Airborne electromagnetic data processing method and system based on ground reference point

Abstract

The application discloses an aviation electromagnetic data processing method and system based on a ground reference point, and comprises the following steps: arranging a transmitting source, collecting first response electromagnetic data corresponding to a plurality of air measuring points; selecting a plurality of ground reference points along the direction of the transmitting source, collecting second response electromagnetic data; performing forward simulation on the first response electromagnetic data after preprocessing, and obtaining a difference value between the ground and the air; loading the difference value into the second response electromagnetic data, and obtaining a reference correction value of any air measuring point in a to-be-detected region by using an interpolation method to obtain third response electromagnetic data; presetting an amplitude difference threshold, comparing the amplitudes of the first response electromagnetic data after preprocessing and the third response electromagnetic data, obtaining a frequency range of motion noise pollution corresponding to an amplitude difference greater than the amplitude difference threshold; and in the frequency range of motion noise pollution, extracting field value amplitudes of the ground reference points and the air measuring points, and performing denoising correction processing by using a least mean square error algorithm.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of geophysical exploration, in particular to an airborne electromagnetic data processing method and system based on a ground reference point. BACKGROUND

[0002] Airborne electromagnetic method is one of the geophysical exploration methods that have attracted much attention in recent years. This method adopts ground / air transmission and air measurement of response magnetic field. This method has the advantages of wide detection range, strong terrain adaptability, low detection cost and high efficiency. The coil sensor is hung on a flight platform such as a helicopter or a drone through a cable for air collection and measurement. Due to the measurement and collection mode and the flight attitude of the flight platform such as the drone, the coil sensor swings and the attitude changes, resulting in motion noise. The collected data is inevitably affected by the motion noise of the coil, and the motion noise is one of the main noise sources of the method, which directly manifests as data polluted by motion noise having a larger amplitude than normal data in the low-frequency part. Therefore, how to reduce or even eliminate the influence of motion noise on airborne electromagnetic data is one of the key factors that determine the quality of airborne electromagnetic data and the detection effect.

[0003] In the Chinese patent with the patent publication number CN114994777A and the name of a ground-air frequency domain electromagnetic motion noise active suppression method, the frequency band range of the main frequency of the transmission current is calculated according to the exploration depth range by using the skin depth formula; the elastic coefficient of the dynamic noise suppression module of the receiving coil sensor is adjusted to reduce the natural frequency f of the receiving coil sensor so that it is not aliasing with the transmission base frequency f01; the angular velocity and acceleration of the pitch direction and the roll direction of the receiving coil sensor are read by the dynamic noise suppression active adjustment module, and the damping coefficient is adjusted according to the angular velocity and acceleration to minimize the acceleration of the receiving coil sensor by changing the damping characteristics.

[0004] In the Chinese patent with the patent publication number CN115097528A and the name of an airborne electromagnetic signal observation device and system carried by a drone, it includes an inner frame, an outer frame and a flexible support connected to each other, the surface of the outer frame is provided with a plurality of hanging points for connecting the cable; the flexible support is installed below the drone, the flexible support is a multi-radiation umbrella structure, the top end of each spoke is connected to the hanging point of the outer frame by a cable; the outer frame is a hollow closed structure, the inner tube of the outer frame is used to install each electronic unit of the receiver; the surface of the outer frame is provided with one or more cable interfaces, the cable interface is used to connect the sensor output signal cable; the inner frame is used to accommodate the inductive magnetic sensor. It can enhance the stability of the sensor in flight and significantly suppress the motion noise.

[0005] In the Chinese patent with the patent publication number CN115356774A and the title "Half-airborne electromagnetic detection device and method based on coaxial coplanar mutual reference coil set", it includes a measurement coil and a reference coil linearly related to additive motion noise, the bandwidths of the reference coil and the measurement coil are consistent and coaxial coplanar, the detection resolution of the reference coil meets the requirement of only being able to distinguish additive motion noise and being unable to distinguish real vertical magnetic field signals, the detection resolution of the measurement coil meets the requirement of being able to distinguish real vertical magnetic field signals and motion noise at the same time, the outer diameter of the reference coil is much smaller than the outer diameter of the measurement coil, and the measurement coil and the reference coil are hard connected or soft connected. The detection method of the invention receives only additive motion noise through one coil, receives additive motion noise and real vertical magnetic field signals through the other coil, and then cancels the additive motion noise received by the two coils to obtain the real vertical magnetic field signal. The device and method of the invention simplify the half-airborne electromagnetic detection measurement system and method, and improve the consistency of the measurement results and the real results.

