Wheel chock recognition method and device, vehicle and storage medium

Abstract

This disclosure relates to a wheel chock identification method, apparatus, vehicle, and storage medium, relating to the field of autonomous driving. The method includes: determining identification information of a target wheel chock in a first image and storage location information characterizing the location of the target wheel chock, wherein the first image is a frame from an image sequence obtained through image acquisition, and the identification information is obtained by tracking the target wheel chock; acquiring association information indicating the wheel chock identification associated with each storage location; determining whether the identification information of the target wheel chock and the storage location information conform to the association information; and correcting the identification information of the target wheel chock if they do not conform to the association information. Thus, by using the association information between the storage location and the wheel chock as auxiliary information to verify the accuracy of the identification information obtained through initial tracking, and correcting it in a timely manner, the accuracy of wheel chock tracking is improved.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving technology, and in particular to wheel chock recognition methods, devices, vehicles, and storage media. Background Technology

[0002] Multi-object tracking refers to identifying moving objects within a given image sequence and mapping these moving objects across different frames to determine their trajectories. In the field of autonomous driving, multi-object tracking is a crucial component, such as tracking wheel chocks. Currently, wheel chock tracking typically employs optical flow or Kalman tracing methods. However, these methods are susceptible to inaccurate identification and poor performance due to factors such as varying lighting conditions and nonlinear vehicle motion. Summary of the Invention

[0003] To overcome the problems existing in the related technologies, this disclosure provides a wheel chock identification method, device, vehicle, and storage medium.

[0004] According to a first aspect of the present disclosure, a wheel chock identification method is provided, the method comprising:

[0005] The identification information of the target wheel stopper in the first image and the storage location information used to characterize the storage location of the target wheel stopper are determined. The first image is a frame image in an image sequence obtained through image acquisition. The identification information is obtained by tracking the target wheel stopper.

[0006] Obtain the association information used to indicate the wheel stop identifier associated with each storage location;

[0007] Determine whether the identification information and storage location information of the target wheel chock match the associated information;

[0008] If the information does not match the association information, the identification information of the target wheel chock is corrected.

[0009] Optionally, the identification information of the target wheel chock is obtained in the following way:

[0010] Determine the detection position information of each wheel stopper in the first image;

[0011] Obtain the predicted position information of each wheel stopper generated based on the second image, where the second image is the frame preceding the first image in the image sequence;

[0012] Determine the target predicted position information that matches the detected position information of the target wheel chock;

[0013] The wheel chock identifier corresponding to the predicted target location information is determined as the identification information of the target wheel chock.

[0014] Optionally, the predicted location information is obtained in the following way:

[0015] Wheel stops are detected in the second image to obtain detection results, which are used to indicate the position of each wheel stop in the second image;

[0016] Using the Kalman filter algorithm, the position of each stopper in the second image is predicted in the next frame of the second image, thus obtaining the predicted position information.

[0017] Optionally, determining the target predicted position information that matches the detection position information of the target wheel chock includes:

[0018] For each of the predicted position information, the relative distance between the target wheel stop and the wheel stop corresponding to the predicted position information is determined based on the detected position information of the target wheel stop.

[0019] The predicted position information corresponding to the smallest relative distance is determined as the target predicted position information that matches the detection position information of the target wheel chock.

[0020] Optionally, determining whether the identification information and storage location information of the target wheel chock match the associated information includes:

[0021] Determine the target wheel stop identifier associated with the target storage location indicated by the target wheel stop storage location information in the associated information;

[0022] Determine whether the identification information of the target wheel chock matches the target wheel chock identifier;

[0023] If the identification information of the target wheel stopper matches the identification of the target wheel stopper, it is determined that the identification information of the target wheel stopper and the storage location information conform to the associated information;

[0024] If the identification information of the target wheel stop does not match the identification of the target wheel stop, it is determined that the identification information of the target wheel stop and the storage location information do not conform to the associated information.

[0025] Optionally, the step of determining the target wheel chock identifier associated with the target storage location indicated by the target wheel chock's storage location information in the associated information includes:

[0026] Determine the number of wheel stops contained in the target storage location;

[0027] If the target storage location contains more than one wheel chock, determine the distribution location of the target wheel chock in the target storage location;

[0028] Among the wheel stopper identifiers associated with the target storage location in the associated information, the wheel stopper identifier corresponding to the wheel stopper that matches the distribution location is determined as the target wheel stopper identifier.

