A method and device for determining pipeline type and direction based on ground penetrating radar

Abstract

The present invention relates to the field of ground penetrating radar, and in particular to a method and device for determining the type and direction of a pipeline based on ground penetrating radar. The method detects and determines the ground type by emitting mixed-frequency electromagnetic waves, determines the corresponding signal attenuation parameter according to the ground type, and emits a vertical ground direction signal, three first-direction incident oblique signals incident at a direction of 60 degrees to the ground, and three second-direction incident oblique signals incident at a direction of 30 degrees to the ground to determine the type and direction of the pipeline. After the pipeline type is determined, the corresponding signal main frequency is determined to adjust the parameters of Gabor filtering to complete signal filtering, eliminate noise interference to improve the recognition accuracy of the pipeline direction, and proposes a complete pipeline direction judgment logic to improve the recognition rate of pipeline material and pipeline direction.

Description

Technical Field

[0001] The invention relates to the field of ground penetrating radar, and in particular to a method and a device for determining the type and direction of a pipeline based on ground penetrating radar. Background Art

[0002] Ground Penetrating Radar (GPR) is a non-destructive detection technology that detects underground structures and objects by emitting electromagnetic waves into the ground and receiving their reflected signals. Ground penetrating radar is widely used in geological exploration, civil engineering, archaeological research, municipal construction and other fields. Its main advantage is that it can quickly and accurately obtain underground information without excavating the ground.

[0003] The working principle of ground penetrating radar is based on the emission and reflection of electromagnetic waves. When the emitted electromagnetic waves encounter the interface of different media, reflection, refraction and scattering will occur. After the receiver receives the reflected signal, it can infer the location, shape and material of the underground object by analyzing its time delay, intensity and waveform characteristics. Pipes of different materials have different reflection characteristics for electromagnetic waves. Existing technologies have difficulty in accurately distinguishing pipelines of different materials, especially in complex geological environments. The accuracy of pipeline material identification is low and the direction of the pipeline is not accurately judged. Summary of the invention

[0004] In view of the problems existing in the prior art, the purpose of the present invention is to provide a method for determining the type and direction of a pipeline based on ground penetrating radar, so as to effectively solve the above problems.

[0005] To achieve the above object, the technical solution of the present invention is:

[0006] Step s1: Transmit mixed frequency electromagnetic waves to detect and determine the ground type.

[0007] Detecting and determining the ground type is the first step in underground pipeline detection. Different ground types (such as soil, concrete, grass, etc.) have different attenuation and reflection characteristics for electromagnetic waves, so determining the ground type is crucial for subsequent signal processing and pipeline identification.

[0008] Step s11: Emitting electromagnetic waves of mixed frequency in a direction perpendicular to the ground.

[0009] Electromagnetic waves containing high frequency (500MHz-1GHz), medium frequency (200MHz-500MHz) and low frequency (100MHz-200MHz) components are emitted perpendicular to the ground. Electromagnetic waves of different frequency bands have different penetration depths and resolutions, and different ground types have different reflection effects on electromagnetic waves. For example, muddy ground is weak for electromagnetic waves of high, medium and low frequencies, sand has a strong reflection effect on high and medium frequency electromagnetic waves, and asphalt ground has a strong reflection characteristic for high frequency electromagnetic waves.

[0010] Step s12: Receive the reflected signal and perform denoising on the signal.

[0011] The reflected signals of different frequency bands are received simultaneously by a multi-channel receiver. The arrival time of the recorded signal can be used to calculate the transmission distance, and the received reflected signals are denoised to provide data support for subsequent processing.

[0012] Step s13: Extract features of the received reflected signal, including signal time domain features and frequency domain features. The time domain features include the maximum instantaneous intensity and average intensity of the signal, and the frequency domain features include the main frequency features. Sum band energy characteristics ;

[0013] Time domain features: extract the instantaneous intensity and average intensity of the reflected signal.

[0014] Maximum instantaneous strength: Calculate the absolute value of each data point of the received signal and take the maximum instantaneous strength I(t).

[0015] Average intensity: Calculate the average value μ of the instantaneous intensity within a certain time window. The calculation formula is as follows:

[0016] ;

[0017] Frequency domain features: Perform fast Fourier transform (FFT) on the reflected signal to extract spectrum features, including main frequency components and frequency band energy. The FFT formula is as follows:

[0018] ;

[0019] Spectral characteristics: Extract the frequency components of the signal through FFT, analyze its spectral characteristics, and extract its main frequency characteristics Sum band energy characteristics .

[0020] Step s14: Complete ground type recognition based on the signal time domain characteristics and frequency domain characteristics.

[0021] Combine time domain and frequency domain features into a comprehensive feature vector:

[0022] ;

[0023] Application of classification model: The comprehensive feature vector is input into the pre-trained classification model to identify the ground type. Sample data of known ground types are used for annotation, and machine learning algorithms (such as support vector machines, random forests, convolutional neural networks, etc.) are used for model training. According to the characteristics of the real-time received signals, the trained classification model is used to identify the ground type and output the ground type, which includes soil, dry sand, wet sand, and asphalt.

