Vehicle identification method and system, storage medium and electronic device

Abstract

The invention discloses a vehicle identification method and system, a storage medium and an electronic device.The method comprises the steps that vehicle detection data, detected by a road surface sensor, of a target vehicle are obtained, and the vehicle detection data comprise data generated when each axle in a set of axles of the target vehicle rolls the road surface sensor; vehicle point cloud data obtained when a vehicle scanning component scans the target vehicle are obtained, and the vehicle point cloud data are used for representing the vehicle outer contour of the target vehicle; and performing axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

Description

Technical Field

[0001] The present application relates to the field of vehicle identification, and in particular, to a vehicle identification method and system, a storage medium and an electronic device. Background Art

[0002] With the rapid development of intelligent transportation technology, vehicle axle type recognition has become one of the focuses of current research. Vehicle axle type recognition technology classifies and identifies different types of vehicles by identifying the number of axles, wheelbase, single or double tires and other characteristic information of the vehicle.

[0003] However, the existing technology mainly uses image processing and other methods to identify the vehicle axle type based on the vehicle's contour information or the vehicle's side image, which is easily affected by external environmental issues (such as light, color, posture, occlusion, etc.).

[0004] It can be seen that the vehicle identification method in the related art has the problem of low accuracy in identifying the vehicle axle type. Summary of the invention

[0005] Embodiments of the present application provide a vehicle identification method and system, a storage medium, and an electronic device to at least solve the problem of low accuracy in identifying vehicle axle types in the vehicle identification method in the related art.

[0006] According to one aspect of an embodiment of the present application, a vehicle identification method is provided, comprising: obtaining vehicle detection data of a target vehicle detected by a road surface sensor, wherein the vehicle detection data includes data generated when each axle of a group of axles of the target vehicle rolls over the road surface sensor; obtaining vehicle point cloud data obtained by a vehicle scanning component when scanning the target vehicle, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; and performing axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

[0007] According to another aspect of an embodiment of the present application, a vehicle identification system is also provided, including: a road surface sensor, used to perform weighing detection on a passing target vehicle to obtain vehicle detection data of the target vehicle, wherein the vehicle detection data includes weighing data generated by each axle of a group of axles of the target vehicle rolling over the road surface sensor; a vehicle scanning component, used to scan the target vehicle to obtain vehicle point cloud data, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; a data processing component, used to obtain the vehicle detection data; obtain the vehicle point cloud data; perform axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain the target axle type information of the target vehicle.

[0008] According to another aspect of the embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned vehicle identification method when running.

[0009] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the vehicle identification method through the computer program.

[0010] In an embodiment of the present application, a method is adopted in which the data detected by a road surface sensor and the data scanned by a vehicle scanning component are combined to determine the axle type information of the vehicle. Since the data detected by the road surface sensor is less affected by the external environment, the data generated when the vehicle rolls over the road surface sensor can be detected more accurately. The scanning data corresponding to the target vehicle in the scanning data of the vehicle scanning component is determined based on the data detected by the road surface sensor. The axle type information of the vehicle is determined based on at least one of the data detected by the road surface sensor and the scanning data corresponding to the target vehicle scanned by the vehicle scanning component. This can reduce the influence of the external environment and achieve the technical effect of improving the accuracy of vehicle axle type recognition, thereby solving the problem of low accuracy of vehicle axle type recognition in the vehicle recognition method in the related art. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present application and, together with the description, serve to explain the principles of the present application.

[0012] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, for ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0013] Figure 1 is a schematic diagram of a hardware environment of an optional vehicle identification method according to an embodiment of the present application;

[0014] Figure 2 is a flow chart of an optional vehicle identification method according to an embodiment of the present application;

[0015] Figure 3 is a schematic diagram of an optional vehicle identification method according to an embodiment of the present application;

[0016] Figure 4 is a schematic diagram of another optional vehicle identification method according to an embodiment of the present application;

[0017] Figure 5 is a schematic diagram of another optional vehicle identification method according to an embodiment of the present application;

[0018] Figure 6 is a schematic diagram of another optional vehicle identification method according to an embodiment of the present application;

[0019] Figure 7 is a schematic diagram of another optional vehicle identification method according to an embodiment of the present application;

[0020] Figure 8 It is a structural block diagram of an optional electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0021] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present application.

[0022] It should be noted that the terms "first", "second", etc. in the specification and claims of the present application and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchangeable where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any of their variations are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device comprising a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0023] According to one aspect of an embodiment of the present application, a vehicle identification method is provided. Optionally, in this embodiment, the vehicle identification method can be applied to Figure 1 In the hardware environment including the detection component 102 and the server 104 shown. Figure 1 As shown, the server 104 is connected to the detection component 102 via a network, and can be used to identify the vehicle axle type based on the detection data of the detection component 102. A database can be set on the server or independently of the server to provide data storage services for the server 104.

[0024] The above network may include but is not limited to at least one of the following: wired network, wireless network. The above wired network may include but is not limited to at least one of the following: wide area network, metropolitan area network, local area network, and the above wireless network may include but is not limited to at least one of the following: WIFI (Wireless Fidelity), Bluetooth. The detection component 102 may include a road surface sensor and a vehicle scanning component, wherein the vehicle scanning component may be but is not limited to a laser sensor.

[0025] The vehicle identification method of the embodiment of the present application can be executed by the server 104, can be executed by the detection component 102, or can be executed by the server 104 and the detection component 102 together. Taking the vehicle identification method of the embodiment executed by the server 104 as an example, Figure 2 is a flow chart of an optional vehicle identification method according to an embodiment of the present application, such as Figure 2 As shown, the process of the method may include the following steps:

[0026] Step S202, obtaining vehicle detection data of the target vehicle detected by the road surface sensor, wherein the vehicle detection data includes data generated when each axle in a group of axles of the target vehicle rolls over the road surface sensor.

[0027] The vehicle identification method in this embodiment can be applied to the scene of classifying and identifying different vehicles. The vehicle axle type may include but is not limited to the number of axles, wheelbase and wheelbase ratio, and single-double tires and other features of the vehicle. By identifying the vehicle axle type, different types of vehicles can be classified and identified. Generally speaking, vehicles with more axles, larger wheelbases, and double-axle axles are large vehicles, while vehicles with fewer axles or smaller wheelbases or single-axle axles are generally small vehicles.

