A geomagnetic differential vehicle detection method and system based on a single geomagnetic sensor

Abstract

The present application relates to a kind of geomagnetic differential vehicle detection method and system based on single geomagnetic sensor, mainly solve the stability of current environment baseline algorithm is low, poor adaptability and camera and radar and other sensor cost is high, is influenced by weather etc.Problems.The present application method will be geomagnetic sensor etc. equidistant deployment in single lane one side lane line, sensor sensing node carries out collection to geomagnetic data, calculates three-axis first-order difference signal and carries out filtering, through the calculation of the first-order difference signal corresponding to the probability of having car by bayesian probability, and then according to the probability of having car vehicle detection, the uncertainty of magnetic field data is considered, provide more reliable detection result.Finally, vehicle passing information is uploaded to intelligent traffic cloud platform through base station, realize the accurate, real-time monitoring of platform to road traffic information.

Description

Technical Field

[0001] This invention belongs to the field of intelligent transportation technology, specifically relating to a geomagnetic differential vehicle detection method and system based on a single geomagnetic sensor. Background Technology

[0002] Currently, the field of intelligent transportation is in a phase of rapid development, and the need for accurate, all-weather, all-time traffic information is becoming increasingly urgent in the context of intelligent highway transportation. Accurate traffic information can help traffic managers better understand vehicle movement on the road, providing valuable data for traffic flow prediction, traffic congestion warning, and traffic accident monitoring.

[0003] There are many methods for detecting traffic flow, with commonly used methods including camera monitoring, millimeter-wave radar detection, and infrared sensor detection. Currently, in single-lane scenarios, cameras and radar sensors are typically used to correlate vehicle trajectories. However, these sensors have specific deployment environments, usually requiring installation on gantry cranes, making large-scale deployment difficult, costly, and their detection accuracy affected by weather conditions.

[0004] A vehicle detection method based on coordinate transformation and matching of millimeter-wave radar and video data is proposed in the patent application "Vehicle Detection Method Based on Coordinate Transformation and Matching of Millimeter-Wave Radar and Video Data" (Application No.: CN202111351146.7, Publication No.: CN114280599A) filed by Zhou Wenqian et al. This method collects data by setting up millimeter-wave radar equipment and high-angle camera equipment. Through coordinate transformation and temporal-spatial data matching, the millimeter-wave radar data is fused with image data. The fused data is then used to train a vehicle detection model, improving the sensitivity and accuracy of the model. However, this method has drawbacks. The deployment environment for sensors such as cameras and radar is specific, typically mounted on gantry frames, making large-scale deployment difficult and costly. Furthermore, the detection accuracy is affected by weather conditions; in harsh environments, vehicle detection may be difficult, further impacting single-lane vehicle trajectory association.

[0005] J. Zhao et al., in their paper "Improved Vision-Based Vehicle Detection and Classification by Optimized YOLOv4" (IEEE Access, vol. 10, pp. 8590-8603), proposed a more accurate vehicle detection and classification model based on YOLOv4 with additional optimizations. The method involves increasing the receptive range in both channel and spatial dimensions by introducing an attention mechanism in the form of a CBAM module. Furthermore, in the FPN part, feature fusion is modified, and additional upsampling operations are performed. Then, the output features are fused again, and the detection results from different layers are combined to improve the detection performance of the proposed YOLOv4_AF model. The drawback of this method is that, by introducing a CBAM module on top of the YOLOv4 model, the computational complexity and time are increased, resulting in lower efficiency and making it unsuitable for real-time vehicle detection in single-lane systems used in intelligent highway systems. Summary of the Invention

[0006] To address the aforementioned problems in the existing technology, this invention provides a geomagnetic differential vehicle detection method and system based on a single geomagnetic sensor. The technical problem to be solved by this invention is achieved through the following technical solution:

[0007] This invention provides a geomagnetic differential vehicle detection method based on a single geomagnetic sensor, applicable to a single lane where multiple geomagnetic sensors are deployed, wherein the multiple geomagnetic sensors are deployed at intervals on one side of the single lane.

