Vehicle detection method, device and equipment based on transverse double geomagnetic fusion

Abstract

The application provides a vehicle detection method, device and equipment based on lateral double geomagnetic fusion in the technical field of vehicle detection, which comprises the following steps: determining the disturbance magnetic field intensity and the disturbance magnetic field change amount at the current time according to the geomagnetic data collected by a geomagnetic sensor; if the disturbance magnetic field intensity at the current time is greater than the disturbance magnetic field noise and the disturbance magnetic field change amount is greater than the disturbance change amount threshold, adding the disturbance magnetic field intensity to the geomagnetic disturbance waveform list; determining the geomagnetic disturbance waveform range, the pulse width and the number of extreme points according to the disturbance magnetic field intensity in the geomagnetic disturbance waveform list; determining the vehicle detection label according to the size relationship between the geomagnetic disturbance waveform range and the plurality of preset range thresholds, the size relationship between the pulse width and the plurality of preset number thresholds, and the size relationship between the number of extreme points and the plurality of preset number thresholds; and performing label fusion according to the vehicle detection labels corresponding to each geomagnetic sensor to determine the vehicle detection result. Thus, the accuracy of vehicle detection can be improved.

Description

TECHNICAL FIELD

[0001] The present disclosure relates to the technical field of vehicle detection, and in particular to a vehicle detection method, device and equipment based on lateral double geomagnetic fusion. BACKGROUND

[0002] In the tunnel intelligent traffic scene, vehicle detection is a key link of traffic state perception and management, and the selection of its technology plays a decisive role in detection accuracy and reliability. The traditional machine vision detection method has shown significant advantages in vehicle detection field due to its high-resolution image acquisition and feature recognition capability. However, due to the constraints of the closed environment of the tunnel, the camera lens is easy to adsorb pollutants such as vehicle exhaust and dust in the tunnel, causing lens dust, which will seriously interfere with the image quality, causing deviation in target feature extraction, and ultimately causing a significant decline in vehicle detection accuracy.

[0003] In related scenarios, vehicle detection technology based on geomagnetic sensors uses the geomagnetic field disturbance caused by the passing of vehicles for vehicle detection, which does not rely on optical imaging and is not affected by environmental factors such as dust and light, and can achieve long-term stable vehicle detection, effectively making up for the performance short board of machine vision detection in harsh environments. However, the geomagnetic vehicle detector is installed in the road in a buried manner, which will cause ground damage and is not conducive to road maintenance, and on the other hand, it is difficult to obtain power, which is not conducive to the continuous operation of the equipment. SUMMARY

[0004] In view of the above-mentioned problems, in combination with the first aspect of the present disclosure, the embodiments of the present disclosure provide a vehicle detection method based on lateral double geomagnetic fusion, which comprises:

[0005] According to the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the side walls of the tunnel cable trench, the disturbance magnetic field strength and the disturbance magnetic field change amount corresponding to the current time are determined, each of the geomagnetic sensors emits a geomagnetic wave to the symmetrically corresponding geomagnetic sensor perpendicular to the side wall of the tunnel cable trench, and the geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic wave;

[0006] In the case that the disturbance magnetic field strength at the current time is greater than the disturbance magnetic field noise and the disturbance magnetic field change amount is greater than the disturbance change amount threshold, the disturbance magnetic field strength is added to the geomagnetic disturbance waveform list, and in the case that the disturbance magnetic field change amount of the continuous preset number of times is less than the waveform stability threshold, the addition of the disturbance magnetic field strength to the geomagnetic disturbance waveform list is stopped;

[0007] According to the disturbance magnetic field strength in the geomagnetic disturbance waveform list, the difference between the maximum disturbance magnetic field strength and the minimum disturbance magnetic field strength is determined to obtain the geomagnetic disturbance waveform range.

[0008] counting the number of the disturbed magnetic field strengths in the list of the geomagnetic disturbance waveforms to obtain a geomagnetic disturbance waveform pulse width, and counting the number of maximum values and minimum values of the disturbed magnetic field strengths in the list of the geomagnetic disturbance waveforms to obtain a number of extreme points;

[0009] According to the size relationship between the geomagnetic disturbance waveform range and a plurality of preset range thresholds, the size relationship between the geomagnetic disturbance waveform pulse width and a plurality of preset number thresholds, and the size relationship between the number of extreme points and the plurality of preset number thresholds, a vehicle detection label is determined, and label fusion is performed according to the vehicle detection labels corresponding to each of the geomagnetic sensors to determine a vehicle detection result.

[0010] In a possible implementation, the determination of the vehicle detection label according to the size relationship between the geomagnetic disturbance waveform range and a first preset range threshold, the size relationship between the geomagnetic disturbance waveform pulse width and a first preset number threshold, and the size relationship between the number of extreme points and the first preset number threshold includes:

[0011] In a case where the number of extreme points is less than a first preset number threshold in the preset number thresholds or the geomagnetic disturbance waveform range is less than or equal to a first preset range threshold in the preset range thresholds, the vehicle detection label is determined as no vehicle.

[0012] In a case where the number of extreme points is greater than a second preset number threshold in the preset number thresholds, if the geomagnetic disturbance waveform range is less than a second preset range threshold in the preset range thresholds, the vehicle detection label is determined as a vehicle, and if the geomagnetic disturbance waveform range is greater than or equal to the second preset range threshold, the vehicle detection label is determined as a suspected vehicle.

[0013] In a case where the number of extreme points is greater than or equal to the first preset number threshold and less than a second preset number threshold, the vehicle detection label is determined according to the number of extreme points, the geomagnetic disturbance waveform pulse width, the geomagnetic disturbance waveform range, and a plurality of preset thresholds.

[0014] In a possible implementation, the determination of the vehicle detection label according to the number of extreme points, the geomagnetic disturbance waveform pulse width, the geomagnetic disturbance waveform range, and a plurality of preset thresholds includes:

[0015] A proportion parameter value is determined according to a ratio between the geomagnetic disturbance waveform range and the geomagnetic disturbance waveform pulse width.

[0016] A waveform feature value is determined according to a product between the proportion parameter value and the number of extreme points.

[0017] The vehicle detection label is determined according to a size relationship between the waveform feature value and a plurality of preset thresholds.

[0018] In one possible implementation, determining the vehicle detection label based on the magnitude relationship between the waveform feature value and multiple preset thresholds includes:

[0019] If the range of the geomagnetic disturbance waveform is greater than or equal to the third preset range threshold among the preset range thresholds, and if the waveform feature value is greater than the first preset threshold among the multiple preset thresholds, then the vehicle detection tag is determined to be a large vehicle.

[0020] If the waveform feature value is greater than the second set threshold among multiple set thresholds and the waveform feature value is less than or equal to the first set threshold, then the vehicle detection tag is determined to be suspected of having a large vehicle.

[0021] If the waveform feature value is less than or equal to the second set threshold, then the vehicle detection tag is determined to be "no vehicle".

[0022] In one possible implementation, the method further includes:

[0023] If the range of the geomagnetic disturbance waveform is less than the third preset range threshold among the preset range thresholds, then an adaptive feature threshold is determined based on the number of extreme points and the range of the geomagnetic disturbance waveform.

[0024] The target feature threshold is determined based on the adaptive feature threshold and the first adaptive coefficient;

[0025] If the waveform feature value is less than or equal to the target feature threshold, and if the waveform feature value is less than or equal to the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "no vehicle"; if the waveform feature value is greater than the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "suspected vehicle".

[0026] If the waveform feature value is greater than the target feature threshold, and the number of extreme points is equal to the first preset number threshold, then the vehicle detection tag is determined to be a suspected vehicle; if the number of extreme points is not equal to the first preset number threshold, then the vehicle detection tag is determined to be a vehicle.

[0027] In one possible implementation, an adaptive feature threshold is determined based on the number of extreme points and the range of the geomagnetic disturbance waveform, including:

[0028] Based on the relationship between the magnitude of the geomagnetic disturbance waveform range and the adaptive pulse width threshold, different adaptive pulse width formulas are selected to determine the adaptive pulse width.

[0029] Based on the relationship between the number of extreme points and the adaptive correction coefficient threshold, different adaptive correction coefficient formulas are selected to determine the adaptive correction coefficient.

[0030] The adaptive feature threshold is determined based on the adaptive pulse width, the adaptive correction coefficient, and the geomagnetic disturbance waveform range.

[0031] In one possible implementation, determining the intensity and change of the disturbed magnetic field at the current moment based on geomagnetic data collected by geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench includes:

[0032] Based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench, the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench are low-pass filtered according to the three axes to obtain filtered geomagnetic data corresponding to each axis in time sequence.