[0024] Step s2: Select a corresponding signal attenuation model according to the determined ground type.

[0025] Select an appropriate electromagnetic wave attenuation model based on the identified ground type. Different ground types (such as soil, sand, asphalt, etc.) have different electromagnetic properties, which will affect the propagation speed and attenuation of electromagnetic waves in the medium. Therefore, different attenuation models need to be selected for different ground types.

[0026] Exponential attenuation model: Applicable to dry sand and asphalt roads. The attenuation of electromagnetic waves in these media is mainly determined by the dielectric constant and conductivity, and the attenuation degree is exponentially related to the propagation distance. The attenuation formula is as follows:

[0027] * ;

[0028] Mixed attenuation model: used for complex media, such as wet sand and soil. The attenuation formula is as follows:

[0029] * ;

[0030] Among them, A(d) is the signal strength when the propagation distance is d, A is the initial signal strength, x is the attenuation coefficient. The specific method of determining the attenuation coefficient can be obtained through experimental measurement. By experimentally measuring the electromagnetic wave propagation characteristics of different ground types, the corresponding attenuation coefficient is obtained, which is usually between 0.01 and 0.1 dB / cm. d is the propagation distance in hundred meters, that is, when the transmission distance is 100m, the value of d is 1. Taking into account the high absorption characteristics of muddy soil and wet soil, it is used to supplement the description of long-distance signal attenuation.

[0031] Step s3: transmitting a vertical ground direction signal, three first-direction oblique incident signals incident at 60 degrees to the ground, and three second-direction oblique incident signals incident at 30 degrees to the ground. The projections of the three first-direction oblique incident signals and the three second-direction oblique incident signals on the horizontal plane are evenly distributed at intervals of 60 degrees. The reflected signal is corrected according to the signal attenuation model, and the characteristics of the reflected signal are extracted to determine the type of the pipeline.

[0032] Pipes made of different materials have different reflection intensities for electromagnetic waves. Metal pipes have the highest reflection intensity, followed by plastic pipes, and concrete pipes have the weakest. Based on the time it takes for the reflected signal to reach the receiver, the propagation speed and path length of the electromagnetic wave are calculated, and the depth and location of the pipeline can be inferred.

[0033] Step s31: the ground penetrating radar transmitter simultaneously transmits seven electromagnetic wave signals;

[0034] The specific signal directions are as follows:

[0035] One vertical ground direction signal: transmitted vertically to the ground, mainly used for preliminary positioning and depth measurement.

[0036] The first direction oblique incident signal: The first direction oblique incident signal includes three signals, which are incident on the ground at an angle of 60 degrees. The angle between the three oblique signals is 120 degrees. The direction of the ground penetrating radar is direction 1, which is used to detect the type and direction of the pipeline.

[0037] Second direction oblique incident signal: The second direction oblique incident signal includes three signals, which are incident on the ground at an angle of 30 degrees. The angle between the three oblique signals is 120 degrees, which is used to detect the type and direction of the pipeline.

[0038] And with the projection of the vertical ground direction signal on the horizontal plane (the location of the ground penetrating radar) as the center point, the projections of the three first-direction oblique incident signals on the horizontal plane are 120 degrees to each other, the projections of the three second-direction oblique incident signals on the horizontal plane are 120 degrees to each other, and the projections of the three first-direction oblique incident signals and the three second-direction oblique incident signals on the horizontal plane are evenly distributed at an angle of 60 degrees. Setting the incident direction in this way can cover the detection range to the greatest extent, without missing the detection area, and ensure the accuracy of the detection effect.

[0039] Step s32: using a multi-channel receiver to simultaneously receive the vertical ground direction signal reflection signal, the three first direction oblique incident signal reflection signals and the three second direction oblique incident signal reflection signals.

[0040] Step s33: Correct the received signal according to the selected attenuation model.

[0041] Select a suitable attenuation model (exponential attenuation model, mixed attenuation model) according to the ground type. If the ground type is dry sand or asphalt road, the attenuation model formula is as follows:

[0042] * ;

[0043] Correspondingly, the correction formula is as follows:

[0044] ;

[0045] If the ground type is wet sand or soil, the attenuation formula is as follows:

[0046] * ;

[0047] Correspondingly, the correction formula is as follows:

[0048] * ;

[0049] Step s34: extract the maximum reflection intensity, average reflection intensity, standard deviation of reflection intensity, main frequency components, frequency band energy characteristics, and signal attenuation coefficient based on the corrected signal, and construct a feature vector for determining the pipeline type, which includes metal pipes, plastic pipes, and concrete pipes.

[0050] Features are extracted from the corrected signal, specifically including time domain features and frequency domain features. The time domain features include instantaneous intensity and average intensity.