[0008] The geomagnetic differential vehicle detection method includes:

[0009] Step 1: Use the geomagnetic sensor to collect triaxial geomagnetic data at the corresponding location;

[0010] Step 2: Obtain the triaxial magnetic field strength difference value based on the triaxial geomagnetic data;

[0011] Step 3: Remove noise from the triaxial magnetic field strength difference values ​​to obtain a smooth triaxial magnetic field first-order difference signal;

[0012] Step 4: Calculate the probability of a vehicle passing by based on the first-order difference signal of the three-axis smooth magnetic field;

[0013] Step 5: Detect vehicles based on the probability of a vehicle passing by corresponding to the first-order difference of the three axes, and determine whether a vehicle has passed by the corresponding position of the geomagnetic sensor.

[0014] Step 6: When a vehicle passes by, obtain the vehicle's passing information and transmit the vehicle passing information to the base station around the single lane, so that the base station can transmit the vehicle passing information to the intelligent transportation cloud platform.

[0015] In one embodiment of the present invention, in step 2, the triaxial magnetic field strength difference value is calculated according to the following formula:

[0016] D i (k)=F i (k)-F i (k-1);

[0017] In the formula, F i (k) represents the i-axis geomagnetic data collected by the geomagnetic sensor at time k, F i (k-1) represents the i-axis geomagnetic data collected by the geomagnetic sensor at time k-1, where D is the geomagnetic data. i (k) represents the difference in magnetic field strength along the i-axis at the location of the geomagnetic sensor at time k, where i = X, Y, Z represents the X, Y, and Z axes.

[0018] In one embodiment of the present invention, step 3 includes:

[0019] The triaxial magnetic field strength difference values ​​are filtered using a low-pass filter to remove noise and obtain a smooth triaxial magnetic field first-order difference signal.

[0020] In one embodiment of the present invention, step 4 includes:

[0021] Step 4.1: Using the first-order difference signal of the three-axis smooth magnetic field as the probability calculation condition, calculate the probability of a vehicle passing by corresponding to the first-order difference of a single axis.

[0022] Step 4.2: Fuse the vehicle passing probabilities corresponding to the first-order difference of the single axle to obtain the vehicle passing probabilities corresponding to the first-order difference of the three axles.

[0023] In one embodiment of the present invention, in step 4.1, the probability of a vehicle passing by corresponding to the first-order difference of the single axle is calculated according to the following formula:

[0024]

[0025] In the formula, e(k) represents the event of whether a vehicle passes through the range of the geomagnetic sensor at time k, j∈{0,1}, e(k)=0 indicates that no vehicle passes through the detection range of the geomagnetic sensor, and e(k)=1 indicates that a vehicle passes through the detection range of the geomagnetic sensor. i (k) represents the first-order differential signal of the smoothed magnetic field along the i-axis at the location of the geomagnetic sensor at time k, where i = X, Y, Z represent the X, Y, and Z axes, respectively.i (x) is the probability density function of the first-order difference signal of the i-axis magnetic field, I i,0 express The corresponding integral interval of the probability of no vehicle passing through, I i,1 express The corresponding integral interval of the probability that a vehicle has passed through.

[0026] In one embodiment of the present invention, in step 4.2, the probability of a vehicle passing by corresponding to the first-order difference of the three axles is calculated according to the following formula:

[0027]

[0028] In the formula, C is a normalization constant, such that the sum of the calculated probabilities of having a vehicle and not having a vehicle corresponding to the first-order difference of the three axles is 1, and ρ c Let c ∈ {0,1} be the correlation coefficient, ρ0 represent the correlation coefficient when no car passes by, ρ1 represent the correlation coefficient when a car passes by, and P(e(k)=j) be the prior probability of the event corresponding to the value of e(k).