[0029] Optionally, the step of correcting the identification information of the target wheel chock includes:

[0030] The identification information of the target wheel chock is corrected to the target wheel chock identifier.

[0031] According to a second aspect of the present disclosure, a wheel chock identification device is provided, the device comprising:

[0032] The first determining module is configured to determine the identification information of the target wheel stopper in the first image and the storage location information used to characterize the storage location of the target wheel stopper. The first image is a frame image in an image sequence obtained through image acquisition, and the identification information is obtained by tracking the target wheel stopper.

[0033] The first acquisition module is configured to acquire association information that indicates the wheel stop identifier associated with each storage location;

[0034] The second determining module is configured to determine whether the identification information and storage location information of the target wheel stopper match the associated information;

[0035] The correction module is configured to correct the identification information of the target wheel chock if it does not conform to the associated information.

[0036] According to a third aspect of the present disclosure, a vehicle is provided, comprising:

[0037] First processor;

[0038] A first memory for storing instructions executable by a first processor;

[0039] The first processor is configured to execute instructions in the first memory to implement the steps of the wheel chock identification method provided in the first aspect of this disclosure.

[0040] According to a fourth aspect of the present disclosure, a computer-readable storage medium is provided, having stored thereon computer program instructions that, when executed by a second processor, implement the steps of the wheel chock identification method provided in the first aspect of the present disclosure.

[0041] The technical solutions provided by the embodiments of this disclosure may include the following beneficial effects:

[0042] The process involves determining the identification information of the target wheel chock in the first image and the storage location information representing the location of the target wheel chock. It also involves acquiring association information indicating the wheel chock identification associated with each storage location. The process then determines whether the target wheel chock identification information and storage location information match the association information. If they do not match, the target wheel chock identification information is corrected. The first image is a frame from an image sequence obtained through image acquisition, and the identification information is obtained by tracking the target wheel chock. Therefore, while tracking the wheel chock, the association information between the storage location and the wheel chock is acquired as auxiliary information to verify the accuracy of the identification information obtained from the initial tracking. Corrections are made promptly when problems exist in the initial tracking results, thereby improving the accuracy of wheel chock tracking. In autonomous driving scenarios, this approach can improve the accuracy of downstream mapping, thus enhancing the implementation of functions such as automatic parking.

[0043] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description

[0044] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0045] Figure 1 This is a flowchart illustrating a wheel chock identification method according to an exemplary embodiment.

[0046] Figure 2 This is a block diagram illustrating a wheel chock identification device according to an exemplary embodiment.

[0047] Figure 3 This is a functional block diagram of a vehicle illustrating an exemplary embodiment. Detailed Implementation

[0048] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.

[0049] It should be noted that all actions involving the acquisition of signals, information, or data in this application are carried out in compliance with the relevant data protection laws and policies of the country where the application is located, and with the authorization granted by the owner of the relevant device.

[0050] Figure 1 This is a flowchart illustrating a wheel chock recognition method according to an exemplary embodiment. This wheel chock recognition method can be applied to scenarios such as autonomous driving, automatic parking, and robot navigation. Figure 1 As shown, the wheel chock identification method provided in this disclosure may include steps 11 to 14.

[0051] In step 11, the identification information of the target wheel chock in the first image and the storage location information used to characterize the storage location of the target wheel chock are determined.

[0052] The first image is a frame from an image sequence obtained through image acquisition. When the wheel chock recognition method provided in this disclosure is applied to an autonomous driving scenario or an automatic parking scenario, the image sequence can be a series of images captured during the driving process of the vehicle through an image acquisition device (e.g., a camera) installed on the vehicle.

[0053] It is worth noting that, typically, the images in an image sequence are taken of a parking area (e.g., a parking lot), therefore, the images in the image sequence usually contain multiple wheel chocks. Thus, in most cases, the first image contains multiple wheel chocks. In such cases, each wheel chock in the first image can be used as a target wheel chock, and the wheel chock recognition method provided in this disclosure can be executed, thereby enabling accuracy verification for each wheel chock. Furthermore, in some cases, if only a specific wheel chock is of interest, then only that specific wheel chock needs to be used as the target wheel chock, thus enabling accuracy verification for that specific wheel chock.

[0054] The identification information of the target wheel chock can be obtained by tracking the target wheel chock.