[0051] Reflection intensity is an important feature for identifying pipeline materials. Pipes of different materials have different reflection intensities for electromagnetic waves. Record each signal Maximum reflection intensity:

[0052] ;

[0053] Calculate the average reflection intensity of each signal:

[0054] ;

[0055] Calculate the standard deviation of the reflection intensity of each signal:

[0056] ;

[0057] Perform fast Fourier transform (FFT) on the reflected signal to extract the spectrum characteristics, including the main frequency components Sum band energy .

[0058] Calculate the signal attenuation coefficient α: Calculate the attenuation coefficient based on the attenuation of signal strength with distance.

[0059] ;

[0060] Among them, I is the initial signal strength of 0, and d is the propagation distance in hundred meters. That is, when the transmission distance is 100m, the value of d is 1.

[0061] The characteristic vector U of the signal is constructed based on the extracted maximum reflection intensity, average reflection intensity, standard deviation of reflection intensity, main frequency components, frequency band energy characteristics, and signal attenuation coefficient to identify the material of the pipeline.

[0062] U=[ , , , , , α] ;

[0063] Use pipe sample data of known materials for testing and annotation, input the extracted time domain, frequency domain and reflection signal features, and use machine learning algorithms such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN) for model training.

[0064] The extracted feature vector U is input into the trained classification model to output the material type of the pipeline, which includes metal pipeline, plastic pipeline and concrete pipeline.

[0065] Step S4: According to the determined pipeline type, an adaptive Gabor filter is used to filter the vertical ground direction reflection signal, the first direction oblique reflection signal, and the second direction oblique reflection signal.

[0066] Pipes made of different materials have different absorption and reflection characteristics of electromagnetic waves, resulting in different spectral characteristics. Metal pipes reflect more high-frequency components, plastic pipes reflect more medium- and low-frequency components, and concrete pipes reflect more low-frequency components. In order to accurately determine the direction of the pipeline, an adaptive Gabor filter is used to filter the reflected signal so as to better retain the effective characteristics of the signal.

[0067] The spatial domain expression of the Gabor filter is:

[0068] ;

[0069] Among them, λ represents the wavelength of the filter, which determines the frequency selectivity of the filter, and σ represents the standard deviation of the Gaussian envelope, which determines the spatial scale of the filter.

[0070] Different wavelengths are selected according to the main frequency of the extracted signal, and the calculation formula is as follows:

[0071] λ = ;

[0072] Dynamically adjusting the standard deviation σ can retain more signal details. The calculation formula is as follows:

[0073] σ = ;

[0074] Where w represents the main frequency of the signal. The value of w is determined according to the determined pipeline type. If the pipeline type is a metal pipeline, the main frequency of the high-frequency electromagnetic wave component is extracted. If the pipe type is plastic pipe, extract the main frequency of the intermediate frequency electromagnetic wave component If the pipe type is a concrete pipe, extract the main frequency of the low-frequency electromagnetic wave component .

[0075] By adaptively adjusting the filter parameters, useful signal features can be better extracted, the recognition rate and accuracy of pipeline direction can be improved, and noise interference can be reduced.

[0076] Step s5: Determine the direction of the pipeline according to the vertical ground reflection signal, the reflection signals of the three first direction oblique incident signals, and the reflection signals of the three second direction oblique incident signals.

[0077] The maximum reflection intensity of each signal is recorded, including the vertical ground reflection signal, the reflection signals of the three first-direction oblique incident signals, and the reflection signals of the three second-direction oblique incident signals, a total of 7 signals with the maximum reflection intensities I1, I2...I7, and the signal with the strongest reflection intensity Imax1 and the second strongest signal Imax2 are selected from the 7 reflection signals, and the average reflection intensity of each signal is calculated, and the signal with the largest average reflection intensity is selected. .

[0078] Compare Imax2 and The magnitude of the intensity value, if Imax2> At this time, it is determined whether the signals corresponding to Imax1 and Imax2 are reflected signals in the direction perpendicular to the ground. If one of Imax1 and Imax2 is a signal in the direction perpendicular to the ground, the projection direction of the other signal on the ground is taken as the direction of the pipeline; for example, if Imax1 corresponds to a signal in the direction perpendicular to the ground, the projection direction of the signal corresponding to Imax2 on the horizontal plane is taken as the direction of the pipeline.

[0079] If there is no signal in the direction perpendicular to the ground in Imax1 and Imax2, the direction of the Imax1 and Imax2 signals is determined, and the direction of the pipeline is determined according to the direction of the Imax1 and Imax2 signals.

[0080] After determining the signal directions corresponding to the Imax1 and Imax2 signals, project the signal directions corresponding to the Imax1 and Imax2 signals onto the horizontal plane, with the projection point of the vertical ground direction signal, that is, the position of the ground penetrating radar, as the center point, and obtain the corresponding horizontal plane projection directions D1 and D2. At this time, determine whether the angle between direction D1 and direction D2 is 180 degrees. If the angle is 180 degrees, direction D1 and direction D2 are connected to form a straight line, and the direction of this straight line is determined as the direction of the pipeline.

