Vehicle axle type identification method and device, electronic device, and storage medium

Abstract

The application relates to a vehicle axle type identification method and device, an electronic device and a storage medium, wherein the vehicle axle type identification method comprises the following steps: in response to the fact that a first position relationship between a wheel axle tracking result of a target vehicle in a real-time acquired to-be-detected image and a preset early warning identifier meets a preset caching condition, updating the to-be-detected image to image caching information until it is determined that the to-be-detected image meets a preset caching termination condition based on the wheel axle tracking result and the image caching information; performing image processing on the to-be-detected image in the image caching information according to local features of wheel axle objects in the image caching information, and performing axle type identification based on a result of the image processing to obtain axle type information of the target vehicle. The application realizes accurate identification of the axle type of the vehicle based on the vehicle image, and does not need to rely on manual reading or laser radar scanning, so that the labor cost and equipment cost for identifying the axle type of the vehicle can be reduced.

Description

Technical Field

[0001] This application relates to the field of image processing, and in particular to methods, apparatus, electronic devices, and storage media for vehicle axle shape recognition. Background Technology

[0002] Currently, based on the technological development background of "smart cities" and "safe cities," the "source control of overloading" of freight vehicles is inextricably linked to the realization of road safety management. Among them, the axle type identification of freight vehicles is the first link in the "source control of overloading" business, and its identification results are a key indicator for measuring the reliability of the "source control of overloading" measures.

[0003] However, current axle identification for freight vehicles often relies on manual reading or LiDAR scanning. Therefore, current axle identification solutions require high manpower costs, and using LiDAR identification systems also incurs high equipment costs.

[0004] There is currently no effective solution to the problem of high costs associated with vehicle axle type recognition in related technologies. Summary of the Invention

[0005] This embodiment provides a vehicle axle type recognition method, apparatus, electronic device, and storage medium to address the problem of high cost in vehicle axle type recognition in related technologies.

[0006] Firstly, this embodiment provides a vehicle axle type recognition method, including:

[0007] In response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching condition, the image to be detected is updated to the image cache information until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset cache termination condition; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle;

[0008] Based on the local features of the wheel axle object in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle type recognition is performed based on the result of the image processing to obtain the axle type information of the target vehicle.

[0009] In some embodiments, the preset caching condition includes: the first wheel axle object tracked from the image to be detected touches the warning identifier;

[0010] The preset cache termination condition includes: no new wheel axle object is tracked from the image to be detected within a preset time period.

[0011] In some embodiments, updating the image to be detected to the image cache information includes:

[0012] During the process of updating the image to be detected to the image cache information, the number of cached frames of the image in the image cache information is counted until it is determined based on the wheel axle tracking result that a new wheel axle object has touched the warning identifier, then the number of cached frames is cleared to zero and counted again.

[0013] In some embodiments, the method further includes:

[0014] If the wheel axle tracking result does not update a new wheel axle object within a preset time period, and the number of cached frames of the image in the image cache information reaches a preset frame threshold, the image to be detected is confirmed to meet the preset cache termination condition.

[0015] In some embodiments, the step of updating the image to be detected to image cache information in response to a first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and a preset warning identifier meeting a preset caching condition, until the image to be detected meets a preset caching termination condition based on the wheel axle tracking result and the image cache information, includes:

[0016] In response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier conforming to the preset caching conditions, the inter-frame step of the updated image cache information is adjusted according to the displacement vector of the target vehicle.

[0017] The image to be detected is updated to the image cache information based on the adjusted inter-frame step size, until the image to be detected meets the preset cache termination condition based on the wheel axle tracking result and the image cache information.

[0018] In some embodiments, the step of performing image processing on the image to be detected in the image cache information based on the local features of the wheel axle object in the image cache information, and performing axle type recognition based on the result of the image processing to obtain the axle type information of the target vehicle includes:

[0019] Based on the local features of the wheel axle object in the image cache information, the image to be detected in the image cache information is stitched together, and the axle type is identified based on the result of the image stitching to obtain the axle type information of the target vehicle.

[0020] In some embodiments, the axle shape recognition based on the image processing results to obtain the axle shape information of the target vehicle includes:

[0021] The image processing result is subjected to wheel axle detection to obtain the first wheel axle detection result;

[0022] Based on the preset spacing encoding rules and the spacing ratio of each axle object contained in the first axle detection result, the spacing of each axle object is encoded and arranged, and the axle type information of the target vehicle is determined based on the encoding arrangement result.

[0023] In some embodiments, in response to a first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and a preset warning identifier, which meets a preset caching condition, the image to be detected is updated to image cache information. Until it is determined, based on the wheel axle tracking result and the image cache information, that the image to be detected meets a preset caching termination condition, the method further includes:

[0024] The wheel axle objects of the target vehicle in the image to be detected are detected based on the trained detection model to obtain the second wheel axle detection result;

[0025] According to the corner distance metric, the second wheel axle detection result is associated with the preset wheel axle tracking list, the wheel axle tracking list is updated based on the association result, and the wheel axle tracking list is identified as the wheel axle tracking result.

[0026] In some embodiments, detecting the wheel axle objects of the target vehicle in the image to be detected based on the trained detection model to obtain a second wheel axle detection result includes:

[0027] The wheel axle objects of the target vehicle in the image to be detected are detected based on the trained detection model to obtain the initial detection results of the wheel axle objects;

[0028] Based on the position information of the wheel axle object and the confidence level of the initial detection result, the initial detection result is filtered to obtain the second wheel axle detection result.