[0006] In the Chinese patent with the patent publication number CN115510921A and the title "Half-airborne frequency domain electromagnetic detection data noise suppression method based on wavelet de-basis de-noising and notch wave fusion", it includes the following steps: step 1, frequency spectrum analysis of the measured signal containing noise; step 2, the measured signal is first processed according to step 2a, and then the processed signal is processed according to step 2b; step 3, the signal data obtained by processing according to step 2 is filtered to remove other random noise through wavelet threshold de-noising based on frequency domain. The invention discloses a new type of combined noise suppression method, which combines wavelet de-basis de-noising and notch wave processing power frequency interference, can effectively suppress the main noise in the half-airborne electromagnetic detection process, and the processed signal meets the requirements.

[0007] The above patents or technologies mainly process the airborne electromagnetic motion noise from the hardware or algorithm level, which can suppress the motion noise to a certain extent, but due to the randomness of the motion noise, the above technologies cannot completely eliminate the influence of the motion noise.

[0008] Therefore, it is urgent to provide an airborne electromagnetic data processing method and system based on ground reference points, which is simple, accurate and reliable in logic. SUMMARY

[0009] To solve the above problems, the technical scheme adopted by the present application is as follows:

[0010] The first part, the present technology provides an airborne electromagnetic data processing method based on ground reference points, which includes the following steps:

[0011] A transmitting source is laid on the ground in the region to be detected, a coil sensor is carried on a flight platform to fly at a constant speed, and first response electromagnetic data corresponding to a plurality of air measurement points are collected;

[0012] A plurality of ground reference points are selected along the direction of the transmitting source, and second response electromagnetic data are collected at the ground reference points by using the coil sensor;

[0013] The first response electromagnetic data and the second response electromagnetic data are respectively preprocessed;

[0014] The first response electromagnetic data after preprocessing are used for forward modeling, and a difference value between the ground and the air is obtained;

[0015] The difference value is loaded into the second response electromagnetic data, an interpolation method is used to obtain a reference correction value of any air measurement point in the region to be detected, and third response electromagnetic data are obtained;

[0016] An amplitude difference threshold value is preset, the first response electromagnetic data after preprocessing and the third response electromagnetic data are compared in amplitude, and a frequency range of motion noise pollution corresponding to an amplitude difference greater than the amplitude difference threshold value is obtained;

[0017] In the frequency range of motion noise pollution, field value amplitudes of the ground reference points and the air measurement points are extracted, and a least mean square error algorithm is used for denoising correction processing.

[0018] In the second part, the technology provides a system using an airborne electromagnetic data processing method based on ground reference points, which comprises:

[0019] The first response electromagnetic data collection module lays a transmitting source on the ground in the region to be detected, carries a coil sensor on a flight platform to fly at a constant speed, and collects first response electromagnetic data corresponding to a plurality of air measurement points;

[0020] The second response electromagnetic data collection module selects a plurality of ground reference points along the direction of the transmitting source, and collects second response electromagnetic data at the ground reference points by using the coil sensor;

[0021] The preprocessing module is connected with the first response electromagnetic data collection module and the second response electromagnetic data collection module, and pre-processes the first response electromagnetic data and the second response electromagnetic data respectively;

[0022] The difference value obtaining module is connected with the preprocessing module, obtains the first response electromagnetic data after preprocessing, performs forward modeling, and obtains a difference value between the ground and the air;

[0023] The third response electromagnetic data obtaining module is connected with the difference value obtaining module, loads the difference value into the second response electromagnetic data, and obtains the reference correction value of any air measurement point in the detection area by using an interpolation method, so as to obtain the third response electromagnetic data.

[0024] The motion noise pollution frequency band determining module is connected with the third response electromagnetic data obtaining module and the preprocessing module, a preset amplitude difference threshold is set, the amplitudes of the preprocessed first response electromagnetic data and the third response electromagnetic data are compared, and the motion noise pollution frequency band range corresponding to the amplitude difference greater than the amplitude difference threshold is obtained.

[0025] The denoising correction processing module is connected with the motion noise pollution frequency band determining module and the preprocessing module, the field value amplitudes of the ground reference points and the air measurement points in the motion noise pollution frequency band range are extracted, and the denoising correction processing is performed by using the least mean square algorithm.