[0033] The actual magnetic field strength corresponding to each of the geomagnetic sensors is determined by summing the squares of the filtered geomagnetic data corresponding to different axes at the same time and taking the arithmetic square root.

[0034] Based on the sampling frequency of the geomagnetic sensor, a sampling interval corresponding to a preset multiple of the sampling frequency is determined, and the actual magnetic field strength corresponding to the current moment is used as the starting point to query the actual magnetic field strength within the sampling interval in reverse order as the target actual sampling intensity.

[0035] Based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target, determine the disturbance magnetic field strength and the change in the disturbance magnetic field at the current moment.

[0036] In one possible implementation, determining the disturbance magnetic field strength and the change in disturbance magnetic field at the current moment based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target includes:

[0037] Calculate the sum of the actual magnetic field strength collected at the current moment and the corresponding actual sampling intensity of the target to obtain the magnetic field strength sum value, and calculate the ratio between the magnetic field strength sum value and the sampling interval length to obtain the mean magnetic field strength;

[0038] Calculate the difference between the actual magnetic field strength collected at the current moment and the mean magnetic field strength, and calculate the square of the difference to obtain the magnetic field strength parameter. Then, sum the magnetic field strength parameter values ​​within the sampling interval and calculate the arithmetic square root of the summed magnetic field strength parameter value to obtain the stable magnetic field strength value.

[0039] Based on whether the stable value of the magnetic field strength is less than a preset stability threshold, it is determined whether the actual magnetic field strength collected at the current moment meets the stability condition. If the actual magnetic field strength collected at the current moment meets the stability condition, the ratio between the magnetic field strength and the sampling interval length is used as the update value to update the reference magnetic field strength.

[0040] If the intensity difference between the actual magnetic field strength collected at the current moment and the updated reference magnetic field strength meets the preset calculation conditions, the intensity difference will be used as the disturbance magnetic field strength corresponding to the current moment.

[0041] The change in the perturbation magnetic field at the current moment is determined based on the perturbation magnetic field strength at the previous moment and the perturbation magnetic field strength at the current moment.

[0042] In conjunction with a second aspect of the present invention, embodiments of the present invention provide a vehicle detection device based on lateral dual geomagnetic fusion, the device comprising:

[0043] The first determining module is configured to determine the intensity and change of the disturbance magnetic field at the current moment based on the geomagnetic data collected by geomagnetic sensors symmetrically installed on the sidewall of the tunnel cable trench. Each of the geomagnetic sensors emits geomagnetic waves perpendicular to the sidewall of the tunnel cable trench towards the corresponding symmetrical geomagnetic sensor, and the geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic waves.

[0044] The second determining module is configured to add the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field strength at the current moment is greater than the disturbance magnetic field noise floor and the disturbance magnetic field change is greater than the disturbance change threshold, and to stop adding the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field change is less than the waveform stability threshold for a consecutive preset number of times.

[0045] The third determining module is configured to determine the difference between the maximum and minimum disturbance magnetic field strengths based on the disturbance magnetic field strengths in the list of geomagnetic disturbance waveforms, and obtain the geomagnetic disturbance waveform range.

[0046] The fourth determining module is configured to count the number of disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the pulse width of the geomagnetic disturbance waveform, and to count the number of maximum and minimum values ​​of the disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the number of extreme points.

[0047] The fifth determining module is configured to determine vehicle detection tags based on the relationship between the range of the geomagnetic disturbance waveform and multiple preset range thresholds, the pulse width of the geomagnetic disturbance waveform and the relationship between the number of extreme points and multiple preset number thresholds, and to perform tag fusion based on the vehicle detection tags corresponding to each geomagnetic sensor to determine the vehicle detection result.

[0048] In conjunction with a third aspect of the present invention, an embodiment of the present invention provides an electronic device, comprising:

[0049] Processor, machine-readable storage medium;

[0050] The machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or code, and the processor is used to execute the programs, instructions or code in the machine-readable storage medium to implement the method described in any one of the first aspects.

[0051] Compared with existing technologies, the vehicle detection method, apparatus, and equipment based on lateral dual geomagnetic fusion provided in this disclosure can achieve at least the following beneficial effects:

[0052] By symmetrically installing geomagnetic sensors on the sidewalls of the tunnel cable trench, ground damage is avoided by burying them in the road surface. This also facilitates power supply to the geomagnetic sensors, ensuring continuous operation. The symmetrically installed sensors collect complementary geomagnetic data, and combined with a vertically emitted geomagnetic wave design, effectively suppresses environmental noise interference with single-point vehicle detection, improving the signal-to-noise ratio of the disturbed magnetic field signal. Furthermore, a dual judgment method using the noise floor and change threshold of the disturbed magnetic field intensity is employed to avoid false triggering of detection due to instantaneous interference. Recording is terminated after multiple consecutive drops below the waveform stability threshold, reducing invalid data storage. A composite feature set is constructed through joint analysis of the geomagnetic disturbance waveform range, pulse width, and number of extreme points, enabling the differentiation between vehicles and non-vehicle targets. Vehicle detection tags generated independently by the symmetrical sensors are then fused to address the missed detection problem caused by single-point sensor occlusion or local interference, improving the accuracy and robustness of vehicle detection in tunnel scenarios. The preset range threshold and number threshold can be dynamically adjusted according to the tunnel environment to adapt to the differentiated needs of different highway tunnels.

[0053] Other features and advantages of this disclosure will be described in detail in the following detailed description section. Attached Figure Description

[0054] The accompanying drawings are provided to further illustrate the present disclosure and form part of the specification. They are used together with the following detailed description to explain the present disclosure, but do not constitute a limitation thereof. In the drawings:

[0055] Figure 1This is a schematic diagram of the execution flow of a vehicle detection method based on lateral dual geomagnetic fusion provided in an embodiment of the present invention.

[0056] Figure 2 This is a schematic diagram of the execution flow of another vehicle detection method based on lateral dual geomagnetic fusion provided in an embodiment of the present invention.

[0057] Figure 3 This is a schematic block diagram of a vehicle detection device based on lateral dual geomagnetic fusion provided in an embodiment of the present invention.

[0058] Figure 4 This is a schematic diagram of exemplary hardware and software components of a vehicle detection device based on lateral dual geomagnetic fusion provided in an embodiment of the present invention. Detailed Implementation

[0059] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0060] The specific embodiments of this disclosure will be described in detail below with reference to the accompanying drawings. It should be understood that the specific embodiments described herein are for illustration and explanation only and are not intended to limit this disclosure.

[0061] This invention provides a vehicle detection method based on lateral dual geomagnetic fusion, see [link to relevant documentation]. Figure 1 As shown, the method includes:

[0062] In step S11, based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewall of the tunnel cable trench, the intensity and change of the disturbance magnetic field at the current moment are determined. Each of the geomagnetic sensors emits geomagnetic waves perpendicular to the sidewall of the tunnel cable trench towards the corresponding symmetrical geomagnetic sensor. The geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic waves.

[0063] The symmetrically installed geomagnetic sensors are deployed symmetrically on both sides of the tunnel cable trench to collect spatially complementary geomagnetic signals. The perturbed magnetic field strength is the difference (in nT) between the magnetic field strength detected by the geomagnetic sensor and the background geomagnetic field. The perturbed magnetic field change is the difference (in nT / s) between the perturbed magnetic field strengths at adjacent moments, reflecting the dynamic rate of change of the magnetic field. Vertically emitted geomagnetic waves are achieved by the geomagnetic sensor actively emitting electromagnetic waves of a specific frequency (simulating geomagnetic signals) and receiving the reflected signals from the symmetrical sensors, thus enhancing signal stability.

[0064] In this embodiment, the symmetrical sensor forms a closed detection path by transmitting / receiving geomagnetic waves and uses the principle of magnetic field vector superposition to eliminate environmental interference (such as slow changes in the Earth's magnetic field). When a vehicle (metal body) passes by, its magnetic field disturbance will affect the sensors on both sides simultaneously, but the disturbance phase is different due to the positional difference. By comparing the data on both sides, the effective disturbance features can be extracted.

[0065] For example, suppose a car passes through a tunnel. The left sensor detects that the magnetic field strength increases from 50,000 nT to 50,020 nT (disturbance +20 nT), while the right sensor detects that it decreases from 50,000 nT to 49,980 nT (disturbance -20 nT). The disturbance changes are calculated as +20 nT / s on the left and -20 nT / s on the right, and the symmetry is used to verify that it is a vehicle disturbance.