[0029] In some embodiments, in response to a first positional relationship between the wheel axle tracking result of a target vehicle in a real-time acquired image to be detected and a preset warning identifier, which meets a preset caching condition, the image to be detected is updated to image cache information. Until it is determined, based on the wheel axle tracking result and the image cache information, that the image to be detected meets a preset caching termination condition, the method further includes:

[0030] Based on the wheel axle tracking results, a second positional relationship is determined between the first tracked wheel axle object in the image to be detected and multiple preset identifiers;

[0031] Based on the second positional relationship, the direction of entry of the target vehicle is determined, and an identifier corresponding to the direction of entry is selected from the plurality of preset identifiers as a warning identifier; wherein, there is a pre-established correspondence between the plurality of preset identifiers and the direction of entry.

[0032] Secondly, this embodiment provides a vehicle axle type recognition device, including: a cache module and a recognition module; wherein:

[0033] The caching module is configured to update the image to be detected to the image cache information in response to a first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and a preset warning identifier, which meets a preset caching condition, until the image to be detected meets a preset caching termination condition based on the wheel axle tracking result and the image cache information; wherein the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle;

[0034] The recognition module is used to perform image processing on the image to be detected in the image cache information based on the local features of the wheel axle object in the image cache information, and to perform axle type recognition based on the result of the image processing to obtain the axle type information of the target vehicle.

[0035] Thirdly, this embodiment provides an electronic device including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the vehicle axle type recognition method described in the first aspect above.

[0036] Fourthly, this embodiment provides a storage medium storing a computer program that, when executed by a processor, implements the vehicle axle type recognition method described in the first aspect above.

[0037] Compared with related technologies, the vehicle axle type recognition method, apparatus, electronic device, and storage medium provided in this embodiment, in response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching conditions, updates the image to be detected to the image cache information until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset caching termination condition; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle; according to the local features of the wheel axle object in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle type recognition is performed based on the image processing result to obtain the axle type information of the target vehicle. It achieves accurate recognition of vehicle axle type based on vehicle images, without relying on manual reading or lidar scanning, thereby reducing the manual and equipment costs of vehicle axle type recognition.

[0038] Details of one or more embodiments of this application are set forth in the following drawings and description to make other features, objects and advantages of this application more readily apparent. Attached Figure Description

[0039] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0040] Figure 1 This is an application scenario diagram of the vehicle axle type recognition method in this embodiment;

[0041] Figure 2 This is a hardware structure block diagram of the terminal of the vehicle axle type recognition method in this embodiment;

[0042] Figure 3 This is a flowchart of the vehicle axle type recognition method in this embodiment;

[0043] Figure 4 This is a schematic diagram of the splicing result in this embodiment;

[0044] Figure 5 This is a schematic diagram of the first wheel axle detection results in this embodiment;

[0045] Figure 6 This is a schematic diagram of the wheel assembly division in this embodiment;

[0046] Figure 7 This is a flowchart of the vehicle axle type recognition method according to a preferred embodiment;

[0047] Figure 8 This is a structural block diagram of the vehicle axle type recognition device in this embodiment. Detailed Implementation

[0048] To better understand the purpose, technical solution, and advantages of this application, the application is described and illustrated below in conjunction with the accompanying drawings and embodiments.

[0049] Unless otherwise defined, the technical or scientific terms used in this application shall have the general meaning as understood by one of ordinary skill in the art to which this application pertains. Words such as “a,” “an,” “an,” “the,” “the,” and “these,” used in this application, do not indicate quantitative limitation and may be singular or plural. The terms “comprising,” “including,” “having,” and any variations thereof used in this application are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or device that comprises a series of steps or modules (units) is not limited to the listed steps or modules (units) but may include steps or modules (units) not listed, or may include other steps or modules (units) inherent to such processes, methods, products, or devices. The terms “connected,” “linked,” and “coupled,” used in this application, are not limited to physical or mechanical connections but may include electrical connections, whether direct or indirect. The term “multiple” used in this application refers to two or more. The "and / or" operator describes the relationship between related objects, indicating that three relationships can exist. For example, "A and / or B" can represent three cases: A alone, A and B simultaneously, and B alone. Typically, the character " / " indicates that the objects before and after it are in an "or" relationship. The terms "first," "second," and "third," etc., used in this application are merely for distinguishing similar objects and do not represent a specific ordering of the objects.

[0050] Figure 1 This diagram illustrates an application scenario of the vehicle axle type recognition method according to this embodiment. The method embodiment provided in this embodiment can be applied to, for example... Figure 1 In the application scenario shown, warning lines L and R can be set within the camera's field of view. When the target vehicle enters from the left side of the camera's field of view, warning line L is activated and identified as a warning identifier; when the target vehicle enters from the right side of the camera's field of view, warning line R is activated and identified as a warning identifier. The target vehicle in this embodiment can be any type of motor vehicle. Specifically, in conjunction with... Figure 1Taking a cargo truck as an example: When the camera detects that the first axle of a cargo truck entering from the left touches the warning line L, the image to be detected is updated to the image cache information. This process continues until, based on the axle tracking results of the cargo truck and the image cache information, it is determined that the camera's field of view has not tracked any new axle objects within a preset time period, at which point the caching of the image to be detected ends. Then, based on the local features of the axle objects in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle shape recognition is performed on the image processing results to obtain the axle shape information of the cargo truck.