[0055] In one possible implementation, the identification information of the target wheel chock in step 11 can be obtained in the following way:

[0056] Determine the detection position information of each wheel stopper in the first image;

[0057] Obtain the predicted position information of each wheel stopper generated based on the second image;

[0058] Determine the target predicted position information that matches the detection position information of the target wheel chock;

[0059] The wheel chock identifier corresponding to the predicted target location information is determined as the identifier information of the target wheel chock.

[0060] The second image is the frame preceding the first image in the image sequence.

[0061] In one possible implementation, the detection position information of each wheel chock in the first image can be obtained by recognizing the wheel chocks in the first image using an image recognition model (or algorithm).

[0062] For example, multiple sets of sample data can be obtained for training a wheel chock recognition model. Each set of sample data may include sample images and annotation information for the wheel chocks in the sample images. Then, the model is trained by using the sample images as input and the annotation information as the target output, resulting in a trained wheel chock recognition model. The wheel chock recognition model trained in this way has the ability to recognize (or detect) wheel chocks in an image. Therefore, a first image can be input into the wheel chock recognition model, and the detection position information of each wheel chock in the first image can be obtained through the output of the wheel chock recognition model.

[0063] In one possible implementation, the predicted position information of each wheel stopper generated based on the second image can be obtained in the following way:

[0064] Wheel block detection is performed on the second image to obtain the detection results;

[0065] Using the Kalman filter algorithm, the position of each wheel stopper in the second image is predicted in the next frame of the second image, thus obtaining the predicted position information.

[0066] The detection results are used to indicate the position of each wheel chock in the second image. The method for detecting wheel chocks in the second image can be the same as that for determining the detection position information of each wheel chock in the first image, as described above. The principle is the same: to identify and detect wheel chocks within the image, and will not be repeated here.

[0067] After obtaining the detection results of the second image, the Kalman filter algorithm can be used to predict the position of each wheel stopper in the next frame of the second image.

[0068] In this disclosure, a wheel stop can be represented by three points: the two endpoints of the wheel stop and the midpoint of the wheel stop. Accordingly, the position information of the wheel stop (whether it is detected position information or predicted position information) can include the positions of the two endpoints of the wheel stop and the position of the midpoint of the wheel stop.

[0069] Initially, state variables (12 variables) can be constructed based on the coordinates of the three points in the image coordinate system (i.e., the two-dimensional coordinate system corresponding to the second image) and their respective coordinate changes in the horizontal and vertical directions. A corresponding state transition matrix (12*12 matrix) is also set. Then, based on the constructed state variables and state transition matrix, the Kalman filter algorithm is used to predict the position of each wheel stop in the second image in the next frame image and generate the predicted position information.

[0070] After determining the detection position information of each wheel chock in the first image and generating the predicted position information of each wheel chock based on the second image, the wheel chocks in the first and second images can be matched based on the detection and predicted position information. Typically, the predicted position information of the wheel chock in the next frame image obtained from the second image should be close to the detection position information of that wheel chock in the first image (i.e., the frame image following the actually acquired second image). Based on this approach, a one-to-one matching of wheel chocks in the first and second images can be achieved.

[0071] In one possible implementation, determining the target predicted position information that matches the detected position information of the target wheel chock may include the following steps:

[0072] For each predicted position information, the relative distance between the target wheel stop and the wheel stop corresponding to the predicted position information is determined based on the detected position information of the target wheel stop.

[0073] The predicted position information corresponding to the smallest relative distance is determined as the target predicted position information that matches the detection position information of the target wheel chock.

[0074] For example, if a target wheel chock is represented by three points—a first point, a second point, and a third point—corresponding to its left endpoint, middle point, and right endpoint, respectively, and the detected position information of the target wheel chock includes the first coordinate of the first point, the second coordinate of the second point, and the third coordinate of the third point, and a certain predicted position information includes the fourth coordinate corresponding to the left endpoint of the wheel chock, the fifth coordinate corresponding to the middle point of the wheel chock, and the sixth coordinate corresponding to the right endpoint of the wheel chock, and the fourth and sixth coordinates corresponding to the endpoints of the wheel chock can form a target straight line, then the relative distance between the target wheel chock and the wheel chock corresponding to the predicted position information can be determined in the following manner:

[0075] Determine the distance from the first point's first coordinate to the target line, and use this distance as the first distance;

[0076] Determine the distance from the second point's second coordinate to the target line, and use this distance as the second distance;

[0077] Determine the distance from the third point's third coordinate to the target line, and use this distance as the third distance.