[0028] Vehicle axle type recognition technology is mainly based on the vehicle's contour information or vehicle side image. By identifying the number of axles, wheelbase and wheelbase ratio of the vehicle, different types of vehicles can be classified and identified. Vehicle axle type recognition technology has a wide range of applications in the field of intelligent transportation, such as intelligent traffic monitoring, traffic congestion analysis, road safety management, etc. Especially in traffic management, vehicle axle type recognition can quickly and accurately identify the vehicle's axle type information for real-time vehicle supervision and decision-making, improve traffic safety levels, and ensure smooth and orderly urban traffic. The realization of vehicle axle type recognition is inseparable from vehicle detection, contour extraction and recognition technologies. Through vehicle detection and contour extraction, the vehicle's contour information can be obtained, and then the vehicle's axle number, wheelbase, single and double tires and other feature information can be extracted.

[0029] At present, there are three main methods for vehicle axle type recognition: image recognition, laser point cloud algorithm and ground sensor-based algorithm. The image recognition-based method mainly realizes automatic recognition of vehicle axle type based on deep learning model by modeling and analyzing vehicle contour and side image. First, a large number of vehicle images are collected and processed to obtain relatively stable images. Then, the model is trained based on sample images. During the training process, for any input vehicle image, the image features in various situations are learned by integrating the rotation, translation and scale scaling of the training image. It is continuously iterated until the model converges, so that it can recognize information such as the number of axles and wheelbase. However, this method has poor robustness to occlusion, illumination changes and posture changes.

[0030] The above-mentioned laser point cloud algorithm usually refers to the use of a laser radar device to scan the surface contour of the vehicle to obtain point cloud data. Based on the point cloud data, clustering or deep learning recognition algorithms can be used to obtain the axle information and tire type information of the scanned vehicle, and then the vehicle axle type can be determined based on the obtained axle information and tire type information. However, this method is easily affected by external environments such as occlusion and weather, and the position is inaccurate when the algorithm is spliced.

[0031] The above-mentioned vehicle axle type recognition system based on ground sensors is mainly composed of contact sensors and central processors. When the vehicle runs over the sensor, the sensor converts the force or deformation of the sensor into an electrical signal. Each pair of axles passing by will generate a pulse of an electrical signal. The number of axles of the vehicle is calculated by counting the number of pulses, and the axle spacing is calculated by combining the vehicle's speed and the distance between sensors. The system can detect the number of axles, wheelbase and speed of the vehicle, has a data interface power supply and data input and output safety isolation design, and has strong anti-interference ability, but the wheelbase recognition is not accurate enough, and it cannot accurately identify single and double tires.

[0032] In order to at least solve some of the above-mentioned problems, in this embodiment, the data detected by the road surface sensor and the data scanned by the vehicle scanning component can be combined. The road surface sensor can be a weighing sensor. The time when the vehicle passes through the road surface sensor and the weighing data of the vehicle can be determined based on the data detected by the road surface sensor, and the contour information of the vehicle can be determined based on the data scanned by the vehicle scanning component.

[0033] In this embodiment, vehicle detection data of a target vehicle detected by a road surface sensor may be obtained. Here, the road surface sensor includes but is not limited to a narrow strip type weighing sensor, a flat plate type weighing sensor, and a deep buried strain sensor. The target vehicle may be a vehicle of an axle type to be identified, and the vehicle detection data may include weighing data generated by each axle in a group of axles of the target vehicle rolling over the weighing sensor, and the weighing data may include information such as pressure and time.

[0034] When the target vehicle passes through the weighing area, it first passes through the upper weighing coil and then passes through the weighing sensor. When the axle rolls over the weighing sensor, part of the wheel contacts the sensor surface and transmits gravity to the weighing sensor through the force transmission structure. The weighing sensor usually has multiple rows (three rows are used as an example in this embodiment), which are installed between the carrier and the road surface (foundation). The weight of the object to be weighed is converted into an electrical signal proportional to the weight through the force-electric conversion components and the pressure transmitted from the weighing platform through the detection circuit inside the weighing sensor. At this time, the waveform signal of the vehicle weight and the upper and lower weighing time of each axle of the vehicle are obtained. The vehicle then passes through the lower weighing coil, and the time when the vehicle leaves the weighing area is recorded.

[0035] Step S204, obtaining vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle.

[0036] In this embodiment, for the target vehicle, the vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle can also be obtained. Here, the vehicle scanning component can be the aforementioned laser sensor, which can include but is not limited to a single-line laser radar and a road test multi-line laser radar. Figure 1 As shown, the vehicle scanning components can be installed above the lane and on both sides of the lane respectively. The vehicle point cloud data can be used to represent the vehicle outer contour of the target vehicle.

[0037] Laser sensors can detect the surrounding environment by emitting laser beams and receiving reflected signals. When scanning a vehicle, the laser sensor usually emits a series of laser beams continuously. When the laser beam intersects the surface of the vehicle, it will be reflected back and received by the laser sensor. By recording the reflection time and intensity of each laser beam, a series of point cloud data of the vehicle surface can be obtained. By calculating the product of the reflection time and the laser light speed, the distance from the laser beam from the sensor to the vehicle surface can be obtained. In addition, these point cloud data can be combined into a vehicle profile in three-dimensional space so that the shape and size information of the vehicle can be obtained. Due to the limited scanning range and resolution of laser sensors, it is usually necessary to use multiple laser sensors to scan the vehicle together to obtain more complete and accurate profile data. In this case, it is also necessary to synchronize the settings of each laser sensor.

[0038] Step S206, performing axle type recognition on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

[0039] Combining the acquired vehicle detection data and vehicle point cloud data, the target vehicle can be identified by axle type, and the target axle type information of the target vehicle can be obtained. The target axle type information here can include the number of axles, wheelbase, wheelbase, single and double tires and other information of the target vehicle.

[0040] For example, the vehicle scanning component is a laser sensor and the road surface sensor is a narrow strip sensor. Figure 3 As shown in the figure, during the data collection stage, the ground coil monitors in real time whether the vehicle enters the monitoring area. When different axles of the vehicle run over the narrow strip sensor, the axle position and running time are fed back to the central processing unit in real time. At the same time, according to the narrow strip trigger moment, the point cloud data collected by the laser sensor in this time period is intercepted and spliced ​​in the time dimension, and then the two data sources are synchronized and stored. After that, the time-synchronized laser point cloud data is spliced ​​or registered to form the complete vehicle outer contour. Finally, the designed algorithm is used to determine the number of vehicle axles, wheelbase, and single and double tires, and the vehicle axle type recognition result is output.

[0041] The position and time data of the vehicle passing by are obtained through the road surface sensor, and the outer contour of the vehicle is obtained through the laser sensor. By preprocessing and calculating these data, the axle type of the vehicle can be accurately identified and the real-time vehicle information can be provided to the user. Its accuracy is higher and it is not affected by light and color. It can effectively solve the problem that the current axle type recognition is easily affected by light, vehicle posture, etc., resulting in a lower recognition rate, and has a wider application prospect.