[0051] The method embodiments provided in this example can be executed on a terminal, such as a computer or smart camera, or on a server. For example, after the camera captures images, it uploads them to the server, where the server can then recognize them. Alternatively, it can be executed in the cloud or on a distributed system. Taking execution on a terminal as an example... Figure 2 This is a hardware structure block diagram of the terminal for the vehicle axle type recognition method in this embodiment. For example... Figure 2 As shown, a terminal may include one or more ( Figure 2 Only one is shown in the diagram. A processor 202 and a memory 204 for storing data are also included. The processor 202 may be, but is not limited to, a microprocessor (MCU) or a programmable logic device (FPGA). The terminal may also include a transmission device 206 for communication functions and an input / output device 208. Those skilled in the art will understand that… Figure 2 The structure shown is for illustrative purposes only and does not limit the structure of the terminal described above. For example, the terminal may also include components that are larger than... Figure 2 The more or fewer components shown, or having the same Figure 2 The different configurations shown are illustrated.

[0052] The memory 204 can be used to store computer programs, such as application software programs and modules, like the computer program corresponding to the vehicle axle type recognition method in this embodiment. The processor 202 executes various functional applications and data processing by running the computer program stored in the memory 204, thereby implementing the above-described method. The memory 204 may include high-speed random access memory, and may also include non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some instances, the memory 204 may further include memory remotely located relative to the processor 202, and these remote memories can be connected to the terminal via a network. Examples of such networks include, but are not limited to, the Internet, corporate intranets, local area networks, mobile communication networks, and combinations thereof.

[0053] Transmission device 206 is used to receive or send data via a network. This network includes a wireless network provided by the terminal's communication provider. In one example, transmission device 206 includes a Network Interface Controller (NIC), which can connect to other network devices via a base station to communicate with the Internet. In another example, transmission device 206 can be a Radio Frequency (RF) module used for wireless communication with the Internet.

[0054] This embodiment provides a vehicle axle type recognition method. Figure 3 This is a flowchart of the vehicle axle type recognition method in this embodiment, as follows: Figure 3 As shown, the process includes the following steps:

[0055] Step S310: In response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching conditions, the image to be detected is updated to the image cache information until the image to be detected meets the preset caching termination conditions based on the wheel axle tracking result and the image cache information; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle.

[0056] The warning identifier can be any graphic within the monitoring screen used to determine whether to enable caching of the image to be detected, based on the needs of the actual application scenario. Examples include warning lines, warning frames, or several warning coordinate points within the monitoring screen, or other graphics suitable for the monitoring scenario. This embodiment does not impose specific limitations. Preferably, the warning identifier can be a warning line. Specifically, the wheel axle tracking result is the tracking result of the wheel axle objects of the target vehicle detected in the image to be detected. The wheel axle tracking result includes the position information and corresponding identification information of all wheel axle objects tracked in the camera's field of view within the image to be detected. Understandably, for the same wheel axle object across different frames, its identification information is unique, and its position information in the wheel axle tracking result is updated in real time as the position of the wheel axle object changes in the image. Furthermore, based on the position information of each wheel axle object in the wheel axle tracking result and the position information of the warning identifier, the first positional relationship can be determined. Preferably, the first tracked axle object in the current time period's axle tracking results can be determined as the first tracked axle object. The current position information of this first axle object is compared with the position information of the warning identifier to determine the first positional relationship. Image cache information refers to the relevant cache information obtained by caching images that meet the caching conditions, such as the cached images, the number of cached frames, and other information that needs to be recorded during the caching process according to the actual application requirements. Specifically, this image cache information can be located in a cache pool.

[0057] The aforementioned preset caching condition indicates that an axle object has touched the aforementioned warning identifier. If it can be determined from the aforementioned first positional relationship that an axle object has touched the warning identifier, then the first positional relationship meets the preset caching condition. The aforementioned caching termination condition indicates that no new axle object has been tracked from the image to be detected within a preset time period, indicating that all axle objects of the target vehicle have appeared in the camera's field of view, or that the target vehicle has left the camera's field of view. In this case, it is not necessary to continue updating the image to be detected to the image cache information, and image processing can be performed based on the current image cache information, such as image stitching based on the current image cache information. Specifically, if it is determined from the axle tracking results and the number of cached frames of the image cache information that no new axle object of the target vehicle has been tracked within the preset time period, it can be determined that the image to be detected meets the preset caching termination condition.

[0058] Preferably, when tracking detected axle objects, the tracking information of the axle objects can be updated to a preset tracking list. This tracking list consists of several nodes, each node including the coordinates, category, and assigned ID (Identity document) of a tracked axle object. Furthermore, this tracking list is continuously updated based on the information of the axle objects detected in the current frame as the frame number of the image to be detected changes. Each axle object in each node contains a tracking bounding box. When the tracking bounding box of the first axle object in the tracking list touches the identifier—for example, when the target vehicle enters the camera's field of view from the left, and the tracking bounding box of the first axle object touches the preset left-side warning line—the image to be detected is updated to the image cache information. During the caching process, a preset cache frame count accumulation field can be used to count the number of cached frames for the axle objects that have touched the warning line. After the next axle object touches the warning line, the cache frame count accumulation field is cleared to zero, and the cache frame count is restarted.

[0059] Understandably, if no new axle object of the target vehicle is tracked within a preset time period, the accumulated cache frame count field will not be cleared during that period. Therefore, by comparing the value of this accumulated cache frame count field (i.e., the cached frame count) with a preset frame count threshold, it can be determined whether no new axle object was tracked within the preset time period. Additionally, whether the aforementioned tracking list is updated with new nodes within the preset time period also indicates whether a new axle object was tracked within that period. Based on this, by combining the cached frame count in the image cache information with the axle tracking results, it can be determined whether to terminate the caching process.