[0029] In one embodiment of the present invention, step 5 includes:

[0030] Step 5.1: In the car-free state, denote the probability of a car passing by corresponding to the first-order difference of the three axes at time k as P. car (k), compare P car (k) and reaching the detection threshold θ arr The size of P car (k)≥θ arr If the vehicle is detected, it will enter the detection state; otherwise, it will remain in the no-vehicle state.

[0031] Step 5.2: Upon reaching the detection state, compare N consecutive N... arr P at each moment within a time window car (k) and reaching the detection threshold θ arr The size, if N consecutive arr P at each moment within a time window car (k) all satisfy P car (k)≥θ arr If the vehicle arrives, the vehicle passes through the current state; otherwise, the system returns to the "no vehicle" state.

[0032] Step 5.3: When the vehicle has passed, compare P. car (k) and the departure detection threshold θ lea The size of P car (k)≤θ lea If the vehicle passes through, it enters the exit detection state; otherwise, it remains in the vehicle passing through state.

[0033] Step 5.4: After leaving the detection state, compare N consecutive N... lea P at each moment within a time window car (k) and the departure detection threshold θ lea The size, if N consecutive lea P at each moment within a time window car (k) all satisfy P car (k)≤θ lea If the vehicle has left and the system returns to a vehicle-free state, then repeat steps 5.1-5.4 to detect subsequent vehicles; otherwise, return to the vehicle-passed state.

[0034] In one embodiment of the present invention, in step 6, the vehicle passing information includes the vehicle arrival time and the vehicle passing time.

[0035] This invention also provides a geomagnetic differential vehicle detection system based on a single geomagnetic sensor, applicable to the geomagnetic differential vehicle detection method based on a single geomagnetic sensor described in any of the above embodiments. The geomagnetic differential vehicle detection system includes multiple sensing devices communicatively connected to a base station. These sensing devices are spaced apart and deployed on one side of the single lane. Each sensing device includes a geomagnetic sensor, a data processing unit, and a communication unit connected in sequence.

[0036] The geomagnetic sensor is used to collect triaxial geomagnetic data at the corresponding location;

[0037] The data processing module is used to obtain triaxial magnetic field strength difference values ​​based on the triaxial geomagnetic data; remove noise from the triaxial magnetic field strength difference values ​​to obtain a triaxial smoothed magnetic field first-order difference signal; calculate the probability of a vehicle passing by corresponding to the triaxial first-order difference based on the triaxial smoothed magnetic field first-order difference signal; detect vehicles based on the probability of a vehicle passing by corresponding to the triaxial first-order difference to determine whether a vehicle has passed by the corresponding location of the geomagnetic sensor; and acquire vehicle passing information when a vehicle passes by.

[0038] The communication unit is used to transmit the vehicle passage information to the base station, so that the base station can be used to transmit the vehicle passage information to the intelligent transportation cloud platform.

[0039] In one embodiment of the present invention, the deployment direction of the sensing device is consistent with the driving direction of the vehicle.

[0040] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0041] 1. The geomagnetic differential vehicle detection method based on a single geomagnetic sensor of the present invention uses the first-order differential signal of geomagnetic data collected by the geomagnetic sensor to detect vehicles, overcoming the influence of magnetic field drift when relying on the environmental magnetic field for vehicle detection in the prior art; in addition, since the first-order differential measures the relative change of the magnetic field between time points, vehicle detection can also be performed when a vehicle stops within the sensor detection range.

[0042] 2. The geomagnetic differential vehicle detection method based on a single geomagnetic sensor of the present invention detects vehicles based on the Bayesian probability of the presence of a vehicle in geomagnetic data, and determines whether a vehicle has passed by based on the probability value. It takes into account the uncertainty of the data, can better adapt to complex real-world scenarios, has good generalization ability, provides more reliable detection results, and avoids the problems of high false detection rate and poor adaptability caused by a single threshold.