[0078] Determine the relative distance between the x-coordinate of the first coordinate and the x-coordinate of the fourth coordinate, and use this distance as the fourth distance;

[0079] Determine the relative distance between the x-coordinate of the second coordinate and the x-coordinate of the fifth coordinate, and use this distance as the fifth distance;

[0080] Determine the relative distance between the x-coordinate of the third coordinate and the x-coordinate of the sixth coordinate, and use this distance as the sixth distance;

[0081] The longitudinal relative distance is determined based on the first distance, the second distance, and the third distance;

[0082] The lateral relative distance is determined based on the fourth, fifth, and sixth distances;

[0083] Based on the longitudinal and lateral relative distances, the relative distance between the target wheel stop and the wheel stop corresponding to the predicted position information is determined.

[0084] According to the calculation method described above, the first, second, and third distances are actually the longitudinal distance changes of the wheel chocks. Therefore, the longitudinal relative distance can be determined using the first, second, and third distances. Similarly, the fourth, fifth, and sixth distances are actually the lateral distance changes of the wheel chocks. Therefore, the lateral relative distance can be determined using the fourth, fifth, and sixth distances. For example, the maximum value among the first, second, and third distances can be used as the longitudinal relative distance. As another example, the maximum value among the fourth, fifth, and sixth distances can be used as the lateral relative distance. Furthermore, the sum of the longitudinal and lateral relative distances can be used as the relative distance between the target wheel chock and the wheel chock corresponding to the predicted position information.

[0085] Using this method, the relative distance between wheel stops is determined from both the horizontal and vertical dimensions, ensuring that the relative distance more accurately reflects the relative position of the wheel stops.

[0086] Following the above method, the relative distance between the target wheel chock and each wheel chock corresponding to the predicted position information can be determined. Furthermore, the predicted position information corresponding to the smallest relative distance can be determined as the target predicted position information that matches the detected position information of the target wheel chock.

[0087] The predicted position information is generated one by one based on each wheel stop in the second image. Therefore, if the wheel stop identifiers in the second image are known, the wheel stop identifiers corresponding to each predicted position information can also be determined. Thus, if the wheel stop identifiers corresponding to the target predicted position information are known, the wheel stop identifiers corresponding to the target predicted position information can be determined as the identifier information of the target wheel stop.

[0088] It should be noted that if the second image and the first image are the first two frames of the image sequence (i.e., the second image is the first frame and the first image is the second frame), then since the wheel stop identifiers in the second image have not yet been generated, the wheel stop identifiers corresponding to the target predicted position information cannot be directly obtained. In this case, based on the matching of wheel stops in the first and second images, the same wheel stop ID can be assigned to two matching wheel stops to determine the wheel stop identifiers of each wheel stop. In subsequent wheel stop identification processes, the wheel stop identifiers assigned in this process can be used, which corresponds to the situation where the wheel stop identifiers corresponding to the target predicted position information are known.

[0089] The method described above actually yields the identification information of the target wheel chock by tracking it.

[0090] Based on the first image, storage location information representing the location of the target wheel chock can also be determined. For the first image, wheel chocks and storage locations can be identified. Based on the relative positions of the wheel chocks and storage locations, the storage location of each wheel chock can be determined. For example, if a wheel chock is located within the area covered by a certain storage location, it means that the wheel chock is within that storage location. Based on this, the storage location of the target wheel chock can be determined, i.e., the storage location information corresponding to the target wheel chock can be obtained. The identification of storage locations in the image can utilize methods such as storage location identification and storage location tracking to identify which storage location appears in the image (e.g., distinguishing different storage locations through different storage location identifiers). The principle is similar to that of wheel chock identification and tracking, and will not be elaborated further here.

[0091] In step 12, the association information used to indicate the wheel stop identifier associated with each storage location is obtained.

[0092] The association information indicates the wheel stop identifier associated with each storage location, thus indicating the wheel stop in each storage location. The association information can be determined based on the tracking of wheel stops and storage locations prior to the first image. For example, based on the tracking results of wheel stops and storage locations in the first image and previous images in the image sequence, the wheel stop identifier of each wheel stop and the storage location identifier of each storage location can be determined. At the same time, based on the relative positional relationship between the wheel stop and the storage location, the wheel stops located within the coverage area of ​​the storage location can be identified, and the wheel stop identifier of such wheel stops can be associated with the storage location identifier of that storage location. In this way, for each storage location, the corresponding wheel stop identifier is associated, thereby forming the aforementioned association information.