[0060] Next, if the first positional relationship between the wheel axle tracking result and the preset warning identifier meets the preset caching conditions, the image to be detected is updated to the image cache information. Understandably, the image to be detected is a series of frames captured by the camera within its field of view, containing images of the wheel axle objects of the target vehicle. For example, the caching process may specifically include: enabling a preset image cache signal `cache_start_flag`, for example, setting its state to 1; starting caching from the current frame image in the image to be detected; storing the current frame image in a preset cache pool; and continuously caching subsequent frames as the number of image frames captured by the camera changes, until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset caching termination condition. Then, the state of the image cache signal `cache_start_flag` is set to 0, ending the caching operation for the image to be detected.

[0061] Step S320: Based on the local features of the wheel axle objects in the image cache information, perform image processing on the image to be detected in the image cache information, and perform axle type recognition based on the image processing results to obtain the axle type information of the target vehicle.

[0062] After caching is complete, the `cache_stop_flag` flag can be set to 1 to enable image processing of the images to be detected from the cached image information in the cache pool. Specifically, this image processing can be image stitching. The set of images to be stitched, composed of the images to be detected in the cache pool, is sequentially input into a preset image stitching algorithm in the order they entered the cache pool to complete the matching and stitching of local features, thereby outputting a complete stitched image and obtaining the image stitching result. Figure 4 This is a schematic diagram of the splicing result in this embodiment. Figure 4 It is known that in a surveillance scenario, the camera can be mounted at a height appropriate to the axles of the target vehicle to ensure that the axles are completely detected within the camera's field of view. However, due to limitations imposed by the camera's own field of view and the vehicle's structure, a single frame of the image to be detected may only contain a portion of the axles, for example, only one axle. Therefore, it is necessary to stitch the images to be detected from the image buffer information in step S310 to obtain the desired result. Figure 4 The image shown contains all the wheel and axle objects of the target vehicle.

[0063] After obtaining the image processing results, these results can be input into a preset wheel and axle detection algorithm to detect all wheel and axle objects. Based on the positional relationships between all wheel and axle objects in the image processing results, the axle type information of the target vehicle can be determined. Preferably, the axle type of the target vehicle can be determined based on the spacing ratio between all detected wheel and axle objects.

[0064] Steps S310 to S320, in response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching conditions, update the image to be detected to the image cache information until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset caching termination condition; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle; according to the local features of the wheel axle object in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle type recognition is performed based on the image processing result to obtain the axle type information of the target vehicle. This achieves accurate identification of vehicle axle type based on vehicle images, without relying on manual reading or LiDAR scanning, thereby reducing the manual and equipment costs of axle type recognition for vehicles.

[0065] Furthermore, in one embodiment, the preset caching conditions include: the first wheel axle object tracked from the image to be detected touches the warning identifier; the preset caching termination conditions include: no new wheel axle object is tracked from the image to be detected within a preset time period.

[0066] Understandably, when the first wheel axle object is detected touching the warning identifier within the camera's field of view, it indicates that a target vehicle has entered the detection area for axle type recognition, thus requiring the activation of axle type recognition for that target vehicle. Correspondingly, if no new wheel axle object is tracked from the image to be detected within a preset time period, it indicates that the target vehicle has left the camera's field of view, or that all wheels of the target vehicle have left the camera's field of view, thus pausing the image to be detected can be terminated. Based on this, this embodiment uses the detection of the first tracked wheel axle object touching the warning identifier as the flag to activate pausing the image to be detected, thereby accurately initiating pausing the image to be detected and using the failure to track a new wheel axle object within a preset time period as the flag to terminate pausing, thereby improving the completeness of the target vehicle's wheel axle area image obtained through subsequent image stitching.

[0067] In one embodiment, updating the image to be detected to the image cache information based on step S310 above may further include the following steps:

[0068] Step S311: During the process of updating the image to be detected to the image cache information, the number of cached frames of the image in the image cache information is counted until it is determined based on the wheel axle tracking result that a new wheel axle object has touched the warning identifier, the number of cached frames is cleared to zero and counted again.

[0069] Specifically, based on the above analysis, it can be seen that the caching of the image to be detected can be terminated if no new wheel axle objects are tracked within a preset time period. During the caching process, a counter field for the number of cached frames can be set. Whenever a new wheel axle object touches the warning identifier, the cached frame count is reset to zero and restarted. If no new wheel axle object touches the warning identifier within a certain period, the cached frame count will continue to accumulate during that period. Therefore, if the number of cached frames exceeds the preset frame count threshold, it indicates that no new wheel axle object has touched the warning identifier within the preset time period. This embodiment, by setting the count of cached frames during the image caching process, can more accurately determine the entry and exit of the target vehicle's wheel axle objects within the camera's field of view, thereby more accurately determining the timing for terminating caching, avoiding caching too much useless information, and improving the efficiency of subsequent image processing and recognition.

[0070] Furthermore, in one embodiment, the above-described vehicle axle type identification method may further include the following steps:

[0071] Step S312: In response to the fact that the wheel axle tracking result has not been updated with a new wheel axle object within a preset time period, and the number of cached frames of the image in the image cache information has reached a preset frame number threshold, the image to be detected is confirmed to meet the preset cache termination condition.