[0043] 3. The geomagnetic differential vehicle detection method based on a single geomagnetic sensor of the present invention overcomes the problems of existing technologies that use cameras or radar sensors, which require installation on a gantry, are difficult to deploy on a large scale, have high costs, and whose detection accuracy is affected by weather. The present invention can be deployed on a large scale and can achieve single-lane vehicle detection at low cost and without the influence of the environment.

[0044] The above description is merely an overview of the technical solution of the present invention. In order to better understand the technical means of the present invention and to implement it in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the present invention more apparent and understandable, preferred embodiments are described in detail below with reference to the accompanying drawings. Attached Figure Description

[0045] Figure 1 This is a flowchart of a geomagnetic differential vehicle detection method based on a single geomagnetic sensor provided in an embodiment of the present invention;

[0046] Figure 2 This is a schematic diagram of the deployment of a geomagnetic sensor provided in an embodiment of the present invention;

[0047] Figure 3 This is a flowchart of a vehicle detection state machine provided in an embodiment of the present invention. Detailed Implementation

[0048] To further illustrate the technical means and effects adopted by the present invention to achieve the intended purpose, the following detailed description, in conjunction with the accompanying drawings and specific embodiments, provides a method and system for geomagnetic differential vehicle detection based on a single geomagnetic sensor proposed in accordance with the present invention.

[0049] The foregoing and other technical contents, features, and effects of the present invention will be clearly presented in the following detailed description of specific embodiments in conjunction with the accompanying drawings. Through the description of the specific embodiments, a more in-depth and concrete understanding can be gained of the technical means and effects adopted by the present invention to achieve its intended purpose. However, the accompanying drawings are for reference and illustration only and are not intended to limit the technical solutions of the present invention.

[0050] In a first aspect, embodiments of the present invention provide a geomagnetic differential vehicle detection method based on a single geomagnetic sensor, applicable to a single lane where multiple geomagnetic sensors are deployed. The multiple geomagnetic sensors are deployed at intervals on one side of the lane, and the deployment direction of the geomagnetic sensors is the same as the vehicle's travel direction.

[0051] Please see Figure 2 , Figure 2 This is a schematic diagram of the deployment of a geomagnetic sensor provided in an embodiment of the present invention. Figure 2 As shown, optionally, multiple geomagnetic sensors of the same model can be selected according to the lane range to be detected, such as the RM3100 model geomagnetic sensor. These multiple geomagnetic sensors are deployed at equal intervals of 8-15 meters on one side of the lane, for example, with an interval of 10 meters. The deployment direction of the geomagnetic sensors is the same as the vehicle's travel direction.

[0052] The geomagnetic differential vehicle detection method based on a single geomagnetic sensor of the present invention overcomes the problems of existing technologies that use cameras or radar sensors, which require installation on a gantry, are difficult to deploy on a large scale, have high costs, and whose detection accuracy is affected by weather. The method can be deployed on a large scale and can achieve single-lane vehicle detection at low cost and without being affected by the environment.

[0053] Please see Figure 1 , Figure 1 This is a flowchart of a geomagnetic differential vehicle detection method based on a single geomagnetic sensor provided in an embodiment of the present invention, as shown below. Figure 1 The geomagnetic differential vehicle detection method based on a single geomagnetic sensor shown in this embodiment includes:

[0054] Step 1: Use a geomagnetic sensor to collect triaxial geomagnetic data at the corresponding location.

[0055] Step 2: Obtain the triaxial magnetic field strength difference value based on the triaxial geomagnetic data.

[0056] In an optional embodiment, the triaxial magnetic field strength difference value is calculated according to the following formula:

[0057] D i (k)=Fi (k)-F i (k-1);

[0058] In the formula, F i (k) represents the i-axis geomagnetic data collected by the geomagnetic sensor at time k, F i (k-1) represents the i-axis geomagnetic data collected by the geomagnetic sensor at time k-1, where D is the geomagnetic data. i (k) represents the difference in magnetic field strength along the i-axis at the location of the geomagnetic sensor at time k, where i = X, Y, Z represents the X, Y, and Z axes.