[0042] Optionally, after obtaining the weighing time and weighing time of each narrow strip sensor of each axle of the vehicle passing through the weighing sensor, as well as the scanning time of each frame of laser point cloud data (i.e., the aforementioned vehicle point cloud data), due to the different sampling frequencies of the weighing sensor and the vehicle scanning component, for the data at a specific moment, the interpolation method can be used to "fill in frames", and then time synchronization matching and filtering are performed based on the laser point cloud data and narrow strip waveform data (i.e., vehicle detection data) in the time period before and after the vehicle runs over the narrow strip, so as to provide data support for subsequent recognition algorithms.

[0043] Through the above steps S202 to S206, the vehicle detection data of the target vehicle detected by the road surface sensor is obtained, wherein the vehicle detection data includes data generated by each axle of a group of axles of the target vehicle rolling over the road surface sensor; the vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle is obtained, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; the axle type of the target vehicle is identified based on the vehicle detection data and the vehicle point cloud data to obtain the target axle type information of the target vehicle, thereby solving the problem of low accuracy of vehicle axle type identification in the vehicle identification method in the related art and improving the accuracy of vehicle axle type identification.

[0044] In an exemplary embodiment, the target vehicle is identified by axle type based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle, including:

[0045] S11, taking each axle as the current axle in turn and performing the following distance determination operations to obtain a reference distance corresponding to each axle: determining a target rolling time corresponding to the current axle based on vehicle detection data, wherein the target rolling time is the time when the current axle rolls over the road surface sensor; acquiring point cloud data of a target vehicle position matching the target rolling time from vehicle point cloud data to obtain reference point cloud data corresponding to each axle; determining the distance between the target vehicle position at the target rolling time and the vehicle scanning component along the driving direction of the target vehicle based on the reference point cloud data to obtain a reference distance corresponding to the current axle;

[0046] S12, determining the distance difference between the reference distances corresponding to two adjacent axles in a group of axles as the wheelbase between the two adjacent axles, and obtaining the wheelbase information of the target vehicle, wherein the target axle type information includes the wheelbase information of the target vehicle.

[0047] Taking into account that the positions of the road surface sensor and the vehicle scanning component are fixed during the driving process of the target vehicle, in this embodiment, the time when different axles of the target vehicle pass through the road surface sensor can be determined based on the vehicle detection data, and then the point cloud data of the vehicle scanning component scanning the target vehicle position within this time can be determined from the vehicle point cloud data, and then the distance between different axles of the vehicle can be determined.

[0048] In this embodiment, each axle can be used as the current axle to perform a distance determination operation in turn to obtain a reference distance corresponding to each axle. The distance determination operation may include: determining the target rolling time corresponding to the current axle based on the vehicle detection data, and then obtaining the point cloud data of the target vehicle position matching the target rolling time from the vehicle point cloud data to obtain the reference point cloud data corresponding to each axle, and determining the distance between the target vehicle position and the vehicle scanning component along the driving direction of the target vehicle at the target rolling time according to the reference point cloud data to obtain the reference distance corresponding to the current axle. Here, the target rolling time may be the time when the current axle rolls over the road surface sensor. The target vehicle position may be the front of the target vehicle, the rear of the target vehicle, or other positions of the target vehicle.

[0049] The above-mentioned point cloud data of the target vehicle position matching the target rolling time may be the point cloud data of the target vehicle position at the target rolling time obtained after frame supplementation according to the aforementioned "interpolation method", or it may be the point cloud data of the target vehicle position that is closest to the target rolling time.

[0050] After obtaining the reference distance corresponding to each axle, the distance difference between the reference distances corresponding to two adjacent axles in a group of axles can be determined as the wheelbase between the two adjacent axles to obtain the wheelbase information of the target vehicle. Here, the distance difference between the reference distances corresponding to the two adjacent axles is the distance difference between the reference distance corresponding to the first axle and the reference distance corresponding to the second axle in the two adjacent axles. Correspondingly, the target axle type information can include the wheelbase information of the target vehicle.

[0051] For example, the target vehicle position is the front of the target vehicle, the road surface sensor is a narrow strip sensor, and the vehicle scanning component is a laser sensor. Figure 4 As shown, when the first axle of the target vehicle rolls over the narrow strip sensor ZT1, the distance X1 between the vehicle head and the laser sensor is determined based on the reference point cloud data in the corresponding time. When the second axle of the target vehicle rolls over the narrow strip sensor ZT1, the distance X2 between the vehicle head and the laser sensor is determined based on the reference point cloud data in the corresponding time. The distance between the first axle and the second axle of the target vehicle can be determined based on the difference between X2 and X1.

[0052] Optionally, the distances between the target vehicle position and the axles of the target vehicle may be determined in sequence, and the distances between different axles may be determined based on the distances between the target vehicle position and different axles. Figure 5 As shown, taking the target vehicle position as the front of the target vehicle, the road surface sensor as the narrow strip sensor, and the vehicle scanning component as the laser sensor as an example, when the first axle of the target vehicle rolls over the narrow strip sensor ZT1, the distance X1 between the front of the vehicle and the laser sensor and the distance X3 between the first axle and the laser sensor are determined according to the reference point cloud data in the corresponding time, and the distance ΔX1 between the front of the vehicle and the first axle is determined according to the difference between X1 and X3. Similarly, the distance ΔX2 between the front of the vehicle and the first axle is determined, and the distance between the first axle and the second axle of the target vehicle can be determined according to the difference between ΔX1 and ΔX2.

[0053] Through this embodiment, based on the time when the axle rolls over the road surface sensor detected by the road surface sensor, the point cloud data scanned by the vehicle scanning component within the corresponding time is determined, and then the wheelbase of the target vehicle is calculated. This can improve the accuracy of the point cloud data used in the calculation process, and thus determine the accuracy of the wheelbase.

[0054] In an exemplary embodiment, determining the distance between the target vehicle position and the vehicle scanning component along the driving direction of the target vehicle at the target rolling time according to the reference point cloud data to obtain a reference distance corresponding to the current axle includes:

[0055] S21, determining the relative distance between the target vehicle position and the vehicle scanning component at the target rolling time and the relative angle between the target vehicle position and the vehicle scanning component at the target rolling time according to the reference point cloud data;

[0056] S22, based on the relative angle, convert the relative distance into a distance along the driving direction to obtain a reference distance corresponding to the current axle.