[0072] As described in step S311 above, on the one hand, the entry and exit status of wheel axle objects within the camera's field of view can be determined based on the number of cached frames in the image cache information. On the other hand, if no new wheel axle objects are added within a preset time period in the real-time wheel axle tracking results, it also indicates that all wheel axle objects of the target vehicle have left the camera's field of view. Therefore, by combining the wheel axle tracking results and the image cache information, the updating of the image to be detected to the image cache information can be stopped when the wheel axle tracking results have not updated with new wheel axle objects within a preset time period and the number of cached frames in the image cache information reaches a preset frame threshold. This embodiment determines the cache termination condition based on the update status of wheel axle objects in the wheel axle tracking results and the number of cached frames, which can more accurately determine the timing of cache termination and reduce the caching of invalid images.

[0073] In another embodiment, based on step S310 above, in response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching conditions, the image to be detected is updated to the image cache information until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset caching termination condition. Specifically, this may include the following steps:

[0074] Step S313: In response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier conforming to the preset cache condition, the inter-frame step distance of the updated image cache information is adjusted according to the displacement vector of the target vehicle.

[0075] The inter-frame step size represents the number of image frames intervening when caching the image to be detected. For example, if the inter-frame step size is 5 frames, then when updating the image to be detected to the image cache information, one image to be detected is cached every 5 frames. The displacement vector of the target vehicle includes the vehicle speed and the vehicle's direction of entry. The vehicle speed can be the average displacement speed of the first axle object of the target vehicle in the N frames before it touches the warning identifier. The direction of entry of the target vehicle can be determined based on the positional relationship between the first axle object and several pre-set identifiers. For example, if the x-coordinate of the center point of the first axle object is less than the x-coordinate of the left warning line, then the target vehicle is entering the camera's field of view from left to right; similarly, if the x-coordinate of the center point of the first axle object is greater than the x-coordinate of the right warning line, then the target vehicle is entering the camera's field of view from right to left. The center of the axle object can be the center point of the detection box of the axle object.

[0076] To ensure the number of cached images is adapted to the displacement vector of the target vehicle, the cached images should contain all wheel and axle objects of the target vehicle. The faster the target vehicle's speed, the smaller the frame interval for caching. Preferably, the inter-frame step size for caching the images to be detected can be adaptively adjusted based on the ratio between the vehicle speed and a set speed threshold. For example, if the target vehicle's speed is v_vehicle, the preset speed threshold is V_set, and the ratio between them is α, then the inter-frame step size Δstep can be 1 / α. When α is 1, it is a full-frame-rate cache.

[0077] Step S314: Update the image to be detected to the image cache information based on the adjusted inter-frame step distance until the image to be detected meets the preset cache termination condition based on the wheel axle tracking result and the image cache information.

[0078] Steps S313 to S314 above adaptively adjust the inter-frame step distance based on the displacement vector of the target vehicle, which can effectively avoid the occurrence of redundant images and improve the efficiency of subsequent image cache information processing.

[0079] In another embodiment, based on step S320 above, image processing is performed on the image to be detected in the image cache information according to the local features of the wheel axle object in the image cache information, and axle type recognition is performed based on the image processing result to obtain the axle type information of the target vehicle. Specifically, this may include:

[0080] S321, based on the local features of the wheel axle objects in the image cache information, perform image stitching on the image to be detected in the image cache information, and perform axle type recognition based on the image stitching result to obtain the axle type information of the target vehicle.

[0081] In this embodiment, by stitching together the images to be detected from different frames based on the local features of the wheel and axle objects, the complete distribution and arrangement of the wheel and axle objects of the target vehicle can be determined. Furthermore, the axle type information of the target vehicle can be determined based on the stitching result, thereby improving the accuracy of axle type recognition.

[0082] In another embodiment, based on the above step S320, axle shape recognition is performed based on the image processing results to obtain the axle shape information of the target vehicle, which may specifically include the following steps:

[0083] Step S322: Perform wheel axle detection on the image processing result to obtain the first wheel axle detection result.

[0084] Specifically, based on a preset wheel and axle detection algorithm, wheel and axle detection can be performed on the stitched image to obtain all the wheel and axle objects of the target vehicle contained therein. It should be noted that before stitching the images to be detected, each frame of the image to be detected only contains a portion of the wheel and axle objects of the target vehicle, such as a single wheel and axle object. After the image stitching process described above, all the wheel and axle objects of the target vehicle are detected. Figure 5 This is a schematic diagram of the first wheel axle detection results in this embodiment. Figure 5 As shown, each box is a detection box for a wheel axle object, and the object in the detection box is a wheel axle object.

[0085] Step S323: Based on the preset spacing coding rules and the spacing ratio of each wheel axle object contained in the first wheel axle detection result, the spacing of each wheel axle object is coded and arranged, and the axle type information of the target vehicle is determined based on the coding arrangement result.

[0086] Similarly Figure 5 Taking this as an example, the spacing between different axles of the target vehicle varies. Therefore, the spacing ratio of each axle object can be encoded and arranged based on a preset spacing encoding rule. For example, when the ratio between the center point distance of two adjacent axle objects and a preset minimum reference distance is less than a preset ratio, the two adjacent axle objects are considered to be in a close proximity state, and their distance is encoded as 0; otherwise, the two adjacent axle objects are considered to be in a spaced-out state, and their distance is encoded as 1. Based on this, Figure 6 This is a schematic diagram of wheel assembly division. The wheels are then divided according to this spacing coding rule. Figure 5 After encoding the distances between the various wheel axle objects, we can obtain the following: Figure 6 The diagram shows the wheel and axle assembly layout. Among them, Figure 6 Different numerical sequences 1 to 6 represent different axle objects, and the axle spacing code value of the target vehicle can be obtained as 10100. Based on this axle spacing code value, that is, the coding arrangement result, the corresponding axle type information of the target vehicle is mapped out.