[0081] If the angle between D1 and D2 is not 180 degrees, a ray R1 is made from the center point. The ray R1 bisects the angle between D1 and D2, and the direction D3 perpendicular to the ray R1 is taken as the direction of the pipeline.

[0082] Imax2< , then take The projection direction of the corresponding signal direction on the horizontal plane is the direction of the pipeline. The larger the mean value of the reflected signal, the more closely it matches the direction of the pipeline.

[0083] The direction of the pipeline is displayed on the radar display device, and the user is prompted to move in the direction of the pipeline to continuously update the pipeline direction information. During the movement, the direction is corrected according to the real-time received signal characteristics to ensure the accuracy of the pipeline direction.

[0084] The present invention proposes a method and device for determining the type and direction of a pipeline based on a ground penetrating radar. The method determines the corresponding signal attenuation parameter by identifying the ground type, and transmits a vertical ground direction signal, three first-direction incident oblique signals incident at a direction of 60 degrees to the ground, and three second-direction incident oblique signals incident at a direction of 30 degrees to the ground to determine the type and direction of the pipeline. After the pipeline type is determined, the corresponding signal main frequency is determined to adjust the parameters of the Gabor filter to complete the signal filtering, eliminate noise interference to improve the recognition accuracy of the pipeline direction, and proposes a complete pipeline direction judgment logic to improve the recognition rate of the pipeline material and the pipeline direction. BRIEF DESCRIPTION OF THE DRAWINGS

[0085] Figure 1 A flow chart showing the pipeline type and direction detected based on ground penetrating radar;

[0086] Figure 2 Project the directional diagram of 7-channel signals on the horizontal plane;

[0087] Figure 3 Schematic diagram of signal incident angle;

[0088] Figure 4This is a schematic diagram of pipeline direction confirmation in one embodiment;

[0089] Figure 5 A schematic diagram of pipeline direction confirmation in another embodiment;

[0090] Figure 6 Schematic diagram of the ground penetrating radar device. DETAILED DESCRIPTION

[0091] The specific embodiments of the present invention are further described below in conjunction with the accompanying drawings. It should be noted that the description of these embodiments is used to help understand the present invention, but does not constitute a limitation of the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0092] It should be noted that, in the description of the present invention, the directions or positional relationships indicated by the terms "up", "down", "left", "right", "front", "back", etc. are descriptions of the structure of the present invention based on the accompanying drawings, and are only for the convenience of describing the present invention, and do not indicate or imply that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as a limitation on the present invention.

[0093] The "first" and "second" in this technical solution are only used to distinguish the names of the same or similar structures, or corresponding structures with similar functions, and are not an arrangement of the importance of these structures, nor do they have a ranking, comparison of size, or other meanings.

[0094] In addition, unless otherwise clearly specified and limited, the terms "installation" and "connection" should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, or it can be a connection between the two structures. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood based on the overall idea of ​​the present invention and the specific context of the present solution.

[0095] Embodiment 1

[0096] like Figure 1 As shown in the figure, the processing flow of detecting pipeline type and direction based on ground penetrating radar is as follows:

[0097] Step s1: Transmit mixed frequency electromagnetic waves to detect and determine the ground type.

[0098] Detecting and determining the ground type is the first step in underground pipeline detection. Different ground types (such as soil, concrete, grass, etc.) have different attenuation and reflection characteristics for electromagnetic waves, so determining the ground type is crucial for subsequent signal processing and pipeline identification.

[0099] Step s11: Emitting electromagnetic waves of mixed frequency in a direction perpendicular to the ground.

[0100] Electromagnetic waves containing high frequency (500MHz-1GHz), medium frequency (200MHz-500MHz) and low frequency (100MHz-200MHz) components are emitted perpendicular to the ground. Electromagnetic waves of different frequency bands have different penetration depths and resolutions, and different ground types have different reflection effects on electromagnetic waves. For example, muddy ground is weak for electromagnetic waves of high, medium and low frequencies, sand has a strong reflection effect on high and medium frequency electromagnetic waves, and asphalt ground has a strong reflection characteristic for high frequency electromagnetic waves.

[0101] Step s12: Receive the reflected signal and perform denoising on the signal.

[0102] The reflected signals of different frequency bands are received simultaneously by a multi-channel receiver. The arrival time of the recorded signal can be used to calculate the transmission distance, and the received reflected signals are denoised to provide data support for subsequent processing.

[0103] Step s13: Extract features of the received reflected signal, including signal time domain features and frequency domain features. The time domain features include the maximum instantaneous intensity and average intensity of the signal, and the frequency domain features include the main frequency features. Sum band energy characteristics .

[0104] Time domain features: extract the instantaneous intensity and average intensity of the reflected signal.

[0105] Maximum instantaneous strength: Calculate the absolute value of each data point of the received signal and take the maximum instantaneous strength I(t).

[0106] Average intensity: Calculate the average value μ of the instantaneous intensity within a certain time window. The calculation formula is as follows:

[0107] ;

[0108] Frequency domain features: Perform fast Fourier transform (FFT) on the reflected signal to extract spectrum features, including main frequency components and frequency band energy. The FFT formula is as follows:

[0109] ;

[0110] Spectral characteristics: Extract the frequency components of the signal through FFT, analyze its spectral characteristics, and extract its main frequency characteristics Sum band energy characteristics .