[0026] Compared with the prior art, the present application has the following beneficial effects:

[0027] The present application ingeniously starts from the root cause of motion noise, corrects the data polluted by motion noise by collecting ground reference point data (not polluted by motion noise), and has good effects compared with other methods.

[0028] The present application ingeniously introduces the least mean square algorithm into the processing of airborne electromagnetic motion noise denoising, and innovatively improves the super parameter setting in the least mean square algorithm according to the characteristics of motion noise, so that the algorithm can meet the convergence accuracy and reduce the calculation amount, and quickly converges.

[0029] In summary, the present application has the advantages of simple logic, accuracy and reliability, and has high practical value and popularization value in the field of geophysical exploration technology. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments. It should be understood that the following drawings only show some embodiments of the present application, and should not be regarded as limiting the scope of protection. For those skilled in the art, other related drawings can also be obtained without creative labor on the basis of these drawings.

[0031] Figure 1 The present application is a logic flow chart.

[0032] Figure 2 The present application is a collection schematic diagram.

[0033] Figure 3 The uniform flight measurement point and reference point position map of the present application.

[0034] Figure 4 The air W1 measurement point field value amplitude map of the present application.

[0035] Figure 5 The forward simulation ground, air field value amplitude and difference comparison map of the present application.

[0036] Figure 6 The air W1 measurement point and ground G1 measurement point corresponding field value amplitude comparison map of the present application.

[0037] Figure 7 The mean square error-time curve map of the present application.

[0038] Figure 8 The air W1 measurement point, ground G1 measurement point and final processing result main frequency amplitude map of the present application. Embodiment

[0039] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application will be further described below in combination with the drawings and embodiments. The embodiments of the present application include but are not limited to the following embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0040] In the present embodiment, the term "and / or" is only used to describe the association relationship of the associated objects, which means that there can be three relationships, for example, A and / or B, which can represent the three cases of A alone, A and B together, and B alone.

[0041] The terms "first" and "second" and the like in the specification and claims of the present embodiment are used to distinguish different objects, and are not used to describe the specific order of the objects. For example, the first target object and the second target object are used to distinguish different target objects, and are not used to describe the specific order of the target objects.

[0042] In the present embodiment, the words "exemplary" or "for example" are used to represent an example, illustration or description. Any embodiment or design scheme described as "exemplary" or "for example" in the present embodiment should not be interpreted as more preferred or more advantageous than other embodiments or design schemes. On the contrary, the words "exemplary" or "for example" are intended to present the relevant concept in a specific way.

[0043] In the description of the embodiments of the present application, the meaning of "a plurality of" is two or more, unless otherwise specified. For example, a plurality of processing units refers to two or more processing units; a plurality of systems refers to two or more systems.

[0044] As shown in Figures 1 to 8 The present embodiment provides a ground reference point-based airborne electromagnetic data processing method, which comprises the following steps:

[0045] First, data acquisition:

[0046] First, lay out the ground transmitter source, collect data at a uniform speed on the designed survey line by the coil sensor carried by the unmanned aerial vehicle, and record the relevant collection parameters. Second, at the corresponding ground position of the survey line, collect a plurality of ground reference point data at certain intervals using the same instrument. The number of ground reference points should be no less than 5% of the number of uniform flight measurement points. The layout diagram is shown in Figure 2 .

[0047] Specifically, according to the terrain conditions, the transmitter source is laid out in a flat terrain. According to the detection depth requirement and the actual grounding resistance, the appropriate transmission main frequency and current size are set, and the transmission parameters such as the transmission endpoint position, length, transmission main frequency and current size are recorded.

[0048] In the present embodiment, the coil sensor and the collection host are carried by the multi-rotor unmanned aerial vehicle to collect data. The multi-rotor unmanned aerial vehicle, the coil sensor and the collection host are connected by a cable. In order to avoid electromagnetic noise of the multi-rotor unmanned aerial vehicle, the distance between the coil sensor and the multi-rotor unmanned aerial vehicle is greater than 10 m, so the length of the hanging cable should be greater than 10 m. The uniform flight speed of the multi-rotor unmanned aerial vehicle should be less than 10 m / s, and the flight parameters such as flight height and speed are recorded. At the same time, the time when the unmanned aerial vehicle flies into and out of the survey line is recorded, which is convenient for subsequent processing. In the present embodiment, the length of the transmitter source is 1624 m, the transmission main frequencies are 25 Hz, 128 Hz, 512 Hz, 1200 Hz and 6400 Hz, and the corresponding current sizes are 5.9 A, 5.9 A, 5.8 A, 5.3 A and 4.2 A. The flight height is 80 m, and the flight speed is 2 m / s.