[0087] Additionally, in one embodiment, when the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meets the preset caching condition, the image to be detected is updated to the image cache information. Until it is determined, based on the wheel axle tracking result and the image cache information, that the image to be detected meets the preset caching termination condition, the above vehicle axle type recognition method may further include:

[0088] Step S331: Detect the wheel axle objects of the target vehicle in the image to be detected according to the trained detection model to obtain the second wheel axle detection result;

[0089] Step S332: According to the corner distance metric method, associate the second wheel axle detection result with the preset wheel axle tracking list, update the wheel axle tracking list based on the association result, and identify the wheel axle tracking list as the wheel axle tracking result.

[0090] For example, updating the wheel axle tracking list based on the second wheel axle detection result can be performed as follows: After detecting a wheel axle object in the target objects, initialize the node information `node_info` and the list length `L` in the wheel axle tracking list `track_list`, where the node information `node_info` includes the tracking target coordinates `rect`, class, and id information. The second wheel axle detection result is the wheel axle detection target set `axis_od_set`. Associate and match this wheel axle detection target set with the tracking nodes in the wheel axle tracking list using corner distance metric. If the association is successful, use the attribute information of the detection boxes in the wheel axle detection target set, such as coordinates and class, to update the information of the corresponding tracking nodes in the wheel axle tracking list. Treat wheel axle detection boxes in the wheel axle detection target set that do not match tracking nodes as new wheel axle objects, initialize the attribute information of these new wheel axle objects as new tracking nodes, update them in the wheel axle tracking list, and assign a tracking id.

[0091] The above association process can be as follows: Assuming the top-left corner coordinates of the detection box BBox of each wheel axle object in the wheel axle detection target set are (x1, y1) and the bottom-right corner coordinates are (x2, y2), and the top-left corner coordinates of the tracking box BBox_track corresponding to the node in the wheel axle tracking chain are (x3, y3) and the bottom-right corner coordinates are (x4, y4), the corner distance metric is calculated as follows:

[0092] First, determine whether the current detection bounding box (BBox) and the tracking bounding box (BBox_track) have any intersection. The determination method is as follows:

[0093]

[0094] Where max() represents the maximum value and min() represents the minimum value. If x satisfies i ≤x j And y i ≤y j If the detection box BBox and the tracking box BBox_track are in an intersecting state, the corner distance metric is calculated as follows:

[0095]

[0096] Where Δd1 is the distance between the top-left corners of the two boxes, Δd2 is the distance between the bottom-right corners of the two boxes, and Δd is the distance between the top-left corner of the detection box and the bottom-right corner of the tracking box. The correlation between the two boxes is [associativity]. score The calculation formula is as follows:

[0097] associate score =1-(Δd1+Δd2) / Δd (3)

[0098] If associate score If the value is greater than the set threshold, the two are considered to be successfully associated; otherwise, the association fails.

[0099] Further, in one embodiment, based on the above step S332, the wheel axle objects of the target vehicle in the image to be detected are detected according to the trained detection model to obtain a second wheel axle detection result. Specifically, this includes: detecting the wheel axle objects of the target vehicle in the image to be detected according to the trained detection model to obtain an initial detection result of the wheel axle objects; and filtering the initial detection result based on the position information of the wheel axle objects and the confidence level of the initial detection result to obtain a second wheel axle detection result.

[0100] For example, for near-field target vehicles, the center point of the detection bounding box for axle objects is often located in the lower half of the image to be detected. Therefore, the axle object can be filtered by determining whether its position information belongs to a preset region. If the axle object is located in the preset region, the valid state bit of the corresponding detection bounding box is set to 1, and the axle object is identified as a valid object; otherwise, the valid state bit is set to 0, and the axle object is identified as an invalid object. Invalid axle objects will not be updated in the axle tracking results. Alternatively, the detected axle objects can also be filtered based on the confidence level of the detection bounding box to improve the reliability of the second axle detection result.

[0101] In addition, if there is no valid wheel axle object in the effective area of ​​the camera's field of view, the axle type recognition function of the target vehicle is in an idle state, the corresponding idle state bit Idle_flag has a state value of 0, and the cached cumulative frame number frame_cnt is set to 0, and the detection of wheel axle objects continues until a valid wheel axle object is detected.

[0102] In another embodiment, when the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meets the preset caching condition, the image to be detected is updated to the image cache information. Until it is determined based on the wheel axle tracking result and the image cache information that the image to be detected meets the preset caching termination condition, the above vehicle axle type recognition method may further include the following steps:

[0103] Step S341: Based on the wheel axle tracking results, determine a second positional relationship between the first tracked wheel axle object in the image to be detected and multiple preset identifiers. This second positional relationship may include the distance or coordinate magnitude between the center point of the wheel axle object and the multiple preset identifiers. For example, the second positional relationship may be the relationship between the x-coordinate of the center point of the tracking frame of the first tracked wheel axle object and the x-coordinates of the left and right warning lines, respectively.

[0104] Step S342: Based on the second positional relationship, determine the direction of entry of the target vehicle, and select the identifier corresponding to the direction of entry from multiple preset identifiers as a warning identifier; wherein, there is a pre-established correspondence between the multiple preset identifiers and the direction of entry.