[0057] When determining the reference distance corresponding to the current axle, the calculation can be performed based on the relative distance and relative angle between the vehicle scanning component and the target vehicle position. Considering that the two-point line between the vehicle scanning component and the target vehicle position may not be parallel to the driving direction of the target vehicle (i.e., the two-point line is not on a plane parallel to the driving direction of the target vehicle and perpendicular to the road plane), when calculating the reference distance, the determined relative angle and relative distance can be converted into an angle and distance on a plane parallel to the driving direction of the target vehicle and perpendicular to the road plane, i.e., the relative angle is converted into the target angle, and the relative distance is converted into the target distance.

[0058] Converting the relative distance into the distance along the driving direction based on the relative angle may mean first converting the relative angle into a target angle, converting the relative distance into a target distance, and then converting the target distance into the distance along the driving direction according to the target angle, thereby obtaining a reference distance corresponding to the current axle.

[0059] For example, the target vehicle position is the front of the target vehicle, the road surface sensor is a narrow strip sensor, and the vehicle scanning component is a laser sensor. Figure 6 As shown, according to the distance L1 (ie, target distance) and angle θ1 (ie, target angle) from the laser sensor to the front of the vehicle, as shown in formula (1), the distance X1 from the laser sensor to the front of the vehicle along the driving direction can be determined.

[0060] X1 = L1 * Sinθ1 (1)

[0061] Optionally, when calculating the distance between the target vehicle position and the axle, the above method can also be used. When the first axle of the vehicle rolls over the narrow strip sensor ZT1, since the distance L2 and the angle θ2 between the laser sensor and the narrow strip sensor are known, the lateral distance X3 between the laser sensor and the first axle can be calculated as shown in formula (2):

[0062] X3= L2 * Sinθ2 (2)

[0063] According to formulas (1) and (2), the distance ΔX1 between the front of the vehicle and the first axis can be obtained, as shown in formula (3):

[0064] ΔX1 = X3- X1 (3)

[0065] Similarly, when the second axle rolls over the narrow strip sensor ZT1, the distance ΔX2 between the vehicle head and the second axle can be obtained by the above method, as shown in formula (4):

[0066] ΔX2 = X4 - X1 (4)

[0067] It should be noted that since the relative position has changed due to the operation of the vehicle, the value of X1 in formula (4) is different from that in formula (3), but the physical meaning remains unchanged.

[0068] According to formulas (3) and (4), the distance ΔX3 between the first axis and the second axis can be calculated, as shown in formula (5).

[0069] ΔX3 = ΔX2 - ΔX1 (5)

[0070] After calculating the relative distance from each axle to the front of the vehicle according to the above method, the spacing between two adjacent axles can be further obtained.

[0071] In addition, when there are three rows of narrow strip sensors, each row of narrow strip sensors can be used to measure the wheelbase, the above-mentioned single wheelbase can be measured three times, and then the average value is taken based on the three calculation results, thereby improving the measurement accuracy.

[0072] Through this embodiment, the reference distance for calculating the wheelbase is determined according to the angle and distance between the vehicle scanning component and the target vehicle position, so that the efficiency of wheelbase measurement can be improved.

[0073] In an exemplary embodiment, based on the relative angle, converting the relative distance into the distance along the driving direction to obtain the reference distance corresponding to the current axle includes:

[0074] S31, when there are multiple relative distances, convert each relative distance into a distance along the driving direction based on a relative angle corresponding to each relative distance, to obtain multiple candidate distances;

[0075] S32, clustering the multiple candidate distances to obtain multiple distance clusters, wherein each distance cluster in the multiple distance clusters includes part of the multiple candidate distances;

[0076] S33, determining the cluster center of the distance cluster containing the most candidate distances among the multiple distance clusters as the reference distance corresponding to the current axle.

[0077] Considering that the target vehicle position is not a point, the target vehicle position scanned by the vehicle scanning component can have multiple relative distances. For example, when the target vehicle position is the front of the target vehicle and the vehicle scanning component is a laser sensor, the laser sensor can have multiple laser scans at different positions of the front of the vehicle at the same time, resulting in multiple relative distances.

[0078] In this embodiment, when there are multiple relative distances, each relative distance can be converted into a distance along the driving direction based on the relative angle corresponding to each relative distance to obtain multiple candidate distances. The multiple candidate distances are then clustered to obtain multiple distance clusters, and the cluster center of the distance cluster with the most candidate distances among the multiple distance clusters is determined as the reference distance corresponding to the current axle. Here, each distance cluster in the multiple distance clusters contains some candidate distances among the multiple candidate distances. The clustering method can be a clustering method using a K-means clustering algorithm.

[0079] Through this embodiment, the multiple relative distances obtained are clustered to determine a reference distance, which can improve the accuracy of wheelbase calculation while avoiding wheelbase calculation errors caused by scanning errors.

[0080] In an exemplary embodiment, the target vehicle is identified by axle type based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle, including:

[0081] S41, acquiring point cloud data corresponding to the wheels on each axle from the vehicle point cloud data;

[0082] S42, taking the wheel on each axle as the current wheel and performing at least one of the following single-tire and double-tire identification operations to obtain single-tire and double-tire information of each axle, wherein the single-tire and double-tire information of each axle is used to indicate whether the wheel on each axle is a single-tire wheel or a double-tire wheel, and the target axle type information includes the single-tire and double-tire information of each axle:

[0083] Based on the point cloud data corresponding to the current wheel, determine a first reference distance between the current wheel and the vehicle scanning component, and determine a second reference distance between the current wheel and the vehicle scanning component, wherein the first reference distance is the distance between the edge of the tire side of the current wheel and the vehicle scanning component in a vertical direction along the driving direction of the target vehicle, and the second reference distance is the distance between the center of the tire side of the current wheel and the vehicle scanning component in a vertical direction; when the distance difference between the first reference distance and the second reference distance is greater than or equal to a preset distance difference threshold, determine that the current wheel is a dual-tire wheel; when the distance difference between the first reference distance and the second reference distance is less than the preset distance difference threshold, determine that the current wheel is a single-tire wheel;

[0084] Matching the point cloud data corresponding to the current wheel with a set of preset point cloud templates, wherein the set of point cloud templates includes a single-tire point cloud template corresponding to a single-tire wheel and a dual-tire point cloud template corresponding to a dual-tire wheel; when the point cloud data corresponding to the current wheel matches the single-tire point cloud template, determining that the current wheel is a single-tire wheel; when the point cloud data corresponding to the current wheel matches the dual-tire point cloud template, determining that the current wheel is a dual-tire wheel;

[0085] The depth information of the point cloud data corresponding to the current wheel is converted into a grayscale image to obtain a target grayscale image; single-tire or double-tire identification is performed on the target grayscale image to obtain a recognition result of the target grayscale image, wherein the recognition result of the target grayscale image is used to indicate whether the current wheel is a single-tire wheel or a double-tire wheel.