[0059] Step 3: Remove noise from the triaxial magnetic field strength difference values ​​to obtain the triaxial smooth magnetic field first-order difference signal.

[0060] Optionally, a low-pass filter can be used to filter the triaxial magnetic field strength difference values ​​to remove noise and obtain a smooth triaxial magnetic field first-order difference signal.

[0061] For example, an FIR low-pass filter can be used, and the Kaiser window function method can be employed to filter the triaxial magnetic field strength difference values. The key parameters of the Kaiser window function method are the window length and the -6dB cutoff frequency; the filter performance is adjusted by modifying these two parameters. In this embodiment, the X and Y axis magnetic field differential filters use a -6dB cutoff frequency of 5Hz, and the Z axis magnetic field differential filter uses a -6dB cutoff frequency of 6Hz.

[0062] Step 4: Calculate the probability of a vehicle passing by based on the first-order difference signal of the three-axis smooth magnetic field.

[0063] In an optional embodiment, step 4 includes:

[0064] Step 4.1: Using the first-order difference signal of the three-axis smooth magnetic field as the probability calculation condition, calculate the probability of a vehicle passing by corresponding to the first-order difference of the single axis.

[0065] The probability of a vehicle passing by corresponding to the first-order difference of a single axle is calculated using the following formula:

[0066]

[0067] In the formula, e(k) represents the event of whether a vehicle passes through the range of the geomagnetic sensor at time k, j∈{0,1}, e(k)=0 indicates that no vehicle passes through the detection range of the geomagnetic sensor, and e(k)=1 indicates that a vehicle passes through the detection range of the geomagnetic sensor. Let f be the first-order differential signal of the smoothed magnetic field along the i-axis at the location corresponding to the geomagnetic sensor at time k, where i = X, Y, Z represent the X, Y, and Z axes, respectively. i (x) is the probability density function of the first-order difference signal of the i-axis magnetic field, Ii,0 express The corresponding integral interval of the probability of no vehicle passing through, I i,1 express The corresponding integral interval of the probability that a vehicle has passed through.

[0068] In this embodiment,

[0069] Step 4.2: Fuse the vehicle passing probabilities corresponding to the first-order difference of a single axle to obtain the vehicle passing probabilities corresponding to the first-order difference of a three-axle axle.

[0070] The probability of a vehicle passing by corresponding to the first-order difference of the three axles is calculated according to the following formula:

[0071]

[0072] In the formula, C is a normalization constant, such that the sum of the probabilities of having a vehicle and not having a vehicle corresponding to the calculated first-order difference of the three axles is... ρ c Let c ∈ {0,1} be the correlation coefficient, which is the correlation coefficient between the product of the first-order difference probabilities of the three axles when there are no vehicles and the joint probability of the first-order difference of the three axles when there are vehicles. Here, ρ0 represents the correlation coefficient when there are no vehicles, and ρ1 represents the correlation coefficient when there are vehicles. In this embodiment, ρ0 = 0.64 and ρ1 = 0.89. P(e(k) = j) is the prior probability of the event corresponding to the value of e(k). The value of P(e(k) = j) can be determined based on the historical traffic flow data of the road. In this embodiment, P(e(k) = 0) = 0.25 and P(e(k) = 1) = 0.75.

[0073] Step 5: Detect vehicles based on the probability of a vehicle passing by corresponding to the first-order difference of the three axes, and determine whether a vehicle has passed by the corresponding location of the geomagnetic sensor.

[0074] Please refer to the above. Figure 3 , Figure 3 This is a flowchart of a vehicle detection state machine provided in an embodiment of the present invention. Figure 3 As shown, in an optional embodiment, step 5 may include the following steps.