[0066] In step S12, when the current disturbance magnetic field strength is greater than the disturbance magnetic field noise floor and the disturbance magnetic field change is greater than the disturbance change threshold, the disturbance magnetic field strength is added to the geomagnetic disturbance waveform list. When the disturbance magnetic field change is less than the waveform stability threshold for a preset number of consecutive times, the addition of the disturbance magnetic field strength to the geomagnetic disturbance waveform list is stopped.

[0067] Among them, the disturbance magnetic field noise floor is the threshold of magnetic field fluctuation caused by environmental noise when no vehicles are passing by (e.g., ±5nT). The disturbance change threshold is the minimum magnetic field change rate that triggers recording (e.g., ±10nT / s). The waveform stability threshold is used to determine the end of the disturbance (e.g., ±2nT / s); if it falls below this value multiple times consecutively, the disturbance is considered to have stopped.

[0068] In this embodiment, transient interference (such as people walking) is filtered out using a dual threshold (noise floor + change amount), recording only continuous and significant magnetic field disturbances. A waveform stabilization threshold is used to terminate invalid recordings and avoid data redundancy.

[0069] For example, at a certain moment, the left sensor detects a disturbance change of +15 nT / s (exceeding the threshold of 10 nT / s), and the current disturbance intensity is +25 nT (exceeding the noise floor of 5 nT), triggering recording. Subsequently, it detects three more changes of +1 nT / s, -0.5 nT / s, and +0.8 nT / s (all below the stability threshold of 2 nT / s), and then stops recording.

[0070] Waveform extraction for vehicle inspection is mainly divided into start extraction and end extraction. The current disturbance magnetic field strength is set to... The strength of the disturbed magnetic field at the previous moment was Change in the disturbed magnetic field The list of geomagnetic disturbance waveforms is as follows .

[0071] Waveform extraction begins: If and This indicates that the geomagnetic data has begun to fluctuate, and the strength of the fluctuating magnetic field at this time will be... Add to the list of geomagnetic disturbance waveforms middle.

[0072] in, Indicates the background noise of the disturbed magnetic field; This is the threshold value for the change in the perturbed magnetic field; the specific value can be obtained through experimental testing. In this invention, It is 5. It is 0.3.

[0073] Waveform end extraction judgment: If continuous Second-rate Less than the waveform stability threshold If the signal is strong enough, the geomagnetic fluctuation is considered to have ended, and the waveform feature extraction process begins; otherwise, the current perturbed magnetic field strength is added. The list of geomagnetic disturbance waveforms is as follows .

[0074] In this invention, It is 10. It is 0.2.

[0075] In step S13, based on the disturbance magnetic field strength in the list of geomagnetic disturbance waveforms, the difference between the maximum and minimum disturbance magnetic field strength is determined to obtain the geomagnetic disturbance waveform range.

[0076] Among them, the geomagnetic disturbance waveform range (i.e., the range) is the difference between the maximum and minimum values ​​in the geomagnetic disturbance waveform (unit: nT), which reflects the range of disturbance amplitude.

[0077] In this embodiment, a larger range indicates a more severe disturbance (e.g., a large vehicle with a high metal content); a smaller range may indicate a small target or interference. The range can be used to distinguish vehicle types (e.g., engineering vehicles vs. maintenance tools). Geomagnetic disturbance waveform range (i.e., range): The disturbance intensity sequence recorded in a certain instance is [+12, +15, +10, -8, -5] nT, and the range is 15 - (-8) = 23nT.

[0078] In step S14, the number of disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms is counted to obtain the pulse width of the geomagnetic disturbance waveform, and the number of maximum and minimum values ​​of the disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms is counted to obtain the number of extreme points.

[0079] The pulse width of the geomagnetic disturbance waveform is the total number of data points (unit: number of sampling points) in the geomagnetic disturbance waveform list, reflecting the duration of the disturbance. The number of extreme points is the number of local maxima and minima in the geomagnetic disturbance waveform list, reflecting the complexity of the magnetic field fluctuations. Pulse width: ,Right now The number of points stored in the database; the number of extreme points: , which is the number of maxima and minima in the waveform.

[0080] In step S15, vehicle detection tags are determined based on the relationship between the range of the geomagnetic disturbance waveform and multiple preset range thresholds, the pulse width of the geomagnetic disturbance waveform, and the relationship between the number of extreme points and multiple preset number thresholds. Then, tag fusion is performed based on the vehicle detection tags corresponding to each geomagnetic sensor to determine the vehicle detection result.

[0081] Among them, the preset range threshold is used to distinguish the range threshold of vehicle type (e.g., range > 20nT indicates a large vehicle). The preset number threshold is used to distinguish the extreme point threshold of vehicle type (e.g., extreme point > 3 indicates a multi-axle vehicle). Label fusion combines the detection results of symmetrical sensors and determines the final vehicle detection result through voting or weighted algorithms.

[0082] In this embodiment of the disclosure, vehicle detection labels (such as "large vehicle", "small vehicle", "interference") are generated by comparing the range, pulse width, extreme points, and threshold values. Symmetrical sensor label fusion can eliminate single-point false detections (such as when one side of the sensor is blocked).

[0083] The aforementioned technical solution utilizes symmetrically installed geomagnetic sensors on the sidewalls of tunnel cable trenches, avoiding burial in the road surface and preventing ground damage. This also facilitates power supply to the geomagnetic sensors, ensuring continuous operation. The symmetrically installed sensors collect complementary geomagnetic data, and combined with a vertically emitted geomagnetic wave design, effectively suppresses environmental noise interference with single-point vehicle detection, improving the signal-to-noise ratio of the disturbed magnetic field signal. Furthermore, a dual judgment based on the noise floor and change threshold of the disturbed magnetic field intensity is employed to prevent false triggering of detection due to transient interference. Recording is terminated after multiple consecutive drops below the waveform stability threshold, reducing invalid data storage. A composite feature set is constructed through joint analysis of the geomagnetic disturbance waveform range, pulse width, and number of extreme points, enabling the differentiation between vehicles and non-vehicle targets. Vehicle detection tags generated independently by the symmetrical sensors are then fused, resolving the issue of missed detections caused by single-point sensor occlusion or local interference, thus improving the accuracy and robustness of vehicle detection in tunnel scenarios. The preset range threshold and number threshold can be dynamically adjusted according to the tunnel environment to adapt to the differentiated needs of different highway tunnels.

[0084] In a possible implementation manner, in step S15, determining the vehicle detection label according to the relationship between the range difference of the geomagnetic disturbance waveform and the first preset range difference threshold, the pulse width of the geomagnetic disturbance waveform, and the relationship between the number of extreme points and the first preset number threshold includes:

[0085] In step S151, when the number of extreme points is less than the first preset number threshold in the preset number threshold or the range difference of the geomagnetic disturbance waveform is less than or equal to the first preset range difference threshold in the preset range difference threshold, determine that the vehicle detection label is no vehicle;

[0086] Among them, the first preset number threshold (T1): the minimum number of extreme points for distinguishing vehicles from non-vehicles (e.g., T1 = 2). The first preset range difference threshold (R1): the maximum range difference determined to be no vehicle (e.g., R1 = 15 nT). No vehicle label: The current disturbance is caused by non-vehicle targets (such as personnel, tools) or environmental noise.

[0087] For example, when the number of extreme points < T1 (the disturbance waveform is simple) or the range difference < R1 (the degree of geomagnetic disturbance is relatively gentle), it is determined to be no vehicle. For example, the geomagnetic disturbance in the current lane caused by a car in the adjacent lane may produce a low range difference or low extreme points.

[0088] In step S152, when the number of extreme points is greater than the second preset number threshold in the preset number threshold, if the range difference of the geomagnetic disturbance waveform is less than the second preset range difference threshold, determine that the vehicle detection label is a vehicle; if the change range difference is greater than or equal to the second preset range difference threshold, determine that the vehicle detection label is suspected to have a vehicle;

[0089] Among them, the second preset number threshold (T2): the minimum number of extreme points determined to be a vehicle (e.g., T2 = 3). The second preset range difference threshold (R2): the range difference threshold for distinguishing vehicle types or interference (e.g., R2 = 25 nT). Vehicle label: The number of extreme points > T2 and the range difference < R2, which conforms to the magnetic field disturbance characteristics of a vehicle.

[0090] Suspected to have a vehicle label: The number of extreme points > T2 but the range difference ≥ R2, which may be a large vehicle or complex interference.

[0091] In the embodiments of the present disclosure, the number of extreme points > T2 indicates that the disturbance is complex (such as the multi-axis structure of a vehicle), and the range difference < R2 is a typical vehicle disturbance (moderate metal content). If the range difference ≥ R2, it may be an extra-large vehicle (such as an engineering vehicle) or interference superposition, and further analysis is required.

