then obtaining the attitude information matrix of the TL model; using the aeromagnetic compensation algorithm to solve and obtain 18 compensation coefficients of the TL model; finally, substituting the 18 compensation coefficients and the original magnetic background data into the TL model to complete the aeromagnetic compensation processing of the original magnetic background data.

6. A magnetic background removal method according to claim 1, characterized in that: The magnetic background raw data after aeromagnetic compensation is subjected to magnetic diurnal variation correction processing, including: Differentiation and processing are performed according to the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload, as follows: in the case where the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload is one, a scalar probe capable of collecting total magnetic field data is synchronously deployed on the ground, and the same time and space reference is adopted as the scalar probe carried by the UAV, and data are collected simultaneously, and the total magnetic field data collected by the scalar probe carried by the UAV is subtracted from the total magnetic field data collected by the scalar probe on the ground to achieve correction of magnetic diurnal variation; in the case where the number of scalar probes capable of collecting total magnetic field data in the magnetic detection payload is greater than one, magnetic diurnal variation correction is achieved by performing horizontal or vertical gradient calculation on the total magnetic field data synchronously collected by multiple scalar probes.

7. The method for removing magnetic background according to claim 1, characterized in that: The step of performing magnetic background removal on the collected data to obtain magnetic detection data after the magnetic background is removed includes: After the target shooting mission is completed, the same acquisition method as before the target shooting mission is adopted to obtain the raw data of magnetic detection, and then the aeromagnetic compensation processing and magnetic diurnal variation correction processing are carried out on it in turn; The original magnetic background data after magnetic compensation and magnetic diurnal variation correction before the target shooting mission is taken out from the magnetic background database; The processed magnetic background raw data and magnetic detection raw data in the measured area are divided according to the route, and classified into magnetic background raw data and magnetic detection raw data of route 1 to route N; According to the processed magnetic background raw data taken from the magnetic background database, the corresponding magnetic background estimation data at m spatially discrete sampling points of route n are calculated, n = 1, 2, ..., N, that is, at the spatially discrete sampling points [(x′1, y′1, z′1), (x′2, y′2, z′2), ... (x′ m ,y′ m ,z′ m )] corresponding to the estimated magnetic background data [M(x′1,y′1,z′1),M(x′2,y′2,z′2),…M(x′ m ,y′ m ,z′ m )]; According to the above method, the magnetic detection data of route 1 to route N are calculated as the corresponding magnetic background estimation data at m spatially discrete sampling points; For [(x′1,y′1,z′1),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )], the processed raw magnetic detection data is subtracted from the corresponding magnetic background estimation data, that is, the magnetic detection data C(x′) of each sampling point after the magnetic background is removed is obtained. m ,y′ m ,z′ m ) = M′(x′ m ,y′ m ,z′ m )-M(x′ m ,y′ m ,z′ m ).

8. The method for removing magnetic background according to claim 1, characterized in that: The processed magnetic background raw data and magnetic detection raw data in the measured area are divided into magnetic background data and magnetic detection raw data of route 1 to route N according to the route, including: The known UAV route planning information and route error are used for classification. Specifically, the UAV route is set to fly along the due north direction, the route accuracy is <0.5m, the number of planned routes is n, n=1,2,…,N, and the horizontal coordinates of each waypoint on route n are x no ; Each spatially discrete sampling point in the magnetic background raw data and magnetic detection raw data has a unique horizontal coordinate x. When x is at (x no -0.5,x no +0.5), the sampling point is classified as a point on route n.

9. The method for removing magnetic background according to claim 1, characterized in that: The calculation of magnetic background estimation data corresponding to m spatially discrete sampling points of route n includes: Assume that the processed raw magnetic background data of route n is represented by spatially discrete sampling points [(x1, y1, z1), (x2, y2, z2), … (x n ,y n ,z n )] and its corresponding magnetic field values ​​[M(x1,y1,z1),M(x2,y2,z2),…M(x n ,y n ,z n )]; Assume that the processed raw magnetic detection data of route n is represented by spatially discrete sampling points [(x′1, y′1, z′1), (x′2, y′2, z′2), … (x′ m ,y′ m ,z′ m )] and its corresponding magnetic field values ​​[M′(x′1,y′1,z′1),M′(x′2,y′2,z′2),…M′(x′ m ,y′ m ,z′ m )]; using the spatial interpolation algorithm, we can obtain the function f(x′ m ,y′ m ,z′ m ), convert [(x′1,y′1,z′1),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )] Substitute into the function f(x′ m ,y′ m ,z′ m ), we get the spatially discrete sampling points [(x′1,y′1,z′1),(x′2,y′2,z′2),…(x′ m ,y′ m ,z′ m )], namely [M(x′1,y′1,z′1),M(x′2,y′2,z′2),…M(x′ m ,y′ m ,z′ m )].

10. A magnetic background removal method according to claim 9, characterized in that: The spatial interpolation algorithm adopts the Kriging interpolation method, that is, the best linear unbiased prediction value of the unknown point is predicted by calculating the weighted average of each known spatial sampling point.

Images