[0111] Step s14: Complete ground type recognition based on the signal time domain characteristics and frequency domain characteristics.

[0112] Combine time domain and frequency domain features into a comprehensive feature vector:

[0113] ;

[0114] Application of classification model: The comprehensive feature vector is input into the pre-trained classification model to identify the ground type. Sample data of known ground types are used for annotation, and machine learning algorithms (such as support vector machines, random forests, convolutional neural networks, etc.) are used for model training. According to the characteristics of the real-time received signals, the trained classification model is used to identify the ground type and output the ground type, which includes soil, dry sand, wet sand, and asphalt.

[0115] Step s2: Select a corresponding signal attenuation model according to the determined ground type.

[0116] Different ground types (such as soil, sand, asphalt, etc.) have different electromagnetic properties, which will affect the propagation speed and attenuation of electromagnetic waves in the medium. Therefore, different attenuation models need to be selected for different ground types.

[0117] Different ground types have different effects on the propagation of electromagnetic waves, mainly including the following aspects:

[0118] Dielectric constant: The dielectric constant of a medium determines the speed at which electromagnetic waves propagate in the medium. The larger the dielectric constant, the slower the electromagnetic wave propagates.

[0119] Conductivity: The conductivity of a medium determines the energy loss of electromagnetic waves in the medium. The greater the conductivity, the faster the electromagnetic waves decay.

[0120] According to the electromagnetic characteristics of the ground type, an appropriate attenuation model can be selected to describe the propagation and attenuation of electromagnetic waves in the medium. Different signal attenuation models are selected according to the ground type:

[0121] Exponential attenuation model: Applicable to dry sand and asphalt roads. The attenuation of electromagnetic waves in these media is mainly determined by the dielectric constant and conductivity, and the attenuation degree is exponentially related to the propagation distance. The attenuation formula is as follows:

[0122] * ;

[0123] Mixed attenuation model: used for complex media, such as wet sand and soil. The attenuation formula is as follows:

[0124] * ;

[0125] Among them, A(d) is the signal strength when the propagation distance is d, A is the initial signal strength, x is the attenuation coefficient. The specific method of determining the attenuation coefficient can be obtained through experimental measurement. By experimentally measuring the electromagnetic wave propagation characteristics of different ground types, the corresponding attenuation coefficient is obtained, which is usually between 0.01 and 0.1 dB / cm. d is the propagation distance in hundred meters, that is, when the transmission distance is 100m, the value of d is 1. Taking into account the high absorption characteristics of muddy soil and wet soil, it is used to supplement the description of long-distance signal attenuation.

[0126] Step s3: transmitting a vertical ground direction signal, three first-direction oblique incident signals incident at 60 degrees to the ground, and three second-direction oblique incident signals incident at 30 degrees to the ground. The projections of the three first-direction oblique incident signals and the three second-direction oblique incident signals on the horizontal plane are evenly distributed at intervals of 60 degrees. The reflected signal is corrected according to the signal attenuation model, and the characteristics of the reflected signal are extracted to determine the type of the pipeline.

[0127] Pipes made of different materials have different reflection intensities for electromagnetic waves. Metal pipes have the highest reflection intensity, followed by plastic pipes, and concrete pipes have the weakest. Based on the time it takes for the reflected signal to reach the receiver, the propagation speed and path length of the electromagnetic wave are calculated, and the depth and location of the pipeline can be inferred.

[0128] Step s31: the ground penetrating radar transmitter simultaneously transmits seven electromagnetic wave signals;

[0129] The specific signal directions are as follows:

[0130] One vertical ground direction signal: transmitted vertically to the ground, mainly used for preliminary positioning and depth measurement.

[0131] The first direction oblique incident signal: The first direction oblique incident signal includes three signals Y1, Y2 and Y3. The three signals are incident on the ground at an angle of 60 degrees. Figure 3 The angle between the three oblique signals is 120 degrees, and their projection on the horizontal plane is shown in the attached figure. Figure 2 As shown, the direction of the ground penetrating radar is taken as direction 1, and the projection direction of the emission direction of signal Y1 on the horizontal plane coincides with direction 1.

[0132] The second direction oblique incident signal: The second direction oblique incident signal includes three signals Z1, Z2 and Z3. The three signals are incident on the ground at an angle of 30 degrees. Figure 3The angle between the three oblique signals is 120 degrees, and their projection on the horizontal plane is shown in the attached figure. Figure 2 As shown, it is used to detect pipeline type and direction.