[0075] Step 5.1: In the car-free state, denote the probability of a car passing by corresponding to the first-order difference of the three axes at time k as P. car (k), compare P car (k) and reaching the detection threshold θ arr The size of P car (k)≥θ arr If the vehicle is detected, it will enter the detection state; otherwise, it will remain in the no-vehicle state.

[0076] Step 5.2: Upon reaching the detection state, compare N consecutive N... arr P at each moment within a time window car (k) and reaching the detection threshold θ arr The size, if N consecutive arr P at each moment within a time window car (k) all satisfy P car (k)≥θ arr If the vehicle arrives, the vehicle will be confirmed to have passed through the current state; otherwise, the system will return to the "no vehicle" state.

[0077] Step 5.3: When the vehicle has passed, compare P. car (k) and the departure detection threshold θ lea The size of P car (k)≤θ lea If the vehicle passes through, it enters the exit detection state; otherwise, it remains in the vehicle passing through state.

[0078] Step 5.4: After leaving the detection state, compare N consecutive N... lea P at each moment within a time window car (k) and the departure detection threshold θ lea The size, if N consecutive lea P at each moment within a time window car (k) all satisfy P car (k)≤θ lea If the vehicle has left and the system returns to a vehicle-free state, then repeat steps 5.1-5.4 to detect subsequent vehicles; otherwise, return to the vehicle-passed state.

[0079] It should be noted that after the detection system is initialized, it first enters a vehicle-free state, and then calculates the probability of a vehicle passing by based on the first-order difference values ​​of the three axes. And observe it. In this embodiment, the detection threshold θ is reached. arr The value is 0.9, N arr The value is 10, which is far from the detection threshold θ. lea The value is 0.3, N lea The value is 10.

[0080] In this embodiment, when a vehicle completes four different state transitions, it indicates that a vehicle has passed within the detection range of the geomagnetic sensor at the current location. Each geomagnetic sensor records the arrival and departure times of each passing vehicle while detecting it, for the purpose of statistical analysis of vehicle passage information.

[0081] Step 6: When a vehicle passes by, obtain the vehicle's passing information and transmit the vehicle passing information to the base station around the single lane, so that the base station can transmit the vehicle passing information to the intelligent transportation cloud platform.

[0082] In this embodiment, the vehicle passage information includes the vehicle arrival time and the vehicle passage time. For example, the transition time from the arrival detection state to the vehicle passage state can be recorded as the arrival time of the passing vehicle, and the transition time from the departure detection state to the vehicle-free state can be recorded as the departure time of the passing vehicle. The vehicle passage time can be calculated using the recorded arrival and departure times of the passing vehicles.

[0083] The geomagnetic differential vehicle detection method based on a single geomagnetic sensor in this invention utilizes the first-order differential signal of geomagnetic data collected by the geomagnetic sensor for vehicle detection, overcoming the influence of magnetic field drift when relying on the ambient magnetic field for vehicle detection in existing technologies. Furthermore, since the first-order difference measures the relative change of the magnetic field between different times, vehicle detection can be performed even when a vehicle is parked within the sensor's detection range. Moreover, vehicle detection is based on the Bayesian probability of a vehicle's presence using geomagnetic data, and the probability value is used to determine whether a vehicle has passed. This takes into account the uncertainty of the data, allowing for better adaptation to complex real-world scenarios, better generalization ability, and more reliable detection results. It avoids the problems of high false detection rates and poor adaptability caused by single threshold methods.

[0084] Secondly, embodiments of the present invention provide a geomagnetic differential vehicle detection system based on a single geomagnetic sensor, applicable to the geomagnetic differential vehicle detection method based on a single geomagnetic sensor provided in the first aspect. The geomagnetic differential vehicle detection system includes multiple sensing devices that are communicatively connected to a base station. The multiple sensing devices are deployed at intervals on one side of a single lane. In this embodiment, the deployment direction of the sensing devices is consistent with the driving direction of the vehicle.