[0093] In step 13, it is determined whether the identification information and storage location information of the target wheel stopper match the associated information.

[0094] In one possible implementation, step 13 may include the following steps:

[0095] Determine the target wheel chock identifier associated with the target storage location indicated by the target wheel chock's storage location information in the associated information;

[0096] Determine whether the identification information of the target wheel chock matches the target wheel chock identification;

[0097] If the identification information of the target wheel stopper matches the target wheel stopper identification, it is determined that the identification information of the target wheel stopper and the storage location information are consistent with the associated information.

[0098] If the identification information of the target wheel stop does not match the target wheel stop identification, it is determined that the identification information of the target wheel stop and the storage location information do not match the associated information.

[0099] In one possible embodiment, step 13, determining the target wheel chock identifier associated with the target storage location indicated by the target wheel chock's storage location information in the association information, may include the following steps:

[0100] Determine the number of wheel stops contained in the target storage location;

[0101] If the target storage location contains more than one wheel chock, determine the distribution location of the target wheel chocks within the target storage location;

[0102] Among the wheel stop identifiers associated with the target storage location in the associated information, the wheel stop identifier corresponding to the wheel stop that matches the distribution location is determined as the target wheel stop identifier.

[0103] For example, the number of wheel stops contained in the target storage location can be determined as follows:

[0104] Determine the first lateral length of the vehicle corresponding to the target storage location;

[0105] Determine the second length of the target wheel stop;

[0106] The number of wheel stops contained in the target storage location is determined based on the ratio of the second length to the first length.

[0107] The function of wheel chocks is to ensure that when a vehicle is parked, the tires contact the wheel chock levers, allowing the vehicle to park accurately and neatly in the parking space. Generally, a parking space may contain one or two wheel chocks. Therefore, a parking space with one wheel chock usually has a longer wheel chock, while a parking space with two wheel chocks will have shorter wheel chocks. Thus, the length of the wheel chocks and the parking space's lateral length relative to the vehicle's length reflect the number of wheel chocks included in the parking space.

[0108] If the ratio of the second length to the first length is less than or equal to 0.5, it means that the second length of the target wheel stop does not exceed half of the first length of the target storage location. Therefore, the target storage location contains 2 wheel stops. If the ratio of the second length to the first length is greater than 0.5, it means that the second length of the target wheel stop exceeds half of the first length of the target storage location. Therefore, the target storage location contains 1 wheel stop.

[0109] If the target storage location contains only one wheel stopper, the wheel stopper identifier associated with the target storage location information can be directly used as the target wheel stopper identifier.

[0110] If the target storage location contains more than one wheel chock, it is necessary to further determine which wheel chock identifier associated with the target storage location should be matched with the wheel chock identifier of the target wheel chock. Therefore, if the target storage location contains more than one wheel chock, the distribution location of the target wheel chock within the target storage location can be determined first. This distribution location could include, for example, the left rear or right rear. Then, based on the determined distribution location, the wheel chock identifier corresponding to the wheel chock identifier that matches the distribution location can be identified as the target wheel chock identifier from the wheel chock identifiers associated with the target storage location in the association information.

[0111] Using the above method, when the target storage location contains more than one wheel chock, the wheel chock identifier used to match the identification information of the target wheel chock is determined by combining the position of the target wheel chock, which helps to verify the identification information of the target wheel chock more accurately.

[0112] After identifying the target wheel stop identifier, the system compares the target wheel stop identifier information with the target wheel stop identifier itself to determine whether the target wheel stop identifier information and the storage location information conform to the associated information. For example, if the target wheel stop identifier information matches the target wheel stop identifier, it can be determined that the target wheel stop identifier information matches the target wheel stop identifier, and therefore, the target wheel stop identifier information and the storage location information conform to the associated information; if the target wheel stop identifier information does not match the target wheel stop identifier, it can be determined that the target wheel stop identifier information and the storage location information do not conform to the associated information.

[0113] In step 14, if the correlation information is not met, the identification information of the target wheel chock is corrected.

[0114] If the identification information and storage location information of the target wheel stop do not match the associated information, it indicates that an error occurred when tracking the first image during the tracking process of the target wheel stop. Therefore, it is necessary to correct the identification information of the target wheel stop to ensure the accuracy of the tracking.

[0115] In one possible implementation, correcting the identification information of the target wheel chock may include the following steps:

[0116] Correct the identification information of the target wheel chock to the target wheel chock identifier.