[0105] Continuing with the example of the left and right warning lines, if the x-coordinate of the center point of the tracking frame of the first axle object is less than the x-coordinate corresponding to the left warning line, then the target vehicle is entering the camera's field of view from left to right. If the x-coordinate of the center point of the tracking frame of the first axle object is greater than the x-coordinate corresponding to the right warning line, then the target vehicle is entering the camera's field of view from right to left. It can be understood that this embodiment can also determine the direction of entry based on other positional relationships between the axle object and the identifier, such as distance or relative angle, which will not be elaborated upon here.

[0106] The present embodiment will now be described and illustrated through preferred embodiments.

[0107] Figure 7 This is a flowchart of the vehicle axle type recognition method according to a preferred embodiment. Figure 7 As shown, the vehicle axle type recognition method includes the following steps:

[0108] Step S701: Read the image to be detected within the field of view captured by the camera;

[0109] Step S702: Perform wheel axle detection and valid target determination on the image to be detected to obtain the first wheel axle detection result;

[0110] Step S703: Based on the corner distance metric, associate the first wheel axle detection result with the preset wheel axle tracking list, update the wheel axle tracking list based on the association result, and obtain the wheel axle tracking result;

[0111] Step S704: If there is no valid wheel axle object in the effective area of ​​the camera's field of view, the status bit of the vehicle axle type recognition method will be set to 0, and the cumulative frame count of the wheel axle object will be cleared until a valid wheel axle object is detected in the camera's field of view.

[0112] Step S705: If the first tracked wheel axle object touches the warning line, enable the warning line;

[0113] Step S706: Based on the positional relationship between the first tracked wheel axle object and the left and right warning lines, determine the direction of entry of the target vehicle, and determine the vehicle speed of the target vehicle based on the image information before the first tracked wheel axle object touches the warning line, and obtain the displacement vector based on the vehicle speed and the direction of entry.

[0114] Step S707: Enable the buffer signal, start updating the image to be detected to the image buffer information, count the number of buffer frames of the image in the image buffer information until a new wheel axle object touches the warning line, clear the count to zero and start counting again.

[0115] Step S708: Determine whether the number of cached frames has reached the preset frame threshold and whether no new wheel axle objects have been added to the wheel axle tracking list within the preset time period; if yes, proceed to step S709; otherwise, continue to step S707.

[0116] Step S709: Terminate the caching of the image to be detected, and perform image stitching on the image to be detected in the image cache information to obtain the target stitching result;

[0117] Step S710: Perform axle type recognition on the target stitching result to obtain the axle type information of the target vehicle;

[0118] Step S711: Report the axle type, number of axles, target splicing result, and corresponding weighbridge load information of the target vehicle.

[0119] This embodiment also provides a vehicle axle type recognition device, which is used to implement the above embodiments and preferred embodiments, and will not be repeated as described previously. The terms "module," "unit," "subunit," etc., used below can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0120] Figure 8 This is a structural block diagram of the vehicle axle type recognition device 80 in this embodiment, as shown below. Figure 8 As shown, the vehicle axle type recognition device 80 includes: a cache module 82 and a recognition module 84; wherein:

[0121] The caching module 82 is used to update the image to be detected to the image cache information in response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier, which meets the preset caching conditions, until the image to be detected meets the preset caching termination condition based on the wheel axle tracking result and the image cache information; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle;

[0122] The recognition module 84 is used to perform image processing on the image to be detected in the image cache information based on the local features of the wheel axle object in the image cache information, and to perform axle type recognition based on the result of image processing to obtain the axle type information of the target vehicle.

[0123] The aforementioned vehicle axle type recognition device 80, in response to a first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and a preset warning identifier meeting a preset caching condition, updates the image to be detected in the image cache information until, based on the wheel axle tracking result and the image cache information, it determines that the image to be detected meets a preset caching termination condition. The image to be detected is a multi-frame image containing the wheel axle objects of the target vehicle. Based on the local features of the wheel axle objects in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle type recognition is performed based on the image processing result to obtain the axle type information of the target vehicle. It achieves accurate recognition of vehicle axle type based on vehicle images, without relying on manual reading or lidar scanning, thereby reducing the labor and equipment costs for vehicle axle type recognition.

[0124] It should be noted that the above modules can be functional modules or program modules, and can be implemented through software or hardware. For modules implemented through hardware, the above modules can reside in the same processor; or the above modules can be located in different processors in any combination.

[0125] This embodiment also provides an electronic device including a memory and a processor, the memory storing a computer program and the processor being configured to run the computer program to perform the steps in any of the above method embodiments.

[0126] Optionally, the electronic device may further include a transmission device and an input / output device, wherein the transmission device is connected to the processor and the input / output device is connected to the processor.

[0127] Optionally, in this embodiment, the processor can be configured to perform the following steps via a computer program:

[0128] In response to the first positional relationship between the wheel axle tracking result of the target vehicle in the real-time acquired image to be detected and the preset warning identifier meeting the preset caching conditions, the image to be detected is updated to the image cache information until, based on the wheel axle tracking result and the image cache information, it is determined that the image to be detected meets the preset caching termination condition, and the caching of the image to be detected ends; wherein, the image to be detected is a multi-frame image containing the wheel axle object of the target vehicle;

[0129] Based on the local features of the wheel axle objects in the image cache information, image processing is performed on the image to be detected in the image cache information, and axle type recognition is performed based on the image processing results to obtain the axle type information of the target vehicle.

[0130] It should be noted that the specific examples in this embodiment can refer to the examples described in the above embodiments and optional implementations, and will not be repeated in this embodiment.