[0092] For example, in a vehicle scenario: the extreme point = 4 (> T2 = 3), the range = 20 nT (< R2 = 25 nT). It is determined as "there is a vehicle", which may be a car. In a suspected vehicle scenario: the extreme point = 5 (> T2 = 3), the range = 30 nT (≥ R2 = 25 nT). The system determines it as "suspected of having a vehicle", which may be due to the presence of other interference sources near the sensor.

[0093] In step S153, when the number of extreme points is greater than or equal to the first preset number threshold and less than the second preset number threshold, the vehicle detection label is determined according to the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple set thresholds.

[0094] Among them, the third preset range threshold (R3): a higher threshold (such as R3 = 40 nT), which is used to distinguish extreme disturbances (such as extra-large vehicles or severe interference). Multiple set thresholds: including a pulse width threshold (W1), a range gradient threshold, etc., which are used to refine the determination.

[0095] In the embodiments of the present disclosure, when the extreme point ≥ T1 and < T2, it is necessary to further analyze multi-dimensional features such as the pulse width and the range gradient. For example: if the pulse width is long (the vehicle passing time is long) and the range gradient is gentle (the disturbance is continuous and stable), it is determined as "extra-large vehicle". If the pulse width is short and the range gradient is steep (instantaneous severe disturbance), it is determined as "severe interference".

[0096] In a possible implementation manner, in step S153, the determining the vehicle detection label according to the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple set thresholds includes:

[0097] In step S1531, a proportional parameter value is determined according to the ratio between the range of the geomagnetic disturbance waveform and the pulse width of the geomagnetic disturbance waveform;

[0098] In the embodiments of the present disclosure, the proportional parameter value is used to distinguish instantaneous severe disturbances (such as a metal tool falling) from continuous and stable disturbances (such as a vehicle passing). Instantaneous disturbances usually show a high range but a short pulse width (large P value), while vehicle disturbances show a medium range but a long pulse width (small P value). Among them, the proportional parameter value is range / pulse width, that is .

[0099] In step S1532, a waveform feature value is determined according to the product of the proportional parameter value and the number of extreme points;

[0100] Among them, the waveform feature value = proportional parameter value * number of extreme points, that is .

[0101] In step S1533, the vehicle detection label is determined according to the magnitude relationship between the waveform feature value and multiple set thresholds.

[0102] In the embodiments of the present disclosure, by comparing the waveform feature value F with the set threshold, and combining the service logic to determine the final label. For example:

[0103] F ≤ F1: The perturbation is simple, and it is determined as "no vehicle".

[0104] F1 < F ≤ F2: The perturbation is complex but does not reach the vehicle characteristics, and it is determined as "suspected vehicle" or "complex interference".

[0105] F > F2: The perturbation is severe and complex, and it is determined as "extra-large vehicle" or "severe interference".

[0106] In a possible implementation manner, in step S1533, determining the vehicle detection label according to the magnitude relationship between the waveform feature value and multiple set thresholds includes:

[0107] In step S21, when the极差 of the geomagnetic disturbance waveform is greater than or equal to the third preset极差 threshold (R3) among the preset极差 thresholds, if the waveform feature value is greater than the first set threshold among the multiple set thresholds, it is determined that the vehicle detection label is "there is a large vehicle";

[0108] Among them, the third preset极差 threshold (R3): a higher threshold (such as R3 = 40 nT), which is used to distinguish extreme perturbations (such as extra-large vehicles or severe interference).

[0109] When the waveform feature value F > the first set threshold (F1), it indicates that the perturbation has the following characteristics:

[0110] High amplitude: The极差 (R) is large, indicating that the magnetic field perturbation is severe (such as a large amount of metal in a large vehicle).

[0111] Long duration: The pulse width (W) is long, indicating that the perturbation is continuous and stable (such as a long passing time of a vehicle).

[0112] Complex waveform: The number of extreme points (N) is large, indicating that the magnetic field fluctuation is complex (such as a multi-axis structure or interference superposition).

[0113] Comprehensive determination: Such perturbations are usually caused by extra-large vehicles (such as engineering vehicles, heavy trucks), and need to be marked as "there is a large vehicle".

[0114] In step S22, if the waveform feature value is greater than the second set threshold among the multiple set thresholds and the waveform feature value is less than or equal to the first set threshold, it is determined that the vehicle detection label is "suspected to have a large vehicle";

[0115] It should be noted that the term "极差" in the original text seems to be a specific parameter in the relevant context, and the direct translation may not be very accurate without more background information. Here it is translated as "极差" first, and you may need to adjust it according to the actual situation. Also, the tags -

[0115] are preserved as they are.When the waveform eigenvalue satisfies the second set threshold (F2) < F ≤ F1, it indicates that the disturbance has the following characteristics:

[0116] Medium amplitude: The combination of the range (R) and the pulse width (W) does not reach the severity of F1, but exceeds the threshold of ordinary vehicles.

[0117] Medium complexity: The number of extreme points (N) is relatively large, but the waveform may be affected by interference (such as the superposition of vehicles and electromagnetic noise).

[0118] Comprehensive determination: Such disturbances may be caused by large vehicles (such as ordinary trucks) or complex interference (such as the simultaneous presence of vehicles and construction equipment), and need to be marked as "suspected large vehicle".

[0119] In step S23, if the waveform eigenvalue is less than or equal to the second set threshold, it is determined that the vehicle detection label is no vehicle.

[0120] When the waveform eigenvalue F ≤ the second set threshold (F2), it indicates that the magnetic field disturbance does not conform to the characteristic law caused by large vehicles in this lane (such as the disturbance of large vehicles in the adjacent lane). No further processing is required, and it is marked as "no vehicle".

[0121] In a possible implementation manner, the method further includes:

[0122] If the range of the geomagnetic disturbance waveform is less than the third preset range threshold among the preset range thresholds, the adaptive characteristic threshold is determined according to the number of extreme points and the range of the change amount;

[0123] Among them, when the range of the change amount is less than the third preset range threshold (F3), it indicates that the current disturbance amplitude is relatively low, mainly concentrated on small and medium-sized vehicles or interference from the adjacent lane. At this time, the threshold needs to be dynamically adjusted to adapt to different vehicle scenarios and avoid misjudgment. Adaptive characteristic threshold: Jointly calculated according to the number of extreme points and the range of the geomagnetic disturbance waveform. Different small vehicles may have different ranges of the geomagnetic disturbance waveform and the number of extreme points due to different iron contents in themselves. In order to avoid misjudgment, the threshold needs to be dynamically adjusted to adapt to different situations.

[0124] According to the adaptive characteristic threshold and the first adaptive coefficient, the target characteristic threshold is determined;

[0125] By dynamically adjusting the target characteristic threshold through the first adaptive coefficient, the tunnel environment with different interference levels can be adapted, and the robustness can be improved.

[0126] If the waveform feature value is less than or equal to the target feature threshold, and if the waveform feature value is less than or equal to the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "no vehicle". If the waveform feature value is greater than the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "suspected vehicle".

[0127] When the waveform feature value is low, it is necessary to further distinguish between "no vehicle" and "suspected vehicle" by using an adaptive feature threshold and a second adaptive coefficient.

[0128] If the waveform feature value is greater than the target feature threshold, and the number of extreme points is equal to the first preset number threshold, then the vehicle detection tag is determined to be a suspected vehicle; if the number of extreme points is not equal to the first preset number threshold, then the vehicle detection tag is determined to be a vehicle.

[0129] When the waveform characteristic value is high, it is necessary to distinguish between "suspected vehicle" and "vehicle" by the number of extreme points and the first preset threshold number.

[0130] The dual-threshold design can adapt to different levels of interference (such as high-interference / low-interference tunnels), improving the accuracy of judgment.

[0131] In one possible implementation, an adaptive feature threshold is determined based on the number of extreme points and the range of the geomagnetic disturbance waveform, including:

[0132] Based on the relationship between the magnitude of the geomagnetic disturbance waveform range and the adaptive pulse width threshold, different adaptive pulse width formulas are selected to determine the adaptive pulse width.

[0133] The determination of the adaptive pulse width requires considering the relationship between the geomagnetic disturbance waveform range and the adaptive pulse width threshold, using a dynamic adjustment method to adapt to scenarios with different waveform complexities.

[0134] If the range of the geomagnetic disturbance waveform is greater than or equal to the adaptive pulse width threshold, it indicates high waveform disturbance (such as in small and medium-sized trucks), and the pulse width needs to be increased to reflect the overall duration of the waveform. If the range of the geomagnetic disturbance waveform is less than the adaptive pulse width threshold, it indicates low waveform disturbance (such as in cars), and the pulse width needs to be reduced. By dynamically adjusting the pulse width, the system can adapt to waveforms of different complexities, improving the accuracy of feature extraction.