[0085] In this embodiment, each sensing device includes a geomagnetic sensor, a data processing unit, and a communication unit connected in sequence. The geomagnetic sensor collects triaxial geomagnetic data at a corresponding location. The data processing module obtains triaxial magnetic field strength difference values ​​from the triaxial geomagnetic data; removes noise from the triaxial magnetic field strength difference values ​​to obtain a smoothed first-order differential signal; calculates the probability of a vehicle passing corresponding to the first-order differential signal; detects vehicles based on the probability of a vehicle passing, determining whether a vehicle has passed at the corresponding location of the geomagnetic sensor; and acquires vehicle passing information when a vehicle passes. The communication unit transmits the vehicle passing information to a base station, which then transmits the vehicle passing information to the intelligent transportation cloud platform.

[0086] In an optional embodiment, the communication unit can utilize 4G communication technology to establish a communication connection between each sensing device and a base station around the single lane.

[0087] For details regarding the geomagnetic differential vehicle detection system based on a single geomagnetic sensor and its corresponding beneficial effects, please refer to the relevant content of the geomagnetic differential vehicle detection method based on a single geomagnetic sensor provided in the first aspect, which will not be repeated here.

[0088] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations are intended to cover non-exclusive inclusion, such that an article or apparatus comprising a list of elements includes not only those elements but also other elements not expressly listed. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the article or apparatus that includes said element. Terms such as "connected" or "linked" are not limited to physical or mechanical connections but can include electrical connections, whether direct or indirect.

[0089] The above description, in conjunction with specific preferred embodiments, provides a further detailed explanation of the present invention. It should not be construed that the specific implementation of the present invention is limited to these descriptions. For those skilled in the art, various simple deductions or substitutions can be made without departing from the concept of the present invention, and all such modifications and substitutions should be considered within the scope of protection of the present invention.

Claims

1. A geomagnetic differential vehicle detection method based on a single geomagnetic sensor, characterized by, The application is suitable for a single lane with multiple geomagnetic sensors arranged on a side lane line of the single lane; The geomagnetic differential vehicle detection method comprises: Step 1: collecting three-axis geomagnetic data of a corresponding position by using the geomagnetic sensor; Step 2: obtaining a three-axis magnetic field intensity difference value according to the three-axis geomagnetic data; Step 3: removing noise of the three-axis magnetic field intensity difference value to obtain a three-axis smooth magnetic field first-order difference signal; Step 4: calculating a three-axis first-order difference corresponding vehicle passing probability according to the three-axis smooth magnetic field first-order difference signal; the step 4 comprises: Step 4.1: taking the three-axis smooth magnetic field first-order difference signal as a probability calculation condition respectively to calculate a single-axis first-order difference corresponding vehicle passing probability; the single-axis first-order difference corresponding vehicle passing probability is calculated according to the following formula: ； wherein is an event of whether a vehicle passes through the detection range of the geomagnetic sensor at a time point, , indicates no vehicle passing through the detection range of the geomagnetic sensor, indicates a vehicle passing through the detection range of the geomagnetic sensor, is a magnetic field signal of the geomagnetic sensor at a position corresponding to the time point, a first-order differential signal of the magnetic field signal of the geomagnetic sensor, indicates , and axes, is a probability density function of the first-order differential signal of the magnetic field signal of the geomagnetic sensor, indicates a corresponding probability integration interval of no vehicle passing, indicates a corresponding probability integration interval of a vehicle passing. Step 4.2: fusing the single-axis first-order difference corresponding vehicle passing probability to obtain the three-axis first-order difference corresponding vehicle passing probability; the three-axis first-order difference corresponding vehicle passing probability is calculated according to the following formula: ； wherein, is a normalization constant such that the sum of the calculated probabilities of having a vehicle and not having a vehicle corresponding to the triaxial first-order difference is 1, is a correlation coefficient, , represents the correlation coefficient when no vehicle passes by, represents the correlation coefficient when a vehicle passes by, is the prior probability of the corresponding event; Step 5: detecting a vehicle according to the three-axis first-order difference corresponding vehicle passing probability to determine whether a vehicle passes through the corresponding position of the geomagnetic sensor; Step 6: acquiring vehicle passing information of the vehicle when the vehicle passes through and transmitting the vehicle passing information to a base station around the single lane to transmit the vehicle passing information to an intelligent traffic cloud platform by using the base station.