[0086] For the single or double tire information of the target vehicle, it is possible to determine whether the wheels of each axle of the target vehicle are single or double tires based on the point cloud data scanned at a certain angle by the vehicle scanning components located on both sides of the road. In this embodiment, the point cloud data corresponding to the wheels on each axle can be obtained from the vehicle point cloud data, and the wheel on each axle is used as the current wheel to perform a single or double tire identification operation to obtain the single or double tire information of each axle. Here, the single or double tire information of each axle can be used to indicate whether the wheel on each axle is a single tire wheel or a double tire wheel, and the target axle type information may include the single or double tire information of each axle.

[0087] Considering that the single or double tire of a vehicle is usually intuitively reflected in the appearance of the wheel, such as a single tire is slightly convex, while a double tire has obvious concave features, the above-mentioned single or double tire identification operation can be: based on the point cloud data corresponding to the current wheel, determine the first reference distance between the current wheel and the vehicle scanning component, and the second reference distance between the current wheel and the vehicle scanning component, and determine whether the current wheel is convex or concave according to the distance difference between the first reference distance and the second reference distance. Here, the first reference distance can be the distance between the edge of the tire side of the current wheel and the vehicle scanning component in the vertical direction along the driving direction of the target vehicle, and the second reference distance can be the distance between the center of the tire side of the current wheel and the vehicle scanning component in the vertical direction.

[0088] The point cloud data corresponding to the current wheel can be used to stitch the vehicle point cloud data of the target vehicle into a complete vehicle using a stitching algorithm or a registration algorithm, and then the outline of each wheel can be selected using a point cloud shape detection algorithm.

[0089] When the distance difference between the first reference distance and the second reference distance is greater than or equal to the preset distance difference threshold, it is determined that the side of the current wheel is concave and the current wheel is a dual-tire wheel; when the distance difference between the first reference distance and the second reference distance is less than the preset distance difference threshold, it is determined that the side of the current wheel is concave and the current wheel is a single-tire wheel.

[0090] The formula for converting polar coordinates to rectangular coordinates can be used to calculate the change in the concave-convex distance from the center point of the tire side to the edge of the tire, that is, the single-double tire identification function can be finally achieved through the depth comparison between the edge and the center of the tire side. Figure 7 As shown, the horizontal distance from the tire side edge to the laser can be expressed as formula (6):

[0091] X5 = L3 * Sinθ3 (6)

[0092] The horizontal distance from the center of the tire side to the laser can be expressed as formula (7):

[0093] X6= L4 * Sinθ4 (7)

[0094] The distance index for judging the concavity and convexity is shown in formula (8):

[0095] L7 = X6 – X5 (8)

[0096] Setting standard value like It is considered to be concave, that is, twin (such as Figure 7 Left wheel), otherwise single tire (such as Figure 7 center right wheel).

[0097] In addition, the above-mentioned single-tire and dual-tire identification operation may also be to match the point cloud data corresponding to the current wheel with a preset set of point cloud templates, and when the point cloud data corresponding to the current wheel matches the single-tire point cloud template, determine that the current wheel is a single-tire wheel; when the point cloud data corresponding to the current wheel matches the dual-tire point cloud template, determine that the current wheel is a dual-tire wheel. Here, a set of point cloud templates may include a single-tire point cloud template corresponding to a single-tire wheel and a dual-tire point cloud template corresponding to a dual-tire wheel.

[0098] The point cloud template matching algorithm searches for the part that best matches a given template in the point cloud data and calculates the similarity between them. This method usually includes steps such as point cloud preprocessing, feature extraction, feature matching, posture estimation, and evaluation feedback. Common point cloud template matching methods include feature-based, geometric shape-based, and deep learning-based. The point cloud template can be used to match the precise tire side profile data to determine whether it is a single tire or a double tire.

[0099] The above single-tire or double-tire identification operation may also be to convert the depth information of the point cloud data corresponding to the current wheel into a grayscale image to obtain a target grayscale image, and perform single-tire or double-tire identification on the target grayscale image to obtain an identification result of the target grayscale image. Here, the identification result of the target grayscale image may be used to indicate whether the current wheel is a single-tire wheel or a double-tire wheel.

[0100] The depth information of the captured, spliced ​​or registered laser point cloud data can be converted into a grayscale image, and then the image recognition method based on deep learning can be used to classify single and double fetuses, which can effectively reduce the complexity of image processing and improve the image processing speed and accuracy.

[0101] Optionally, considering the large difference in width between a single tire and a double tire, in this embodiment, it is also possible to determine whether it is a single tire or a double tire based on the rolling width measured by the weighing sensor.

[0102] Through this embodiment, the single and double tire information of the target vehicle is determined based on the concave and convex shape of the wheels of the target vehicle, which can reduce the complexity of single and double tire identification and improve the recognition efficiency of the vehicle axle type.

[0103] In an exemplary embodiment, acquiring point cloud data corresponding to the wheels on each axle from the vehicle point cloud data includes:

[0104] S51, determining a rolling time corresponding to each axle based on the vehicle detection data, wherein the rolling time corresponding to each axle is the time for each axle to roll over a road surface sensor;

[0105] S52, based on the rolling time corresponding to each axle and the driving speed of the target vehicle, determining the time period for the wheel on each axle to pass through the scanning surface of the vehicle scanning component, and obtaining the scanning time period corresponding to the wheel on each axle;

[0106] S53, acquiring point cloud data within a scanning time period corresponding to the wheels on each axle from the vehicle point cloud data, to obtain point cloud data corresponding to the wheels on each axle.

[0107] In order to improve the efficiency of wheel contour recognition, the aforementioned point cloud data corresponding to the current wheel can also be obtained by directly splicing the intercepted point cloud data of the wheel part. In this embodiment, the point cloud data of the wheel part can be determined according to the speed of the target vehicle and the time when the target vehicle rolls over the road surface sensor.

[0108] The rolling time corresponding to each axle can be determined based on the vehicle detection data. According to the determined rolling time and the driving speed of the target vehicle, the time period for the wheel on each axle to pass through the scanning surface of the vehicle scanning component can be determined, thereby obtaining the scanning time period corresponding to the wheel on each axle. Here, the rolling time corresponding to each axle can be the time for each axle to roll over the road surface sensor.

[0109] According to the scanning time period corresponding to the wheels on each axle, point cloud data can be intercepted from the vehicle point cloud data to obtain the point cloud data within the scanning time period corresponding to the wheels on each axle, thereby obtaining the point cloud data corresponding to the wheels on each axle.

[0110] The driving speed of the target vehicle can be calculated based on the data detected by the road sensor. The point cloud data obtained within the scanning time period corresponding to the wheels on each axle can be spliced ​​into the point cloud data of the wheel part using a splicing algorithm or a registration algorithm.