[0135] Based on the relationship between the number of extreme points and the adaptive correction coefficient threshold, different adaptive correction coefficient formulas are selected to determine the adaptive correction coefficient.

[0136] The determination of the adaptive correction coefficient needs to consider the relationship between the number of extreme points, the range of the geomagnetic disturbance waveform, and the threshold of the adaptive correction coefficient, so as to dynamically adjust the calculation method of the correction coefficient to adapt to scenarios with different waveform amplitudes.

[0137] The adaptive feature threshold is determined based on the adaptive pulse width, the adaptive correction coefficient, and the number of extreme points.

[0138] The determination of the adaptive feature threshold requires combining the adaptive pulse width, adaptive correction coefficient, and the number of extreme points. By integrating these factors, the adaptive feature threshold can dynamically adapt to the waveform characteristics of different scenarios, improving the accuracy and robustness of vehicle detection. For example: For high-complexity waveforms: increasing the adaptive pulse width increases the adaptive correction coefficient, and the adaptive feature threshold reflects strong disturbances. For low-complexity waveforms: decreasing the adaptive pulse width decreases the adaptive correction coefficient, and the adaptive feature threshold reflects weak disturbances.

[0139] For example, see Figure 2 As shown, if the range is less than 3000, an adaptive calculation of the target feature threshold (plus feature threshold) is used (the adaptive formula is constructed through extensive statistical analysis and pattern summarization). If the waveform feature value (waveform plus feature value) is greater than... Plus feature threshold and If the value is greater than 1, it is considered "confirmed vehicle"; otherwise, it is considered "suspicious vehicle". If the feature threshold is multiplied by 1, it is considered a "suspected vehicle"; otherwise, it is considered a "not a vehicle". To determine the vehicle control ratio (default is 1). The two values ​​are the control ratio for suspected vehicles (default is 0.2), and can be adjusted according to the actual tunnel detection situation.

[0140] If the range is greater than or equal to 3000, it indicates that a vehicle with a strong magnetic field has passed by, most likely a large vehicle. Therefore, the plus feature value is directly compared with a set threshold (a hyperparameter obtained through statistics). Determine a threshold for large vehicles (default is 80). This is the threshold for suspected large vehicles (default value is 50).

[0141] For example, for features as ; ; We can conclude that: ; ;

[0142] From the number of extreme points Furthermore, since the range is less than 3000, it falls under the adaptive calculation and judgment part. According to the formula, we can obtain:

[0143]

[0144] Conditional judgment:

[0145] The final output is the "Confirm Vehicle" label.

[0146] In one possible implementation, step S11, determining the intensity and change of the disturbed magnetic field at the current moment based on the geomagnetic data collected by geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench, includes:

[0147] In step S111, based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench, the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench are low-pass filtered according to the three axes to obtain filtered geomagnetic data corresponding to each axis in time sequence.

[0148] In this embodiment of the disclosure, the triaxial (X, Y, Z) geomagnetic data collected by symmetrically installed geomagnetic sensors are subjected to low-pass filtering to remove high-frequency noise (such as power equipment interference and electromagnetic pulses) and retain low-frequency effective signals (such as magnetic field disturbances caused by vehicles).

[0149] Filtering method: Use a finite impulse response (FIR) or infinite impulse response (IIR) low-pass filter, and set the cutoff frequency according to the frequency range of magnetic field disturbance caused by the vehicle (usually 0.1~10Hz).

[0150] Axis-by-axis processing: Since the components of the geomagnetic field along each axis may be affected by different interference sources (such as cables distributed along the X-axis in tunnels and vehicles moving along the Y-axis), they need to be filtered independently to avoid inter-axis interference.

[0151] Timing synchronization: The filtered data must maintain temporal consistency to ensure the accuracy of subsequent analysis.

[0152] Low-pass filtering is used to eliminate high-frequency noise and improve the signal-to-noise ratio of magnetic field data.

[0153] In step S112, the arithmetic square root of the sum of the squares of the filtered geomagnetic data corresponding to different axes at the same time is obtained to determine the actual magnetic field strength corresponding to each geomagnetic sensor.

[0154] The actual magnetic field strength of the geomagnetic sensor is obtained by summing the squares of the triaxial filtered geomagnetic data at the same time and then taking the arithmetic square root. This actual magnetic field strength reflects the total amplitude of the geomagnetic field and is directly related to the intensity of the disturbance magnetic field caused by vehicles. This eliminates the amplitude cancellation problem caused by directional differences in the triaxial data. Standardizing the dimensions facilitates subsequent comparisons with the background magnetic field or threshold values.

[0155] In step S113, based on the sampling frequency of the geomagnetic sensor, a sampling interval corresponding to a preset multiple of the sampling frequency is determined, and the actual magnetic field strength corresponding to the current moment is used as the starting point to query the actual magnetic field strength within the sampling interval in reverse order as the target actual sampling strength.

[0156] In this embodiment of the disclosure, based on the sampling frequency of the geomagnetic sensor, a sampling interval ([t−k / fs,t]) corresponding to a preset multiple of the sampling frequency is determined, and the actual magnetic field strength at the current time (t) is used as the starting point to query the actual magnetic field strength within this interval in reverse order as the target actual sampling intensity:

[0157] Sampling interval selection: k is usually taken as 5 to 10, corresponding to a sampling interval of 0.5 to 1 second, in order to cover short-term disturbances caused by vehicles.

[0158] Reverse order query: Traverse the data within the sampling interval backward from the current time to ensure that the target sampling intensity is related to the magnetic field state at the current time.

[0159] By limiting the analysis range through preset sampling intervals, interference from irrelevant historical data is avoided. At the same time, reverse query ensures that the target sampling intensity reflects the magnetic field background or disturbance trend at the current moment.

[0160] In step S114, the disturbance magnetic field strength and the change in disturbance magnetic field at the current moment are determined based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target.

[0161] In this embodiment of the disclosure, the intensity of the disturbed magnetic field is used to quantify the degree of magnetic field disturbance at the current moment, serving as a direct basis for vehicle detection. The change in the disturbed magnetic field is used to quantify the dynamic characteristics of the magnetic field disturbance, assisting in distinguishing vehicle types (e.g., fast-moving vehicles cause high changes).

[0162] The above technical solution eliminates high-frequency noise and retains effective disturbance signals through triaxial low-pass filtering and vector synthesis, thereby improving data quality. By pre-setting sampling intervals and reverse-order queries, a background magnetic field model is dynamically established to adapt to the complex electromagnetic environment within tunnels. Combining the intensity and variation of the disturbed magnetic field, the disturbance is quantified from both amplitude and dynamic characteristics, improving the accuracy of vehicle detection. Short-time sampling intervals and instantaneous change calculations ensure real-time system response, while low-pass filtering and background modeling enhance anti-interference capabilities. In this way, reliable disturbance magnetic field characteristics can be extracted from raw geomagnetic data.

[0163] In one possible implementation, step S114, determining the disturbance magnetic field strength and the change in disturbance magnetic field at the current moment based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target, includes:

[0164] In step S1141, the sum of the actual magnetic field strength collected at the current moment and the corresponding actual sampling intensity of the target is calculated to obtain the sum of magnetic field strengths, and the ratio between the sum of magnetic field strengths and the length of the sampling interval is calculated to obtain the mean magnetic field strength.

[0165] This process involves calculating the sum of the current actual magnetic field strength and the target actual sampling strength, and then dividing this sum by the sampling interval length (N, i.e., the number of sampling points) to obtain the mean magnetic field strength. The mean magnetic field strength reflects the average level of the magnetic field strength within the sampling interval and is used to assess the stability of the current magnetic field state. This process can eliminate the influence of transient noise on the magnetic field strength. By calculating the mean, the current moment's data is combined with historical sampling data to form a dynamic background magnetic field model, adapting to the time-varying nature of the electromagnetic environment within the tunnel.

[0166] In step S1142, the difference between the actual magnetic field strength collected at the current time and the mean magnetic field strength is calculated, and the square of the difference is calculated to obtain the magnetic field strength parameter. The magnetic field strength parameter within the sampling interval is summed, and the arithmetic square root of the summed magnetic field strength parameter is calculated to obtain the stable magnetic field strength value.