[0049] In addition, when collecting data at the ground reference points, ground reference points are collected at certain intervals on the corresponding ground position of the survey line. The transmission parameters are consistent with those in the uniform flight, and the coil sensor and the collection host are consistent with those described above. After reaching the ground collection point, the coil sensor needs to be placed horizontally and statically on the ground for collection. The collection time of a single ground collection point is greater than 60 s, and the start and end times of the ground collection point collection are recorded, which is convenient for subsequent processing. At the same time, it needs to be emphasized that the ground reference points must cover the whole survey line position, and the number should not be less than 5% of the number of uniform flight measurement points.

[0050] In the embodiment, the ground reference point data is collected by way one. Unequal interval collection is taken according to actual situation, and the number of collection is 14, as shown in Figure 3 .

[0051] Secondly, data preprocessing, i.e. data processing of the first response electromagnetic data:

[0052] In the embodiment, the data preprocessing objects include the uniform flight data and the ground reference point data. It mainly includes data format conversion, data arrangement, measurement point coordinate projection, field value calculation, data normalization, etc. Among them, the data format conversion is to convert the original data from binary format to ASCII code, including the original time sequence data and GPS data. In addition, the data arrangement is to select the time sequence data and GPS data collected by the uniform flight according to the time recorded when the multi-rotor unmanned aerial vehicle flies in the first step; in addition, to select the time sequence data and GPS data collected by the ground reference point according to the time recorded when the ground reference point collects data in the first step. In addition, the measurement point coordinate projection of the embodiment is to project the converted GPS coordinates of the uniform flight and the ground reference point GPS coordinates together to obtain the relative position diagram of the measurement point and the reference point, as shown in Figure 3 .

[0053] In the embodiment, the field value calculation is to convert the time sequence data from the time domain to the frequency domain, and the specific calculation process is as follows:

[0054] The discrete Fourier transform (DFT) is used to convert the time sequence into frequency domain samples, and the analysis is as follows: is the sampling interval, wherein the sampling interval must meet the following conditions:

[0055] (1)

[0056] Wherein, n represents the serial number of the time domain sample, represents the cut-off frequency, and the sampling length .

[0057] In order to realize the discretization in the Fourier transform, the following discrete formula is obtained:

[0058] (2)

[0059] Wherein, represents the frequency interval; N represents the number of sampling points; represents the field value; represents the serial number of the frequency domain sample; represents the time sequence; ​

[0060] Therefore, according to the above formula, we can obtain:

[0061] (3)

[0062] in, It represents the set of field signals in the frequency domain; e represents the base of the natural logarithm; i is the imaginary unit.

[0063] In this embodiment, data normalization refers to subtracting the emission parameters and gain factor from the field values. The specific calculation formula is as follows:

[0064] (4)

[0065] Where X represents the original field value data. The values ​​are normalized field values, where G is the gain factor, L is the emitter length, and I is the emitter current.

[0066] The third step is the processing of reference point data, which is the processing of the second response electromagnetic data:

[0067] Typically, the number of ground reference points is much smaller than the number of measurement points collected during uniform flight. Therefore, interpolation based on the ground reference points is necessary to obtain the corresponding number of correction data. First, based on the launch and flight parameters recorded in the first step, the difference between the ground and air parameters needs to be obtained through forward modeling. The forward modeling calculation process is quite complex; the field value calculation formula is given directly here, as follows:

[0068] (5)

[0069] in, Indicates the permeability of free space; This represents the current of the emission source; Indicates the offset distance; This indicates the transmission and reception distance of the measuring point relative to the line element; Indicates the length of the emission source; Indicates the ground reflectance coefficient; Indicates the vertical direction; Indicates the location of the sampling point; This represents a first-order Bessel function.

[0070] In this embodiment, the vertical magnetic field values ​​in the air are obtained using formula (5). Vertical magnetic field value of the ground .

[0071] Based on the vertical magnetic field value in the air Vertical magnetic field value of the ground Calculate the difference between the ground and the air. Its expression is:

[0072] (6).

[0073] As Figure 5 shown, it is the ground and the air vertical magnetic field Bz value curve simulated by forward modeling, and the difference value curve of the two.