[0133] And with the projection of the vertical ground direction signal on the horizontal plane (i.e. the location of the ground penetrating radar) as the center point, the projections of the three first-direction oblique incident signals on the horizontal plane are at an angle of 120 degrees to each other, and the projections of the three second-direction oblique incident signals on the horizontal plane are at an angle of 120 degrees to each other, and the projections of the three first-direction oblique incident signals and the three second-direction oblique incident signals on the horizontal plane are evenly distributed at an angle of 60 degrees, as shown in FIG. Figure 2 As shown, setting the incident direction in this way can cover the detection range to the greatest extent, without missing any detected area, thus ensuring the accuracy of the detection effect.

[0134] Step s32: using a multi-channel receiver to simultaneously receive the vertical ground direction signal reflection signal, the three first direction oblique incident signal reflection signals and the three second direction oblique incident signal reflection signals.

[0135] Step s33: Correct the received signal according to the selected attenuation model.

[0136] Select a suitable attenuation model (exponential attenuation model, mixed attenuation model) according to the ground type. If the ground type is dry sand or asphalt road, the attenuation model formula is as follows:

[0137] * ;

[0138] Correspondingly, the correction formula is as follows:

[0139] ;

[0140] If the ground type is wet sand or soil, the attenuation formula is as follows:

[0141] * ;

[0142] Correspondingly, the correction formula is as follows:

[0143] * ;

[0144] Step s34: extract the maximum reflection intensity, average reflection intensity, standard deviation of reflection intensity, main frequency components, frequency band energy characteristics, and signal attenuation coefficient based on the corrected signal, and construct a feature vector for determining the pipeline type, which includes metal pipes, plastic pipes, and concrete pipes.

[0145] Features are extracted from the corrected signal, specifically including time domain features and frequency domain features. The time domain features include instantaneous intensity and average intensity.

[0146] Reflection intensity is an important feature for identifying pipeline materials. Pipes of different materials have different reflection intensities for electromagnetic waves. Record each signal Maximum reflection intensity:

[0147] ;

[0148] Calculate the average reflection intensity of each signal:

[0149] ;

[0150] Calculate the standard deviation of the reflection intensity of each signal:

[0151] ;

[0152] Perform fast Fourier transform (FFT) on the reflected signal to extract the spectrum characteristics, including the main frequency components Sum band energy .

[0153] Calculate the signal attenuation coefficient α: Calculate the attenuation coefficient based on the attenuation of signal strength with distance.

[0154] ;

[0155] Among them, I is the initial signal strength of 0, and d is the propagation distance in hundred meters. That is, when the transmission distance is 100m, the value of d is 1.

[0156] The characteristic vector U of the signal is constructed based on the extracted maximum reflection intensity, average reflection intensity, standard deviation of reflection intensity, main frequency components, frequency band energy characteristics, and signal attenuation coefficient to identify the material of the pipeline.

[0157] U=[ , , , , ,α];

[0158] Use pipe sample data of known materials for testing and annotation, input the extracted time domain, frequency domain and reflection signal features, and use machine learning algorithms such as support vector machine (SVM), random forest (RF), convolutional neural network (CNN) for model training.

[0159] The extracted feature vector U is input into the trained classification model to output the material type of the pipeline, which includes metal pipeline, plastic pipeline and concrete pipeline.

[0160] Step S4: According to the determined pipeline type, an adaptive Gabor filter is used to filter the vertical ground direction reflection signal, the first direction oblique reflection signal, and the second direction oblique reflection signal.

[0161] Pipes made of different materials have different absorption and reflection characteristics of electromagnetic waves, resulting in different spectral characteristics. Metal pipes reflect more high-frequency components, plastic pipes reflect more medium- and low-frequency components, and concrete pipes reflect more low-frequency components. In order to accurately determine the direction of the pipeline, an adaptive Gabor filter is used to filter the reflected signal so as to better retain the effective characteristics of the signal.

[0162] The spatial domain expression of the Gabor filter is:

[0163] ;

[0164] Among them, λ represents the wavelength of the filter, which determines the frequency selectivity of the filter, and σ represents the standard deviation of the Gaussian envelope, which determines the spatial scale of the filter.

[0165] Different wavelengths are selected according to the main frequency of the extracted signal, and the calculation formula is as follows:

[0166] λ = ;

[0167] Dynamically adjusting the standard deviation σ can retain more signal details. The calculation formula is as follows:

[0168] σ = ;

[0169] Where w represents the main frequency of the signal. The value of w is determined according to the determined pipeline type. If the pipeline type is a metal pipeline, the main frequency of the high-frequency electromagnetic wave component is extracted. If the pipe type is plastic pipe, extract the main frequency of the intermediate frequency electromagnetic wave component If the pipe type is a concrete pipe, extract the main frequency of the low-frequency electromagnetic wave component .

[0170] By adaptively adjusting the filter parameters, useful signal features can be better extracted, the recognition rate and accuracy of pipeline direction can be improved, and noise interference can be reduced.

[0171] Step s5: Determine the direction of the pipeline according to the vertical ground reflection signal, the reflection signals of the three first direction oblique incident signals, and the reflection signals of the three second direction oblique incident signals.