[0131] Furthermore, in conjunction with the vehicle axle type recognition method provided in the above embodiments, this embodiment can also provide a storage medium for implementation. This storage medium stores a computer program; when executed by a processor, the computer program implements any of the vehicle axle type recognition methods described in the above embodiments.

[0132] It should be noted that the information and data involved in this application (including but not limited to data used for analysis, stored data, and displayed data) comply with relevant laws and regulations.

[0133] Obviously, the accompanying drawings are merely some examples or embodiments of this application. Those skilled in the art can apply this application to other similar situations based on these drawings without any creative effort. Furthermore, it is understood that although the work done in this development process may be complex and lengthy, for those skilled in the art, certain design, manufacturing, or production modifications made based on the technical content disclosed in this application are merely conventional technical means and should not be considered as insufficient disclosure of this application.

[0134] The term "embodiment" in this application refers to a specific feature, structure, or characteristic described in connection with an embodiment that may be included in at least one embodiment of this application. The appearance of this phrase in various places in the specification does not necessarily imply the same embodiment, nor does it imply that it is mutually exclusive with or independent of other embodiments. It will be clearly or implicitly understood by those skilled in the art that the embodiments described in this application may be combined with other embodiments without conflict.

[0135] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of patent protection. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the appended claims.

Claims

1. A vehicle axle type identification method characterized by, The method comprises: in response to a first positional relationship between a wheel axle tracking result of a target vehicle in a real-time acquired to-be-detected image and a preset early warning identifier meeting a preset caching condition, updating the to-be-detected image into image caching information until it is determined, based on the wheel axle tracking result and the image caching information, that the to-be-detected image meets a preset caching termination condition; wherein the to-be-detected image is a plurality of images containing a wheel axle object of the target vehicle; the preset caching condition comprises that a first wheel axle object tracked from the to-be-detected image touches the early warning identifier; and the preset caching termination condition comprises that no new wheel axle object is tracked from the to-be-detected image within a preset time period; performing image processing on the to-be-detected image in the image caching information according to local features of the wheel axle object in the image caching information, and performing axle type identification based on a result of the image processing to obtain axle type information of the target vehicle.

2. The vehicle axle type identification method according to claim 1, characterized by, The updating of the to-be-detected image into the image caching information comprises: in the process of updating the to-be-detected image into the image caching information, counting a caching frame number of images in the image caching information until it is determined, based on the wheel axle tracking result, that there is a new wheel axle object touching the early warning identifier, clearing and re-counting the caching frame number.

3. The vehicle axle type identification method according to claim 2, characterized in that, The method further comprises: in response to the wheel axle tracking result not updating a new wheel axle object within a preset time period and the caching frame number of images in the image caching information reaching a preset frame number threshold, confirming that the to-be-detected image meets the preset caching termination condition.

4. The vehicle axle type identification method according to claim 1, characterized by, The updating of the to-be-detected image into the image caching information until it is determined, based on the wheel axle tracking result and the image caching information, that the to-be-detected image meets a preset caching termination condition, comprises: in response to a first positional relationship between a wheel axle tracking result of a target vehicle in a real-time acquired to-be-detected image and a preset early warning identifier meeting a preset caching condition, adjusting an inter-frame step distance of updating image caching information according to a displacement vector of the target vehicle; updating the to-be-detected image into the image caching information based on the adjusted inter-frame step distance until it is determined, based on the wheel axle tracking result and the image caching information, that the to-be-detected image meets a preset caching termination condition.

5. The vehicle axle type identification method of claim 1, wherein The performing of image processing on the to-be-detected image in the image caching information according to local features of the wheel axle object in the image caching information, and the performing of axle type identification based on a result of the image processing to obtain axle type information of the target vehicle, comprises: performing image stitching on the to-be-detected image in the image caching information according to local features of the wheel axle object in the image caching information, and performing axle type identification based on a result of the image stitching to obtain axle type information of the target vehicle.

6. The vehicle axle type identification method of claim 1, wherein The performing of axle type identification based on a result of the image processing to obtain axle type information of the target vehicle, comprises: perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; 7. The vehicle axle type identification method of claim 1, wherein perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; 8. The vehicle axle type identification method according to claim 7, characterized by perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; 9. The vehicle axle type identification method according to any one of claims 1 to 8, characterized by, perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; 10. A vehicle axle type identification apparatus characterized by comprising: perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of the image processing to obtain a first wheel axle detection result; perform wheel axle detection on a result of The cache module is configured to update the to-be-detected image into image cache information in response to a first positional relationship between a wheel axle tracking result of a target vehicle in the to-be-detected image acquired in real time and a preset early warning identifier meeting a preset cache condition, until it is determined that the to-be-detected image meets a preset cache termination condition based on the wheel axle tracking result and the image cache information; the to-be-detected image is a plurality of images containing a wheel axle object of the target vehicle; the preset cache condition includes that a first wheel axle object tracked from the to-be-detected image reaches the early warning identifier; and the preset cache termination condition includes that no new wheel axle object is tracked from the to-be-detected image within a preset time period. The identification module is configured to perform image processing on the to-be-detected image in the image cache information according to a local feature of the wheel axle object in the image cache information, and perform axle type identification based on a result of the image processing to obtain axle type information of the target vehicle. 11.An electronic device comprising a memory and a processor, the electronic device characterized by, The memory stores a computer program, and the processor is configured to execute the computer program to perform the vehicle axle type identification method in any one of claims 1 to 9.

12. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to implement the steps of the vehicle axle type identification method in any one of claims 1 to 9.

Images