[0167] The process involves calculating the difference between the actual magnetic field strength at the current moment and the mean magnetic field strength, squaring this difference to obtain the magnetic field strength parameter, summing all parameter values ​​within the sampling interval, and taking the square root to obtain the stable magnetic field strength value. This stable value reflects the degree of fluctuation in magnetic field strength within the sampling interval and is used to quantify magnetic field stability. This allows for the assessment of whether the magnetic field is affected by vehicle disturbances or noise interference, providing a quantitative basis for determining stability conditions. Through square root and square root operations, magnetic field fluctuations are converted into scalar values ​​for easy comparison with preset thresholds.

[0168] In step S1143, based on whether the stable value of the magnetic field strength is less than a preset stability threshold, it is determined whether the actual magnetic field strength collected at the current moment meets the stability condition. If the actual magnetic field strength collected at the current moment meets the stability condition, the ratio between the magnetic field strength and the sampling interval length is used as the update value to update the reference magnetic field strength.

[0169] In this embodiment, it is determined whether the stable value of the magnetic field strength is less than a preset stability threshold. If the condition is met, the current magnetic field state is considered stable, and the average magnetic field strength is used as the update value to update the reference magnetic field strength. The reference magnetic field strength reflects the background magnetic field level when there is no vehicle disturbance and needs to be dynamically updated to adapt to environmental changes (such as temperature drift and geomagnetic diurnal variation). This ensures that the reference magnetic field strength matches the current environment and avoids misjudgment of disturbances caused by environmental changes. By judging the stability condition, it is ensured that the reference is updated only when the magnetic field is stable, avoiding interference from vehicle disturbances or noise with the reference value.

[0170] In step S1144, if the intensity difference between the actual magnetic field strength collected at the current moment and the updated reference magnetic field strength meets the preset calculation conditions, the intensity difference is used as the disturbance magnetic field strength corresponding to the current moment.

[0171] In this embodiment, when the difference between the actual magnetic field strength at the current moment and the updated reference magnetic field strength meets preset calculation conditions, this difference is used as the disturbance magnetic field strength at the current moment. The disturbance magnetic field strength reflects the degree to which the magnetic field caused by the vehicle deviates from the background magnetic field and is a core indicator for vehicle detection. The amplitude of vehicle disturbance is quantified. Minor fluctuations (such as noise) are filtered to avoid misjudgment. Preset calculation conditions (such as a minimum threshold) ensure that only significant disturbances are responded to, improving detection reliability.

[0172] In step S1145, the change in the disturbance magnetic field at the current moment is determined based on the disturbance magnetic field strength at the current moment and the disturbance magnetic field strength at the previous moment.

[0173] In this embodiment, the change in the disturbed magnetic field is calculated based on the current and previous magnetic field strengths. This change reflects the dynamic characteristics of the magnetic field disturbance and is used to distinguish whether a vehicle has passed by. This assists in vehicle type identification and improves adaptability to complex motion scenarios. The instantaneous rate of change of the magnetic field disturbance is quantified through time difference calculation.

[0174] To summarize the above example, a geomagnetic sensor can sample raw geomagnetic data at 100Hz. Due to instability and noise in the data during acquisition, filtering and baseline determination are necessary. Specifically:

[0175] Step 1: Input the raw triaxial geomagnetic data at time t , , (It is a direct digital signal, and t also represents the number of sample points acquired.)

[0176] Step 2: Perform IIR low-pass filtering on the raw triaxial geomagnetic data, denoted as... , , ;

[0177] Step 3: Convert the filtered triaxial geomagnetic data into geomagnetic intensity. The calculation formula is:

[0178]

[0179] Step 4: Baseline value calculation and update.

[0180] In the process of geomagnetic vehicle detection, the judgment is based on the sudden shift in magnetic field strength. In order to obtain the shift in magnetic field strength, the magnetic field strength under the current environment without vehicle disturbance is used as the reference value, and the shift in magnetic field strength is obtained by subtracting the reference value from the measured geomagnetic strength.

[0181] Benchmark value calculation rules: Define a benchmark strength list with a size of [size missing]. Double the sampling frequency for storing continuous Intensity data per second will be continuous The average magnetic field strength under conditions of no vehicle interference is used as the baseline value. The stability value is used depending on whether vehicle interference is present. The evaluation is conducted using the following formula:

[0182]

[0183] in Sampling frequency, This represents the geomagnetic intensity. If the stable value is less than the stable threshold (default is 50), it is assumed that there is no vehicle interference, and the baseline value can be calculated; otherwise, the baseline value is not updated. The baseline value calculation formula is:

[0184]

[0185] Step 5: Calculation of the intensity of the disturbed magnetic field.

[0186] If the baseline value is greater than 0, it indicates that baseline data already exists. Calculate the intensity of the disturbing magnetic field:

[0187] .

[0188] The above technical solution dynamically updates the reference magnetic field strength through mean value calculation and stability condition judgment, adapting to the time-varying characteristics of the electromagnetic environment within the tunnel. Noise interference is eliminated and the strength of significantly disturbing magnetic fields is quantified through stable value calculation and minimum threshold filtering. Combining the disturbing magnetic field strength (amplitude) and its change (dynamic characteristics), vehicle disturbances are described from multiple dimensions, improving detection accuracy. Short-time sampling intervals and real-time reference updates ensure rapid system response to environmental changes, while multi-level condition judgments enhance anti-interference capabilities. It can extract high-quality disturbing magnetic field features from raw geomagnetic data.

[0189] This invention provides a vehicle detection device based on lateral dual geomagnetic fusion, see [link / reference]. Figure 3 As shown, the device includes:

[0190] The first determining module 210 is configured to determine the intensity and change of the disturbance magnetic field at the current moment based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewall of the tunnel cable trench. Each of the geomagnetic sensors emits geomagnetic waves perpendicular to the sidewall of the tunnel cable trench to the corresponding symmetrical geomagnetic sensor, and the geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic waves.

[0191] The second determining module 220 is configured to add the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field strength at the current moment is greater than the disturbance magnetic field noise floor and the disturbance magnetic field change is greater than the disturbance change threshold, and to stop adding the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field change is less than the waveform stability threshold for a consecutive preset number of times.

[0192] The third determining module 230 is configured to determine the difference between the maximum disturbance magnetic field strength and the minimum disturbance magnetic field change based on the disturbance magnetic field strength in the list of geomagnetic disturbance waveforms, and obtain the geomagnetic disturbance waveform range.

[0193] The fourth determining module 240 is configured to count the number of disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the pulse width of the geomagnetic disturbance waveform, and to count the number of maximum and minimum values ​​of the disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the number of extreme points.

[0194] The fifth determining module 250 is configured to determine vehicle detection tags based on the relationship between the range of the geomagnetic disturbance waveform and multiple preset range thresholds, the pulse width of the geomagnetic disturbance waveform and the relationship between the number of extreme points and multiple preset number thresholds, and to perform tag fusion based on the vehicle detection tags corresponding to each geomagnetic sensor to determine the vehicle detection result.

[0195] In one possible implementation, the fifth determining module 250 is configured as follows:

[0196] If the number of extreme points is less than the first preset number threshold among the preset number thresholds, or if the range of the geomagnetic disturbance waveform is less than or equal to the first preset range threshold among the preset range thresholds, the vehicle detection tag is determined to be no vehicle.

[0197] If the number of extreme points is greater than the second preset number threshold among the preset number thresholds, and the range of the geomagnetic disturbance waveform is less than the second preset range threshold among the preset range thresholds, then the vehicle detection tag is determined to have a vehicle; if the range of the geomagnetic disturbance waveform is greater than or equal to the second preset range threshold, then the vehicle detection tag is determined to have a suspected vehicle.

[0198] If the number of extreme points is greater than or equal to the first preset threshold and less than the second preset threshold, then the vehicle detection tag is determined based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds.

[0199] In one possible implementation, determining the vehicle detection tag based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds includes:

[0200] The proportional parameter is determined based on the ratio between the range of the geomagnetic disturbance waveform and the pulse width of the geomagnetic disturbance waveform;

[0201] The waveform characteristic value is determined by multiplying the proportional parameter value by the number of extreme points.

[0202] The vehicle detection tag is determined based on the relationship between the waveform feature value and multiple set thresholds.

[0203] In one possible implementation, the fifth determining module 250 is configured as follows:

[0204] If the range of the geomagnetic disturbance waveform is greater than or equal to the third preset range threshold among the preset range thresholds, and if the waveform feature value is greater than the first preset threshold among the multiple preset thresholds, then the vehicle detection tag is determined to be a large vehicle.

[0205] If the waveform feature value is greater than the second set threshold among multiple set thresholds and the waveform feature value is less than or equal to the first set threshold, then the vehicle detection tag is determined to be suspected of having a large vehicle.