[0111] Through this embodiment, based on the detection data of the road surface sensor, the time period when the vehicle scanning component scans the wheel is determined, and then the corresponding point cloud data is intercepted and spliced ​​into wheel point cloud data. This can improve the efficiency of wheel contour recognition while ensuring the accuracy of wheel contour recognition.

[0112] In an exemplary embodiment, the vehicle detection data includes vehicle waveform data, which is waveform data generated by each axle rolling over a road surface sensor in sequence; the axle type of the target vehicle is identified based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle, including:

[0113] S61, determining the number of preset waveforms contained in the waveform indicated in the vehicle waveform data as the number of axles of the target vehicle, wherein the preset waveform is a preset waveform generated by the axle-rolling road surface sensor, and the target axle type information includes the number of axles of the target vehicle.

[0114] Since the axle of the target vehicle generates different pressures on the road surface sensor when passing through the road surface sensor, the road surface sensor can detect multiple pressure waves that are continuous in time but different in size. In this embodiment, the number of preset waveforms contained in the waveform indicated in the vehicle waveform data can be determined as the number of axles of the target vehicle. Here, the preset waveform can be a preset waveform generated by the axle rolling over the road surface sensor. The target axle type information can include the number of axles of the target vehicle.

[0115] Taking the road surface sensor as a narrow strip sensor as an example, after a vehicle passes through the upper and lower weighing coils of the ground narrow strip sensor, the waveform generated by a single vehicle can be recorded using the time it takes to pass through the coils and the waveform of the axle crushing can be screened out, and then the number of axles can be confirmed based on the number of waveforms recorded by a single narrow strip. That is, during the time a vehicle passes through the weighing area, if the first row of narrow strips records six waveforms, it can be determined that the vehicle is a six-axle vehicle, that is, the vehicle has six axles.

[0116] In addition, considering that some vehicles have multiple rows of double-axle or triple-axle axles, for example, the axles near the front of a large truck are generally single-axle, and the axles near the body or the rear are generally double-axle or triple-axle. In this embodiment, not only can the total number of axles of the vehicle be determined based on the number of waveforms generated when the vehicle passes, but also each axle group can be determined to be single-axle, double-axle or multi-axle based on the time interval between two adjacent waveforms. For example, if the interval between two waveforms is large, it can be explained that the axles corresponding to the two waveforms are axles of two axle groups, and if the interval between the two waveforms is small, it can be explained that the axles corresponding to the two waveforms are axles of one axle group.

[0117] In addition, it is also possible to determine whether the two axles belong to one axle group according to the distance between the two adjacent grounded axles. The method for determining the distance between the two axles can be the same as described in the above embodiment, and this embodiment will not be described in detail.

[0118] Through this embodiment, the axle number of the target vehicle is determined based on the waveform data detected by the road surface sensor, so that the efficiency of determining the axle number can be improved.

[0119] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0120] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM (Read-Only Memory) / RAM (Random Access Memory), a magnetic disk, or an optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, a computer, a server, or a network device, etc.) to execute the methods of each embodiment of the present application.

[0121] According to another aspect of the embodiment of the present application, a vehicle identification system for implementing the above vehicle identification method is also provided. The vehicle identification system may include:

[0122] A road surface sensor is used to weigh and detect a passing target vehicle to obtain vehicle detection data of the target vehicle, wherein the vehicle detection data includes data generated by each axle of a group of axles of the target vehicle rolling over the road surface sensor;

[0123] A vehicle scanning component is used to scan a target vehicle to obtain vehicle point cloud data, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle;

[0124] The data processing component is used to obtain vehicle detection data; obtain vehicle point cloud data; perform axle type recognition on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

[0125] It should be noted that the data processing component may be a component on a server or a processing device that executes the above determination of the target shaft type information, such as a processor, a controller, etc. The method of determining the target shaft type information is similar to that in the above embodiment, and has been described, so it will not be repeated here.

[0126] Through the above-mentioned vehicle identification system, vehicle detection data of the target vehicle detected by the road surface sensor is obtained, wherein the vehicle detection data includes data generated by each axle of a group of axles of the target vehicle rolling over the road surface sensor; vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle is obtained, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; axle type recognition is performed on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain the target axle type information of the target vehicle, thereby solving the problem of low accuracy of vehicle axle type recognition in the vehicle identification method in the related art and improving the accuracy of vehicle axle type recognition.

[0127] In an exemplary embodiment, the data processing component is also used to perform the following distance determination operations on each axle in turn as the current axle to obtain a reference distance corresponding to each axle: based on the vehicle detection data, determine the target rolling time corresponding to the current axle, wherein the target rolling time is the time when the current axle rolls over the road surface sensor; obtain point cloud data of the target vehicle position matching the target rolling time from the vehicle point cloud data, and obtain reference point cloud data corresponding to each axle; according to the reference point cloud data, determine the distance between the target vehicle position at the target rolling time and the vehicle scanning component along the driving direction of the target vehicle, and obtain the reference distance corresponding to the current axle; determine the distance difference between the reference distances corresponding to two adjacent axles in a group of axles as the wheelbase between the two adjacent axles, and obtain the wheelbase information of the target vehicle, wherein the target axle type information includes the wheelbase information of the target vehicle.

[0128] In an exemplary embodiment, the data processing component is also used to determine the relative distance between the target vehicle position and the vehicle scanning component at the target crushing time and the relative angle between the target vehicle position and the vehicle scanning component at the target crushing time based on the reference point cloud data; based on the relative angle, the relative distance is converted into a distance along the driving direction to obtain a reference distance corresponding to the current axle.

[0129] In an exemplary embodiment, the data processing component is also used to, when there are multiple relative distances, convert each relative distance into a distance along the driving direction based on the relative angle corresponding to each relative distance, so as to obtain multiple candidate distances; cluster the multiple candidate distances to obtain multiple distance clusters, wherein each distance cluster in the multiple distance clusters contains some candidate distances in the multiple candidate distances; and determine the cluster center of the distance cluster that contains the most candidate distances among the multiple distance clusters as the reference distance corresponding to the current axle.