[0172] Record the maximum reflection intensity of each signal, including the vertical ground reflection signal, the reflection signals of the three first-direction oblique incident signals, and the reflection signals of the three second-direction oblique incident signals, for a total of 7 signals with maximum reflection intensities I1, I2...I7. Select the signal with the strongest reflection intensity Imax1 and the second strongest signal Imax2 from the 7 reflection signals, calculate the average reflection intensity of each signal, and select the signal with the largest average reflection intensity. .

[0173] Compare Imax2 and The magnitude of the intensity value, if Imax2> At this time, it is determined whether the signals corresponding to Imax1 and Imax2 are reflected signals in the direction perpendicular to the ground. If one of Imax1 and Imax2 is a signal in the direction perpendicular to the ground, the projection direction of the other signal on the horizontal plane is taken as the direction of the pipeline; for example, if Imax1 corresponds to a signal in the direction perpendicular to the ground, the projection direction of the signal corresponding to Imax2 on the horizontal plane is taken as the direction of the pipeline.

[0174] If there is no signal in the direction perpendicular to the ground in Imax1 and Imax2, the direction of the Imax1 and Imax2 signals is determined, and the direction of the pipeline is determined according to the direction of the Imax1 and Imax2 signals.

[0175] After determining the signal directions corresponding to the Imax1 and Imax2 signals, the signal directions corresponding to the Imax1 and Imax2 signals are projected onto the horizontal plane, with the projection point of the vertical ground direction signal, i.e., the position of the ground penetrating radar, as the center point, and the corresponding horizontal plane projection directions D1 and D2 are obtained. At this time, it is determined whether the angle between the direction D1 and the direction D2 is 180 degrees. In a possible embodiment, Figure 4 As shown, if the angle is 180 degrees, direction D1 and direction D2 are connected to form a straight line, and the direction of this straight line is determined as the direction of the pipeline.

[0176] If the angle between D1 and D2 is not 180 degrees, in a possible embodiment, Figure 5 As shown, at this time, a ray R1 is made from the center point, the ray R1 bisects the angle between D1 and D2, and the direction perpendicular to the ray R1 is taken as the direction of the pipeline.

[0177] Imax2< , then take The projection direction of the corresponding signal direction on the horizontal plane is the direction of the pipeline. Because the mean value of the reflected signal is large, it means that it is highly matched with the direction of the pipeline, and its direction can be directly taken as the direction of the pipeline.

[0178] The direction of the pipeline is displayed on the radar display device, and the user is prompted to move in the direction of the pipeline to continuously update the pipeline direction information. During the movement, the direction is corrected according to the real-time received signal characteristics to ensure the accuracy of the pipeline direction.

[0179] Embodiment 2

[0180] like Figure 6 As shown, a device for determining the type and direction of a pipeline based on a ground penetrating radar includes an electromagnetic wave transmitting module 61 for transmitting electromagnetic waves; an electromagnetic wave receiving module 62 for receiving the transmitted electromagnetic waves; a data processing module 63 for completing the above steps s1 to s5 in determining the ground type, the type of the pipeline and the direction of the pipeline; and a display module 64 for displaying the ground type and direction obtained by detection.

[0181] The embodiments of the present invention are described in detail above with reference to the accompanying drawings, but the present invention is not limited to the described embodiments. For those skilled in the art, various changes, modifications, substitutions and variations of these embodiments are made without departing from the principles and spirit of the present invention, and still fall within the scope of protection of the present invention.

Claims

1. A method for determining pipeline type and direction based on ground penetrating radar, characterized in that: Step s1: transmitting mixed frequency electromagnetic waves to detect and determine the ground type; Step s2: selecting a corresponding signal attenuation model according to the determined ground type; Step s3: transmitting a vertical ground direction signal, three first-direction oblique incident signals incident at a direction of 60 degrees to the ground, and three second-direction oblique incident signals incident at a direction of 30 degrees to the ground, the projections of the three first-direction oblique incident signals and the three second-direction oblique incident signals on the horizontal plane are evenly distributed at intervals of 60 degrees, and the reflected signal is corrected according to the signal attenuation model determined in the step s2, and the characteristics of the reflected signal are extracted to determine the type of the pipeline; Step S4: according to the determined pipeline type, using adaptive Gabor filtering to filter the reflection signal in the direction perpendicular to the ground, the reflection signal of the oblique incident signal in the first direction, and the reflection signal of the oblique incident signal in the second direction; Step s5: Determine the pipeline direction according to the reflected signal processed by the adaptive Gabor filter.

2. The method for determining pipeline type and direction based on ground penetrating radar according to claim 1, characterized in that: The step s1 specifically comprises the following steps: Step s11: emitting electromagnetic waves of mixed frequency in a direction perpendicular to the ground; Step s12: receiving the reflected signal and performing denoising on the signal; Step s13: Extract features of the received reflected signal, including signal time domain features and frequency domain features. The time domain features include the maximum instantaneous intensity and average intensity of the signal, and the frequency domain features include the main frequency features. Sum band energy characteristics ; Step s14: Complete ground type recognition based on the signal time domain characteristics and frequency domain characteristics.