[0206] If the waveform feature value is less than or equal to the second set threshold, then the vehicle detection tag is determined to be "no vehicle".

[0207] In one possible implementation, the fifth determining module 250 is configured as follows:

[0208] If the range of the geomagnetic disturbance waveform is less than the third preset range threshold among the preset range thresholds, then an adaptive feature threshold is determined based on the number of extreme points and the range of the geomagnetic disturbance waveform.

[0209] The target feature threshold is determined based on the adaptive feature threshold and the first adaptive coefficient;

[0210] If the waveform feature value is less than or equal to the target feature threshold, and if the waveform feature value is less than or equal to the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "no vehicle". If the waveform feature value is greater than the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "suspected vehicle".

[0211] If the waveform feature value is greater than the target feature threshold, and the number of extreme points is equal to the first preset number threshold, then the vehicle detection tag is determined to be a suspected vehicle; if the number of extreme points is not equal to the first preset number threshold, then the vehicle detection tag is determined to be a vehicle.

[0212] In one possible implementation, the fifth determining module 250 is configured as follows:

[0213] Based on the relationship between the magnitude of the geomagnetic disturbance waveform range and the adaptive pulse width threshold, different adaptive pulse width formulas are selected to determine the adaptive pulse width.

[0214] Based on the relationship between the number of extreme points and the adaptive correction coefficient threshold, different adaptive correction coefficient formulas are selected to determine the adaptive correction coefficient.

[0215] The adaptive feature threshold is determined based on the adaptive pulse width, the adaptive correction coefficient, and the geomagnetic disturbance waveform range.

[0216] In one possible implementation, the root first determining module 210 is configured as follows:

[0217] Based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench, the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench are low-pass filtered according to the three axes to obtain filtered geomagnetic data corresponding to each axis in time sequence.

[0218] The actual magnetic field strength corresponding to each of the geomagnetic sensors is determined by summing the squares of the filtered geomagnetic data corresponding to different axes at the same time and taking the arithmetic square root.

[0219] Based on the sampling frequency of the geomagnetic sensor, a sampling interval corresponding to a preset multiple of the sampling frequency is determined, and the actual magnetic field strength corresponding to the current moment is used as the starting point to query the actual magnetic field strength within the sampling interval in reverse order as the target actual sampling intensity.

[0220] Based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target, determine the disturbance magnetic field strength and the change in the disturbance magnetic field at the current moment.

[0221] In one possible implementation, the first determining module 210 is configured as follows:

[0222] Calculate the sum of the actual magnetic field strength collected at the current moment and the corresponding actual sampling intensity of the target to obtain the magnetic field strength sum value, and calculate the ratio between the magnetic field strength sum value and the sampling interval length to obtain the mean magnetic field strength;

[0223] Calculate the difference between the actual magnetic field strength collected at the current moment and the mean magnetic field strength, and calculate the square of the difference to obtain the magnetic field strength parameter. Then, sum the magnetic field strength parameter values ​​within the sampling interval and calculate the arithmetic square root of the summed magnetic field strength parameter value to obtain the stable magnetic field strength value.

[0224] Based on whether the stable value of the magnetic field strength is less than a preset stability threshold, it is determined whether the actual magnetic field strength collected at the current moment meets the stability condition. If the actual magnetic field strength collected at the current moment meets the stability condition, the ratio between the magnetic field strength and the sampling interval length is used as the update value to update the reference magnetic field strength.

[0225] If the intensity difference between the actual magnetic field strength collected at the current moment and the updated reference magnetic field strength meets the preset calculation conditions, the intensity difference will be used as the disturbance magnetic field strength corresponding to the current moment.

[0226] The change in the perturbation magnetic field at the current moment is determined based on the perturbation magnetic field strength at the previous moment and the perturbation magnetic field strength at the current moment.

[0227] This invention provides an electronic device, comprising:

[0228] Processor, machine-readable storage medium;

[0229] The machine-readable storage medium is connected to the processor. The machine-readable storage medium is used to store programs, instructions, or code. The processor is used to execute the programs, instructions, or code in the machine-readable storage medium to implement the method described in any of the foregoing embodiments.

[0230] Figure 4 The vehicle detection device 100 based on lateral dual geomagnetic fusion shown includes a processor 1001 and a memory 1003. The processor 1001 and the memory 1003 are connected, for example, via a bus 1002. Optionally, the vehicle detection device 100 based on lateral dual geomagnetic fusion may further include a communication component 1004, which can be used for data interaction between the device 100 and other devices, such as data transmission and / or data reception. It should be noted that in actual scheduling, the communication component 1004 is not limited to one, and the structure of this vehicle detection device 100 based on lateral dual geomagnetic fusion does not constitute a limitation on the embodiments of this application.

[0231] Processor 1001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 1001 may also be a combination that implements computing functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0232] Bus 1002 may include a pathway for transmitting information between the aforementioned components. Bus 1002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Bus 1002 can be divided into address bus, data bus, control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0233] The memory 1003 may be ROM (Read Only Memory) or other types of static storage devices capable of storing static information and instructions, RAM (Random Access Memory) or other types of dynamic storage devices capable of storing information and instructions, or EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media, other magnetic storage devices, or any other medium capable of carrying or storing program code and capable of being read by a computer, without limitation herein.

[0234] The memory 1003 is used to store program code for executing embodiments of the present disclosure, and its execution is controlled by the processor 1001. The processor 1001 is used to execute the program code stored in the memory 1003 to implement the steps shown in the foregoing embodiments of the vehicle detection method based on lateral dual geomagnetic fusion.

[0235] The preferred embodiments of the present disclosure have been described in detail above with reference to the accompanying drawings. However, the present disclosure is not limited to the specific details of the above embodiments. Within the scope of the technical concept of the present disclosure, various changes, modifications, substitutions and variations can be made to these embodiments, and all such changes, modifications, substitutions and variations fall within the protection scope of the present disclosure.

[0236] It should also be noted that the various specific technical features described in the above embodiments can be combined in any suitable manner without contradiction, and such combinations should also be considered as part of this disclosure. To avoid unnecessary repetition, this disclosure will not further describe the various possible combinations. The technical scope of this application is not limited to the contents of the specification, but must be determined according to the scope of the claims.

Claims

1. A vehicle detection method based on lateral dual geomagnetic fusion, characterized in that, The method includes: Based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewall of the tunnel cable trench, the intensity and change of the disturbance magnetic field at the current moment are determined. Each of the geomagnetic sensors emits geomagnetic waves perpendicular to the sidewall of the tunnel cable trench towards the corresponding symmetrical geomagnetic sensor, and the geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic waves. If the current disturbance magnetic field strength is greater than the disturbance magnetic field noise floor and the disturbance magnetic field change is greater than the disturbance change threshold, the disturbance magnetic field strength is added to the geomagnetic disturbance waveform list. If the disturbance magnetic field change is less than the waveform stability threshold for a preset number of consecutive times, the addition of the disturbance magnetic field strength to the geomagnetic disturbance waveform list is stopped. Based on the disturbance magnetic field strengths listed in the geomagnetic disturbance waveform list, the difference between the maximum and minimum disturbance magnetic field strengths is determined to obtain the geomagnetic disturbance waveform range. The number of disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms is counted to obtain the pulse width of the geomagnetic disturbance waveform. The number of maximum and minimum values ​​of the disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms is also counted to obtain the number of extreme points. Based on the relationship between the range of the geomagnetic disturbance waveform and multiple preset range thresholds, the pulse width of the geomagnetic disturbance waveform, and the relationship between the number of extreme points and multiple preset number thresholds, vehicle detection tags are determined, and tag fusion is performed based on the vehicle detection tags corresponding to each geomagnetic sensor to determine the vehicle detection result. The step of determining the vehicle detection tag based on the relationship between the range of the geomagnetic disturbance waveform and a first preset range threshold, the pulse width of the geomagnetic disturbance waveform, and the relationship between the number of extreme points and a first preset number threshold includes: If the number of extreme points is less than the first preset number threshold among the preset number thresholds, or if the range of the geomagnetic disturbance waveform is less than or equal to the first preset range threshold among the preset range thresholds, the vehicle detection tag is determined to be no vehicle. If the number of extreme points is greater than the second preset number threshold among the preset number thresholds, and the range of the geomagnetic disturbance waveform is less than the second preset range threshold among the preset range thresholds, then the vehicle detection tag is determined to have a vehicle; if the range of the geomagnetic disturbance waveform is greater than or equal to the second preset range threshold, then the vehicle detection tag is determined to have a suspected vehicle. If the number of extreme points is greater than or equal to the first preset threshold and less than the second preset threshold, then the vehicle detection tag is determined based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds. The step of determining the vehicle detection tag based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds includes: The proportional parameter is determined based on the ratio between the range of the geomagnetic disturbance waveform and the pulse width of the geomagnetic disturbance waveform; The waveform characteristic value is determined by multiplying the proportional parameter value by the number of extreme points. The vehicle detection tag is determined based on the relationship between the waveform feature value and multiple set thresholds.