2. The single geomagnetic sensor-based geomagnetic differential vehicle detection method according to claim 1, characterized by, In the step 2, the three-axis magnetic field intensity difference value is calculated according to the following formula: ； In the formula, is the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, is the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, is the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, the axis geomagnetic data of the corresponding position collected by the geomagnetic sensor at the moment, represents , and axis.

3. The single geomagnetic sensor-based geomagnetic differential vehicle detection method according to claim 1, characterized by, The step 3 comprises: filtering the three-axis magnetic field intensity difference value by using a low-pass filter to remove noise and obtain a three-axis smooth magnetic field first-order difference signal.

4. The single geomagnetic sensor-based geomagnetic differential vehicle detection method according to claim 1, characterized by, The step 5 comprises: Step 5.1: In the no-car state, if the three-axis first-order difference of the moment is greater than the detection threshold , the car passes through the probability is recorded as , and the size of is compared with the arrival detection threshold , if , the arrival detection state is entered, otherwise it remains in the no-car state; Step 5.2: When reaching the detection state, compare the size of each time window with the arrival detection threshold in succession If each time window in succession satisfies , then determine that the vehicle has arrived and enters the vehicle passing state, otherwise jump back to the no vehicle state; Step 5.3: In the vehicle passing state, compare the size of the distance to the exit detection threshold If then enter the exit detection state, otherwise remain in the vehicle passing state; Step 5.4: In the leaving detection state, compare the size of each time window with the leaving detection threshold in the continuous time windows, if each time window in the continuous time windows meets , determine that the vehicle leaves and jump to the no vehicle state and repeat steps 5.1-5.4 for the subsequent vehicle detection, otherwise jump back to the vehicle passing state.

5. The single geomagnetic sensor-based geomagnetic differential vehicle detection method according to claim 1, characterized in that, In the step 6, the vehicle passing information comprises a vehicle arrival time and a vehicle passing time.

6. A geomagnetic differential vehicle detection system based on a single geomagnetic sensor, characterized in that, The geomagnetic differential vehicle detection system is suitable for the geomagnetic differential vehicle detection method based on a single geomagnetic sensor according to any one of claims 1-5, and comprises multiple sensing devices in communication connection with a base station, the multiple sensing devices are arranged on a side lane line of the single lane at intervals, each sensing device comprises a geomagnetic sensor, a data processing unit and a communication unit connected in sequence, wherein The geomagnetic sensor is used to collect three-axis geomagnetic data of a corresponding position; The data processing unit is used to obtain a three-axis magnetic field intensity difference value according to the three-axis geomagnetic data, remove noise of the three-axis magnetic field intensity difference value to obtain a three-axis smooth magnetic field first-order difference signal, calculate a three-axis first-order difference corresponding vehicle passing probability according to the three-axis smooth magnetic field first-order difference signal, detect a vehicle according to the three-axis first-order difference corresponding vehicle passing probability to determine whether a vehicle passes through the corresponding position of the geomagnetic sensor, and acquire vehicle passing information of the vehicle when the vehicle passes through; The communication unit is used to transmit the vehicle passing information to the base station to transmit the vehicle passing information to an intelligent traffic cloud platform by using the base station.

7. The single geomagnetic sensor based geomagnetic differential vehicle detection system according to claim 6, characterized in that, The sensing device is arranged in the same direction as the driving direction of the vehicle.

Images