3. The method for determining pipeline type and direction based on ground penetrating radar according to claim 1, characterized in that: The ground types include soil, dry sand, wet sand, and asphalt.

4. The method for determining pipeline type and direction based on ground penetrating radar according to claim 3 is characterized in that: The signal attenuation model includes an exponential attenuation model and a mixed attenuation model. If the ground type is dry sand or asphalt road, the exponential attenuation model is selected, and its calculation formula is: * ； If the ground type is wet sand or soil, the mixed attenuation model is selected, and its calculation formula is: * ； Where A(d) is the signal strength when the propagation distance is d, A is the initial signal strength, and x is the attenuation coefficient.

5. The method for determining pipeline type and direction based on ground penetrating radar according to claim 4, characterized in that: The step s3 specifically includes the following processing steps: Step s31: the ground penetrating radar transmitter simultaneously transmits seven electromagnetic wave signals; Step s32: using a multi-channel receiver to simultaneously receive the vertical ground direction signal reflection signal, the three first direction oblique incident signal reflection signals and the three second direction oblique incident signal reflection signals; Step s33: correcting the received signal according to the selected attenuation model; If the ground type is dry sand or asphalt road, the correction formula is: ； If the ground type is wet sand or soil, the correction formula is: * ； Step s34: extract the maximum reflection intensity, average reflection intensity, standard deviation of reflection intensity, main frequency components, frequency band energy characteristics, and signal attenuation coefficient based on the corrected signal, and construct a feature vector for determining the pipeline type, which includes metal pipes, plastic pipes, and concrete pipes.

6. The method for determining pipeline type and direction based on ground penetrating radar according to claim 5, characterized in that: The adaptive Gabor filtering process in step S4 is as follows: The spatial domain expression of the Gabor filter is: ； Among them, λ represents the wavelength of the filter, which determines the frequency selectivity of the filter, and σ represents the standard deviation of the Gaussian envelope, which determines the spatial scale of the filter; Different wavelengths are selected according to the main frequency of the extracted signal, and the calculation formula is as follows: λ= ; Dynamically adjusting the standard deviation σ can retain more signal details. The calculation formula is as follows: σ= ; Where w represents the main frequency of the signal. The value of w is determined according to the determined pipeline type. If the pipeline type is a metal pipeline, the main frequency of the high-frequency electromagnetic wave component is extracted. If the pipe type is plastic pipe, extract the main frequency of the intermediate frequency electromagnetic wave component If the pipe type is a concrete pipe, extract the main frequency of the low-frequency electromagnetic wave component .

7. The method for determining pipeline type and direction based on ground penetrating radar according to claim 1, characterized in that: The step s5 specifically includes: Record the maximum reflection intensity of each signal, including the vertical ground reflection signal, the reflection signals of the three first-direction oblique incident signals, and the reflection signals of the three second-direction oblique incident signals, for a total of 7 signals with maximum reflection intensities I1, I2...I7. Select the signal with the strongest reflection intensity Imax1 and the second strongest signal Imax2 from the 7 reflection signals, calculate the average reflection intensity of each signal, and select the signal with the largest average reflection intensity. ; Compare Imax2 and The magnitude of the intensity value, if Imax2> At this time, it is determined whether the signals corresponding to Imax1 and Imax2 are reflected signals in the direction perpendicular to the ground. If one of Imax1 and Imax2 is a signal in the direction perpendicular to the ground, the projection direction of the other signal on the ground is taken as the direction of the pipeline. If there is no signal in the direction perpendicular to the ground in Imax1 and Imax2, the direction of the Imax1 and Imax2 signals is determined, and the direction of the pipeline is determined according to the direction of the Imax1 and Imax2 signals; After determining the signal directions corresponding to the Imax1 and Imax2 signals, project the signal directions corresponding to the Imax1 and Imax2 signals onto the horizontal plane, with the projection point position of the vertical ground direction signal as the center point, and obtain the corresponding horizontal plane projection directions D1 and D2. At this time, determine whether the angle between direction D1 and direction D2 is 180 degrees. If the angle is 180 degrees, direction D1 and direction D2 are connected to form a straight line, and the direction of this straight line is determined as the direction of the pipeline. If the angle between D1 and D2 is not 180 degrees, a ray R1 is made from the center point. The ray R1 bisects the angle between D1 and D2, and the direction D3 perpendicular to the ray R1 is taken as the direction of the pipeline.

8. The method for determining pipeline type and direction based on ground penetrating radar according to claim 7, characterized in that: Imax2< , then take The projection direction of the corresponding signal direction on the horizontal plane is the direction of the pipeline.

9. A device for determining pipeline type and direction based on ground penetrating radar, characterized in that: It includes an electromagnetic wave transmitting module for transmitting electromagnetic waves; an electromagnetic wave receiving module for receiving the transmitted electromagnetic waves; a data processing module for implementing a method for determining the type and direction of a pipeline based on a ground penetrating radar as described in any one of claims 1 to 7; and a display module for displaying the type and direction of the ground obtained by detection.

Images