2. The vehicle detection method based on lateral dual geomagnetic fusion according to claim 1, characterized in that, The step of determining the vehicle detection tag based on the relationship between the waveform feature value and multiple preset thresholds includes: If the range of the geomagnetic disturbance waveform is greater than or equal to the third preset range threshold among the preset range thresholds, and if the waveform feature value is greater than the first preset threshold among the multiple preset thresholds, then the vehicle detection tag is determined to be a large vehicle. If the waveform feature value is greater than the second set threshold among multiple set thresholds and the waveform feature value is less than or equal to the first set threshold, then the vehicle detection tag is determined to be suspected of having a large vehicle. If the waveform feature value is less than or equal to the second set threshold, then the vehicle detection tag is determined to be "no vehicle".

3. The vehicle detection method based on lateral dual geomagnetic fusion according to claim 1, characterized in that, The method further includes: If the range of the geomagnetic disturbance waveform is less than the third preset range threshold among the preset range thresholds, then an adaptive feature threshold is determined based on the number of extreme points and the range of the geomagnetic disturbance waveform. The target feature threshold is determined based on the adaptive feature threshold and the first adaptive coefficient; If the waveform feature value is less than or equal to the target feature threshold, and if the waveform feature value is less than or equal to the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "no vehicle"; if the waveform feature value is greater than the product of the adaptive feature threshold and the second adaptive coefficient, the vehicle detection label is determined to be "suspected vehicle". If the waveform feature value is greater than the target feature threshold, and the number of extreme points is equal to the first preset number threshold, then the vehicle detection tag is determined to be a suspected vehicle; if the number of extreme points is not equal to the first preset number threshold, then the vehicle detection tag is determined to be a vehicle.

4. The vehicle detection method based on lateral dual geomagnetic fusion according to claim 3, characterized in that, Based on the number of extreme points and the range of the geomagnetic disturbance waveform, an adaptive feature threshold is determined, including: Based on the relationship between the magnitude of the geomagnetic disturbance waveform range and the adaptive pulse width threshold, different adaptive pulse width formulas are selected to determine the adaptive pulse width. Based on the relationship between the number of extreme points and the adaptive correction coefficient threshold, different adaptive correction coefficient formulas are selected to determine the adaptive correction coefficient. The adaptive feature threshold is determined based on the adaptive pulse width, the adaptive correction coefficient, and the geomagnetic disturbance waveform range.

5. The vehicle detection method based on lateral dual geomagnetic fusion according to any one of claims 1-4, characterized in that, The determination of the current moment's disturbance magnetic field strength and change based on geomagnetic data collected by geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench includes: Based on the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench, the geomagnetic data collected by the geomagnetic sensors symmetrically installed on the sidewalls of the tunnel cable trench are low-pass filtered according to the three axes to obtain filtered geomagnetic data corresponding to each axis in time sequence. The actual magnetic field strength corresponding to each of the geomagnetic sensors is determined by summing the squares of the filtered geomagnetic data corresponding to different axes at the same time and taking the arithmetic square root. Based on the sampling frequency of the geomagnetic sensor, a sampling interval corresponding to a preset multiple of the sampling frequency is determined, and the actual magnetic field strength corresponding to the current moment is used as the starting point to query the actual magnetic field strength within the sampling interval in reverse order as the target actual sampling intensity. Based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target, determine the disturbance magnetic field strength and the change in the disturbance magnetic field at the current moment.

6. The vehicle detection method based on lateral dual geomagnetic fusion according to claim 5, characterized in that, The step of determining the disturbance magnetic field strength and the change in disturbance magnetic field at the current moment based on the actual magnetic field strength at the current moment and the actual sampling intensity of the target includes: Calculate the sum of the actual magnetic field strength collected at the current moment and the corresponding actual sampling intensity of the target to obtain the magnetic field strength sum value, and calculate the ratio between the magnetic field strength sum value and the sampling interval length to obtain the mean magnetic field strength; Calculate the difference between the actual magnetic field strength collected at the current moment and the mean magnetic field strength, and calculate the square of the difference to obtain the magnetic field strength parameter. Then, sum the magnetic field strength parameter values ​​within the sampling interval and calculate the arithmetic square root of the summed magnetic field strength parameter value to obtain the stable magnetic field strength value. Based on whether the stable value of the magnetic field strength is less than a preset stability threshold, it is determined whether the actual magnetic field strength collected at the current moment meets the stability condition. If the actual magnetic field strength collected at the current moment meets the stability condition, the ratio between the magnetic field strength and the sampling interval length is used as the update value to update the reference magnetic field strength. If the intensity difference between the actual magnetic field strength collected at the current moment and the updated reference magnetic field strength meets the preset calculation conditions, the intensity difference will be used as the disturbance magnetic field strength corresponding to the current moment. The change in the perturbation magnetic field at the current moment is determined based on the perturbation magnetic field strength at the previous moment and the perturbation magnetic field strength at the current moment.

7. A vehicle detection device based on lateral dual geomagnetic fusion, characterized in that, The device includes: The first determining module is configured to determine the intensity and change of the disturbance magnetic field at the current moment based on the geomagnetic data collected by geomagnetic sensors symmetrically installed on the sidewall of the tunnel cable trench. Each of the geomagnetic sensors emits geomagnetic waves perpendicular to the sidewall of the tunnel cable trench towards the corresponding symmetrical geomagnetic sensor, and the geomagnetic sensor obtains the geomagnetic data based on the received geomagnetic waves. The second determining module is configured to add the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field strength at the current moment is greater than the disturbance magnetic field noise floor and the disturbance magnetic field change is greater than the disturbance change threshold, and to stop adding the disturbance magnetic field strength to the geomagnetic disturbance waveform list when the disturbance magnetic field change is less than the waveform stability threshold for a consecutive preset number of times. The third determining module is configured to determine the difference between the maximum and minimum disturbance magnetic field strengths based on the disturbance magnetic field strengths in the list of geomagnetic disturbance waveforms, and obtain the geomagnetic disturbance waveform range. The fourth determining module is configured to count the number of disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the pulse width of the geomagnetic disturbance waveform, and to count the number of maximum and minimum values ​​of the disturbance magnetic field intensities in the list of geomagnetic disturbance waveforms to obtain the number of extreme points. The fifth determining module is configured to determine the vehicle detection tag based on the relationship between the range of the geomagnetic disturbance waveform and multiple preset range thresholds, the pulse width of the geomagnetic disturbance waveform and the relationship between the number of extreme points and multiple preset number thresholds, and to perform tag fusion based on the vehicle detection tags corresponding to each geomagnetic sensor to determine the vehicle detection result. The fifth determining module is configured as follows: If the number of extreme points is less than the first preset number threshold among the preset number thresholds, or if the range of the geomagnetic disturbance waveform is less than or equal to the first preset range threshold among the preset range thresholds, the vehicle detection tag is determined to be no vehicle. If the number of extreme points is greater than the second preset number threshold among the preset number thresholds, and the range of the geomagnetic disturbance waveform is less than the second preset range threshold among the preset range thresholds, then the vehicle detection tag is determined to have a vehicle; if the range of the geomagnetic disturbance waveform is greater than or equal to the second preset range threshold, then the vehicle detection tag is determined to have a suspected vehicle. If the number of extreme points is greater than or equal to the first preset threshold and less than the second preset threshold, then the vehicle detection tag is determined based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds. The step of determining the vehicle detection tag based on the number of extreme points, the pulse width of the geomagnetic disturbance waveform, the range of the geomagnetic disturbance waveform, and multiple preset thresholds includes: The proportional parameter is determined based on the ratio between the range of the geomagnetic disturbance waveform and the pulse width of the geomagnetic disturbance waveform; The waveform characteristic value is determined by multiplying the proportional parameter value by the number of extreme points. The vehicle detection tag is determined based on the relationship between the waveform feature value and multiple set thresholds.

8. An electronic device, characterized in that, include: Processor, machine-readable storage medium; The machine-readable storage medium is connected to the processor, the machine-readable storage medium is used to store programs, instructions or code, and the processor is used to execute the programs, instructions or code in the machine-readable storage medium to implement the method of any one of claims 1-6.

Images