[0130] In an exemplary embodiment, the data processing component is also used to obtain point cloud data corresponding to the wheels on each axle from the vehicle point cloud data; perform at least one of the following single-tire and double-tire identification operations on the wheel on each axle as the current wheel to obtain the single-tire and double-tire information of each axle, wherein the single-tire and double-tire information of each axle is used to indicate whether the wheel on each axle is a single-tire wheel or a double-tire wheel, and the target axle type information includes the single-tire and double-tire information of each axle: based on the point cloud data corresponding to the current wheel, determine a first reference distance between the current wheel and the vehicle scanning component, and determine a second reference distance between the current wheel and the vehicle scanning component, wherein the first reference distance is the distance between the side edge of the tire of the current wheel and the vehicle scanning component in the vertical direction along the driving direction of the target vehicle, and the second reference distance is the distance between the side center of the tire of the current wheel and the vehicle scanning component in the vertical direction; the distance between the first reference distance and the second reference distance When the difference between the first reference distance and the second reference distance is greater than or equal to the preset distance difference threshold, it is determined that the current wheel is a dual-tire wheel; when the difference between the first reference distance and the second reference distance is less than the preset distance difference threshold, it is determined that the current wheel is a single-tire wheel; the point cloud data corresponding to the current wheel is matched with a preset set of point cloud templates, wherein a set of point cloud templates includes a single-tire point cloud template corresponding to a single-tire wheel and a dual-tire point cloud template corresponding to a dual-tire wheel; when the point cloud data corresponding to the current wheel matches the single-tire point cloud template, it is determined that the current wheel is a single-tire wheel; when the point cloud data corresponding to the current wheel matches the dual-tire point cloud template, it is determined that the current wheel is a dual-tire wheel; the depth information of the point cloud data corresponding to the current wheel is converted into a grayscale image to obtain a target grayscale image; single-tire and dual-tire identification is performed on the target grayscale image to obtain a recognition result of the target grayscale image, wherein the recognition result of the target grayscale image is used to indicate whether the current wheel is a single-tire wheel or a dual-tire wheel.

[0131] In an exemplary embodiment, based on vehicle detection data, the rolling time corresponding to each axle is determined, wherein the rolling time corresponding to each axle is the time for each axle to roll over a road surface sensor; based on the rolling time corresponding to each axle and the driving speed of the target vehicle, the time period for the wheels on each axle to pass through a scanning surface of a vehicle scanning component is determined to obtain a scanning time period corresponding to the wheels on each axle; point cloud data within the scanning time period corresponding to the wheels on each axle is obtained from the vehicle point cloud data to obtain point cloud data corresponding to the wheels on each axle.

[0132] In an exemplary embodiment, the data processing component is also used to determine the number of preset waveforms contained in the waveform indicated in the vehicle waveform data as the number of axles of the target vehicle, wherein the preset waveform is a preset waveform generated by the axle rolling road surface sensor, and the target axle type information includes the number of axles of the target vehicle.

[0133] It should be noted that the examples and application scenarios implemented by the above modules and corresponding steps are the same, but are not limited to the contents disclosed in the above embodiments. Figure 1 In the hardware environment shown, it can be implemented by software or by hardware, wherein the hardware environment includes a network environment.

[0134] According to another aspect of the embodiment of the present application, a storage medium is also provided. Optionally, in this embodiment, the storage medium can be used to execute the program code of any of the above vehicle identification methods in the embodiment of the present application.

[0135] Optionally, in this embodiment, the storage medium may be located on at least one network device among a plurality of network devices in the network shown in the above embodiment.

[0136] Optionally, in this embodiment, the storage medium is configured to store program codes for executing the following steps:

[0137] S1, obtaining vehicle detection data of a target vehicle detected by a road surface sensor, wherein the vehicle detection data includes data generated when each axle in a group of axles of the target vehicle rolls over the road surface sensor;

[0138] S2, obtaining vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle;

[0139] S3, performing axle type recognition on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

[0140] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, which will not be described in detail in this embodiment.

[0141] Optionally, in this embodiment, the storage medium may include, but is not limited to, various media that can store program codes, such as a USB flash drive, a ROM, a RAM, a mobile hard disk, a magnetic disk, or an optical disk.

[0142] According to another aspect of an embodiment of the present application, an electronic device for implementing the above-mentioned vehicle identification method is also provided. The electronic device may be a server, a terminal, or a combination thereof.

[0143] Figure 8 is a structural block diagram of an optional electronic device according to an embodiment of the present application, such as Figure 8As shown, it includes a processor 802, a communication interface 804, a memory 806 and a communication bus 808, wherein the processor 802, the communication interface 804 and the memory 806 communicate with each other through the communication bus 808, wherein,

[0144] Memory 806, used for storing computer programs;

[0145] The processor 802 is used to implement the following steps when executing the computer program stored in the memory 806:

[0146] S1, obtaining vehicle detection data of a target vehicle detected by a road surface sensor, wherein the vehicle detection data includes data generated when each axle in a group of axles of the target vehicle rolls over the road surface sensor;

[0147] S2, obtaining vehicle point cloud data obtained by the vehicle scanning component when scanning the target vehicle, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle;

[0148] S3, performing axle type recognition on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

[0149] Optionally, the communication bus may be a PCI (Peripheral Component Interconnect) bus, or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 8 The communication interface is used for communication between the electronic device and other devices.

[0150] The memory may include RAM, or may include non-volatile memory, such as at least one disk memory. Optionally, the memory may also be at least one storage device located away from the aforementioned processor.

[0151] The above-mentioned processor can be a general-purpose processor, which can include but not be limited to: CPU (Central Processing Unit), NP (Network Processor), etc.; it can also be DSP (Digital Signal Processing), ASIC (Application Specific Integrated Circuit), FPGA (Field-Programmable Gate Array) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components.

[0152] Optionally, the specific examples in this embodiment may refer to the examples described in the above embodiments, and this embodiment will not be described in detail here.

[0153] It can be understood by those skilled in the art that Figure 8 The structure shown is for illustration only. The device for implementing the above-mentioned vehicle identification method may be a terminal device, which may be a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, a mobile Internet device (Mobile Internet Devices, MID), a PAD, and other terminal devices. Figure 8 The structure of the electronic device is not limited. Figure 8 More or fewer components (such as network interfaces, display devices, etc.) shown in, or having Figure 8 Different configurations shown.

[0154] A person of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing the hardware related to the terminal device through a program, and the program can be stored in a computer-readable storage medium, which can include: a flash drive, ROM, RAM, a magnetic disk or an optical disk, etc.

[0155] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0156] If the integrated units in the above embodiments are implemented in the form of software functional units and sold or used as independent products, they can be stored in the above computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, which is stored in a storage medium and includes several instructions for enabling one or more computer devices (which can be personal computers, servers or network devices, etc.) to execute all or part of the steps of the above methods in each embodiment of the present application.

[0157] In the above embodiments of the present application, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0158] In the several embodiments provided in the present application, it should be understood that the disclosed client can be implemented in other ways. Among them, the device embodiments described above are only schematic, for example, the division of units is only a logical function division, and there may be other division methods in actual implementation, for example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0159] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution provided in this embodiment.