[0074] On this basis, the embodiment obtains the reference correction value of the full frequency band in the survey line / region range by interpolation.

[0075] Fourth step, noise pollution frequency band judgment:

[0076] On the basis of the third step, the difference in amplitude between the data of the air uniform flight and the ground reference point data is compared and analyzed to judge the frequency band range of the motion noise pollution. Generally, the signal directly affected by the motion noise is that the field value amplitude is larger than the normal signal, but not all frequency bands will be polluted by the motion noise, so it is necessary to judge the frequency band range of the motion noise pollution to reduce the subsequent calculation amount.

[0077] In the embodiment, a plurality of pairs of uniform flight measuring points and ground reference points at the same position are selected to draw their field value amplitudes in batches. By comparing the amplitudes, the frequency band whose uniform flight amplitude is much larger than the ground reference point amplitude is judged as the frequency band polluted by the motion noise.

[0078] In the embodiment, taking the air W1 measuring point and the ground reference point G1 at the same position as an example, as Figure 6 shown, the air W1 measuring point is polluted by the motion noise, and the field value amplitude in the low frequency band is larger than that of the ground reference point G1. Figure 6 It can be seen from

[0079] Fifth step, data denoising correction:

[0080] According to the noise pollution range judged in the fourth step, the field value amplitudes of the reference point and the corresponding measuring point data in the range are extracted, the measuring point field value amplitude is taken as the input signal, and the reference point field value amplitude is taken as the expected signal. According to the characteristics of the motion noise, the improved least mean square algorithm is used for denoising correction processing. Specifically:

[0081] Firstly, define a filter initial coefficient which is 0 by default, and then input the input signal (measuring point field value amplitude) and the expected signal (reference point field value amplitude) into the filter respectively.

[0082] Calculate the output signal of the filter, and calculate the mean square error between the output signal and the expected signal.

[0083] According to the size of the mean square error, the filter coefficient is continuously adjusted so that the mean square error is continuously reduced, which can be achieved by the following formula:

[0084] (7)

[0085] wherein, denotes the learning rate; denotes the mean square error between the output signal and the expected signal; denotes the input signal; denotes the filter coefficient of the n+1th time; denotes the filter coefficient of the nth time;

[0086] The expression of the mean square error between the output signal and the expected signal is:

[0087] (8)

[0088] wherein, N denotes the sample quantity; y(i) denotes the actual output value of the ith sample; denotes the expected output value of the ith sample. The mean square error-time curve is shown in Figure 7 .

[0089] In this embodiment, the learning rate u is an important hyperparameter, which is usually a constant. A larger learning rate has a faster convergence speed, but the convergence accuracy is not enough. A smaller learning rate improves the algorithm convergence accuracy, but more iterations are needed to converge. Therefore, it is usually necessary to determine the best learning rate through multiple experiments.

[0090] This embodiment is aimed at the motion noise characteristics, i.e. low frequency pollution is large and high frequency pollution is small. According to the frequency size, the learning rate is dynamically adjusted. In the low frequency band (large pollution), the learning rate is small, which improves the convergence accuracy. In the high frequency band (small pollution), the learning rate is large, which improves the convergence speed. The expression of the learning rate is:

[0091] (9)

[0092] wherein, denotes the proportional coefficient; denotes the frequency.

[0093] The range of the learning rate of this embodiment is controlled between 0.1-1. When the mean square error reaches the minimum, the output signal is the signal after removing the motion noise, and the final processing result is shown in Figure 8 .

[0094] The above embodiments are only the preferred embodiments of the present application, and are not intended to limit the protection scope of the present application, and any changes made on the basis of the design principles of the present application and non-creative labor shall belong to the protection scope of the present application.

Claims

1. A method for processing airborne electromagnetic data based on a ground reference point, characterized in that, The method comprises the following steps: A transmitting source is laid on the ground of a region to be detected, a coil sensor is carried on a flight platform to fly at a uniform speed, and first response electromagnetic data corresponding to a plurality of air measurement points are collected; A plurality of ground reference points are selected along the direction of the transmitting source, and second response electromagnetic data are collected at the ground reference points by using the coil sensor; The first response electromagnetic data and the second response electromagnetic data are preprocessed respectively; Forward modeling is performed on the preprocessed first response electromagnetic data, and a difference value between the ground and the air is obtained; The difference value is loaded into the second response electromagnetic data, and an interpolation method is used to obtain a reference correction value of any air measurement point in the region to be detected, thereby obtaining third response electromagnetic data; A preset amplitude difference threshold is set, and amplitude comparison is performed on the preprocessed first response electromagnetic data and the third response electromagnetic data, thereby obtaining a frequency range of motion noise pollution corresponding to an amplitude difference greater than the amplitude difference threshold; In the frequency range of motion noise pollution, field value amplitudes of the ground reference points and the air measurement points are extracted, and a least mean square difference algorithm is used for denoising correction processing.