[0160] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, each unit may exist physically separately, or at least two units may be integrated into one unit. The above integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0161] The above are only preferred implementations of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles of the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A vehicle identification method, characterized in that: include: Acquire vehicle detection data of a target vehicle detected by a road surface sensor, wherein the vehicle detection data includes data generated when each axle of a group of axles of the target vehicle rolls over the road surface sensor; Acquire vehicle point cloud data obtained by a vehicle scanning component when scanning the target vehicle, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; The target vehicle is identified by axle type recognition based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

2. The method according to claim 1, characterized in that The step of performing axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle includes: The following distance determination operations are performed in sequence with each axle as the current axle to obtain a reference distance corresponding to each axle: based on the vehicle detection data, a target rolling time corresponding to the current axle is determined, wherein the target rolling time is the time when the current axle rolls over the road sensor; point cloud data of a target vehicle position matching the target rolling time is acquired from the vehicle point cloud data to obtain reference point cloud data corresponding to each axle; based on the reference point cloud data, a distance between the target vehicle position and the vehicle scanning component at the target rolling time along the driving direction of the target vehicle is determined to obtain a reference distance corresponding to the current axle; The distance difference between the reference distances corresponding to two adjacent axles in the group of axles is determined as the wheelbase between the two adjacent axles to obtain the wheelbase information of the target vehicle, wherein the target axle type information includes the wheelbase information of the target vehicle.

3. The method according to claim 2, characterized in that The step of determining, based on the reference point cloud data, a distance between the target vehicle position and the vehicle scanning component at the target rolling time along the driving direction of the target vehicle to obtain a reference distance corresponding to the current axle includes: Determine, based on the reference point cloud data, a relative distance between the target vehicle position and the vehicle scanning component at the target rolling time and a relative angle between the target vehicle position and the vehicle scanning component at the target rolling time; Based on the relative angle, the relative distance is converted into a distance along the driving direction to obtain a reference distance corresponding to the current axle.

4. The method according to claim 3, characterized in that The converting the relative distance into a distance along the driving direction based on the relative angle to obtain a reference distance corresponding to the current axle includes: In the case where there are multiple relative distances, based on the relative angle corresponding to each relative distance, each relative distance is converted into a distance along the driving direction to obtain multiple candidate distances; Clustering the multiple candidate distances to obtain multiple distance clusters, wherein each distance cluster in the multiple distance clusters includes part of the multiple candidate distances; The cluster center of the distance cluster having the most candidate distances among the multiple distance clusters is determined as the reference distance corresponding to the current axle.

5. The method according to claim 1, characterized in that The step of performing axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle includes: Respectively acquiring point cloud data corresponding to the wheels on each axle from the vehicle point cloud data; The wheel on each axle is taken as the current wheel and at least one of the following single-tire and double-tire identification operations is performed to obtain the single-tire and double-tire information of each axle, wherein the single-tire and double-tire information of each axle is used to indicate whether the wheel on each axle is a single-tire wheel or a double-tire wheel, and the target axle type information includes the single-tire and double-tire information of each axle: Based on the point cloud data corresponding to the current wheel, determine a first reference distance between the current wheel and the vehicle scanning component, and determine a second reference distance between the current wheel and the vehicle scanning component, wherein the first reference distance is the distance between the edge of the tire side of the current wheel and the vehicle scanning component in a vertical direction along the driving direction of the target vehicle, and the second reference distance is the distance between the center of the tire side of the current wheel and the vehicle scanning component along the vertical direction; when the distance difference between the first reference distance and the second reference distance is greater than or equal to a preset distance difference threshold, determine that the current wheel is a dual-tire wheel; when the distance difference between the first reference distance and the second reference distance is less than the preset distance difference threshold, determine that the current wheel is a single-tire wheel; Matching the point cloud data corresponding to the current wheel with a set of preset point cloud templates, wherein the set of point cloud templates includes a single-tire point cloud template corresponding to a single-tire wheel and a dual-tire point cloud template corresponding to a dual-tire wheel; when the point cloud data corresponding to the current wheel matches the single-tire point cloud template, determining that the current wheel is a single-tire wheel; when the point cloud data corresponding to the current wheel matches the dual-tire point cloud template, determining that the current wheel is a dual-tire wheel; The depth information of the point cloud data corresponding to the current wheel is converted into a grayscale image to obtain a target grayscale image; single-tire or double-tire identification is performed on the target grayscale image to obtain a recognition result of the target grayscale image, wherein the recognition result of the target grayscale image is used to indicate whether the current wheel is a single-tire wheel or a double-tire wheel.

6. The method according to claim 5, characterized in that The step of respectively acquiring point cloud data corresponding to the wheels on each axle from the vehicle point cloud data comprises: Determine the rolling time corresponding to each axle based on the vehicle detection data, wherein the rolling time corresponding to each axle is the time when each axle rolls over the road surface sensor; Based on the rolling time corresponding to each axle and the travel speed of the target vehicle, determine the time period during which the wheel on each axle passes through the scanning surface of the vehicle scanning component to obtain the scanning time period corresponding to the wheel on each axle; The point cloud data within a scanning time period corresponding to the wheels on each axle is acquired from the vehicle point cloud data to obtain the point cloud data corresponding to the wheels on each axle.

7. The method according to any one of claims 1 to 6, characterized in that The vehicle detection data includes vehicle waveform data, and the vehicle waveform data is waveform data generated by each axle rolling over the road surface sensor in sequence; The step of performing axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle includes: The number of preset waveforms contained in the waveform indicated in the vehicle waveform data is determined as the number of axles of the target vehicle, wherein the preset waveform is a preset waveform generated by the axle rolling over the road sensor, and the target axle type information includes the number of axles of the target vehicle.

8. A vehicle identification system, characterized in that: include: A road surface sensor, used to weigh and detect a passing target vehicle to obtain vehicle detection data of the target vehicle, wherein the vehicle detection data includes data generated when each axle of a group of axles of the target vehicle rolls over the road surface sensor; A vehicle scanning component, used for scanning the target vehicle to obtain vehicle point cloud data, wherein the vehicle point cloud data is used to represent the vehicle outer contour of the target vehicle; The data processing component is used to obtain the vehicle detection data; obtain the vehicle point cloud data; and perform axle type identification on the target vehicle based on the vehicle detection data and the vehicle point cloud data to obtain target axle type information of the target vehicle.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method according to any one of claims 1 to 7 when executed.

10. An electronic device comprising a memory and a processor, characterized in that: A computer program is stored in the memory, and the processor is configured to execute the method according to any one of claims 1 to 7 through the computer program.

Images