2. The ground reference point based airborne electromagnetic data processing method of claim 1, wherein, The number of ground reference points corresponding to the second response electromagnetic data is not less than 5% of the number of air measurement points corresponding to the first response electromagnetic data.

3. The ground reference point based airborne electromagnetic data processing method of claim 1, wherein, The first response electromagnetic data and the second response electromagnetic data are preprocessed respectively; the preprocessing includes format conversion, data arrangement, measurement point coordinate projection, field value calculation, and data normalization.

4. The ground reference point based airborne electromagnetic data processing method of claim 1, wherein, Forward modeling is performed on the preprocessed first response electromagnetic data, and a difference value between the ground and the air is obtained, including: The vertical magnetic field value is calculated by using field value calculation method which is expressed as: （5） wherein, denotes the vacuum permeability; denotes the current of the transmitter; denotes the offset distance; denotes the offset distance of the measuring point relative to the line element; denotes the length of the transmitter; denotes the ground reflection coefficient; denotes the vertical direction; denotes the sampling point position; denotes the 1st order Bessel function; e denotes the natural logarithm base; The vertical magnetic field values in the air and on the ground are respectively calculated by using formula (5) and formula (6) ; The difference between the values of the vertical magnetic field in the air and on the ground is determined and the values of the vertical magnetic field on the ground The difference between the values of the vertical magnetic field in the air and on the ground is determined which is expressed as: （6）。 5. The ground reference point based airborne electromagnetic data processing method of claim 1, wherein, In the frequency range of motion noise pollution, field value amplitudes of the ground reference points and the air measurement points are extracted, and a least mean square difference algorithm is used for denoising correction processing, including: The expression of the least mean square difference algorithm is: （7） wherein, denotes a learning rate; denotes a mean squared error between the output signal and the desired signal; denotes an input signal; denotes an (n+1)th filter coefficient; denotes an nth filter coefficient; mean square error between the output signal and the desired signal The expression for the mean square error between the output signal and the desired signal is: （8） wherein N represents the number of samples; y(i) represents the actual output value of the ith sample; represents the expected output value of the ith sample.

6. The ground reference point based airborne electromagnetic data processing method of claim 5, wherein, the learning rate The expression for the learning rate is: （9） wherein represents a proportionality coefficient; represents a frequency.

7. A system for processing airborne electromagnetic data based on a ground reference point according to any one of claims 1 to 6, characterized in that, including: The first response electromagnetic data collection module lays a transmitting source on the ground of a region to be detected, carries a coil sensor on a flight platform to fly at a uniform speed, and collects first response electromagnetic data corresponding to a plurality of air measurement points; The second response electromagnetic data collection module selects a plurality of ground reference points along the direction of the transmitting source, and collects second response electromagnetic data at the ground reference points by using the coil sensor; The preprocessing module is connected with the first response electromagnetic data collection module and the second response electromagnetic data collection module, and preprocesses the first response electromagnetic data and the second response electromagnetic data respectively; The difference value obtaining module is connected with the preprocessing module, obtains the preprocessed first response electromagnetic data, performs forward modeling, and obtains a difference value between the ground and the air; The third response electromagnetic data obtaining module is connected with the difference value obtaining module, loads the difference value into the second response electromagnetic data, and uses an interpolation method to obtain a reference correction value of any air measurement point in the region to be detected, thereby obtaining third response electromagnetic data. The motion noise pollution frequency band determination module is connected with the third response electromagnetic data calculation module and the preprocessing module, a preset amplitude difference threshold is set, the amplitudes of the preprocessed first response electromagnetic data and the third response electromagnetic data are compared, and a motion noise pollution frequency band range corresponding to an amplitude difference greater than the amplitude difference threshold is obtained; The denoising correction processing module is connected with the motion noise pollution frequency band determination module and the preprocessing module, the field value amplitudes of the ground reference points and the air measurement points in the motion noise pollution frequency band range are extracted, and a least mean square error algorithm is used for denoising correction processing.

Images