[0117] During the above processing, it can be determined that the target wheel chock identifier determined based on the associated information is the identifier information that the target wheel chock should correspond to. Therefore, if the identifier information of the target wheel chock is incorrect, the identifier information of the target wheel chock can be corrected using the target wheel chock identifier, that is, the identifier information of the target wheel chock is corrected to the target wheel chock identifier.

[0118] If the identification information and storage location information of the target wheel chock match the associated information, it means that the tracking of the target wheel chock is still accurate. Therefore, no correction is needed, and the identification information of the target wheel chock can be maintained.

[0119] The above technical solution determines the identification information of the target wheel chock in the first image and the storage location information representing the location of the target wheel chock. It also obtains the association information indicating the wheel chock identification associated with each storage location. Furthermore, it determines whether the identification information of the target wheel chock and the storage location information conform to the association information. If they do not conform, the identification information of the target wheel chock is corrected. The first image is a frame from an image sequence obtained through image acquisition, and the identification information is obtained by tracking the target wheel chock. Therefore, while tracking the wheel chock, the association information between the storage location and the wheel chock is obtained as auxiliary information to verify the accuracy of the identification information obtained from the initial tracking. Corrections are made promptly when there are problems with the initial tracking results, thereby improving the accuracy of wheel chock tracking. In autonomous driving scenarios, this can improve the accuracy of downstream mapping, thereby enhancing the implementation of functions such as automatic parking.

[0120] Figure 2 This is a block diagram illustrating a wheel chock recognition device according to an exemplary embodiment. (Refer to...) Figure 2 The device 20 includes:

[0121] The first determining module 21 is configured to determine the identification information of the target wheel stopper in the first image and the storage location information used to characterize the storage location of the target wheel stopper. The first image is a frame image in an image sequence obtained through image acquisition, and the identification information is obtained by tracking the target wheel stopper.

[0122] The first acquisition module 22 is configured to acquire association information for indicating the wheel stop identifier associated with each storage location;

[0123] The second determining module 23 is configured to determine whether the identification information and storage location information of the target wheel stopper match the associated information;

[0124] The correction module 24 is configured to correct the identification information of the target wheel chock if it does not conform to the associated information.

[0125] Optionally, the identification information of the target wheel chock is obtained through the following module:

[0126] The third determining module is configured to determine the detection position information of each wheel stopper in the first image;

[0127] The second acquisition module is configured to acquire the predicted position information of each wheel stopper generated based on the second image, wherein the second image is the frame image preceding the first image in the image sequence;

[0128] The fourth determining module is configured to determine the target predicted position information that matches the detection position information of the target wheel chock;

[0129] The fifth determining module is configured to determine the wheel chock identifier corresponding to the predicted target location information as the identifier information of the target wheel chock.

[0130] Optionally, the predicted location information is obtained through the following modules:

[0131] The detection module is configured to detect wheel stops in the second image and obtain detection results, which are used to indicate the position of each wheel stop in the second image;

[0132] The prediction module is configured to use a Kalman filter algorithm to predict the position of each wheel stopper in the second image in the next frame of the second image, thereby obtaining the predicted position information.

[0133] Optionally, the fourth determining module includes:

[0134] The first determining submodule is configured to, for each of the predicted position information, determine the relative distance between the target wheel stopper and the wheel stopper corresponding to the predicted position information based on the detected position information of the target wheel stopper;

[0135] The second determining submodule is configured to determine the predicted position information corresponding to the minimum relative distance as the target predicted position information that matches the detection position information of the target wheel chock.

[0136] Optionally, the second determining module 23 includes:

[0137] The third determining submodule is configured to determine the target wheel stop identifier associated with the target storage location indicated by the storage location information of the target wheel stop in the associated information;

[0138] The fourth determining submodule is configured to determine whether the identification information of the target wheel chock matches the target wheel chock identifier;

[0139] The fifth determining submodule is configured to determine that the identification information and storage location information of the target wheel stopper match the identification information of the target wheel stopper if the identification information of the target wheel stopper matches the identification information of the target wheel stopper.

[0140] The sixth determination submodule is configured to determine that if the identification information of the target wheel stopper does not match the identification of the target wheel stopper, the identification information of the target wheel stopper and the storage location information do not match the associated information.

[0141] Optionally, the third determining submodule includes:

[0142] The seventh determining submodule is configured to determine the number of wheel stops contained in the target storage location;

[0143] The eighth determining submodule is configured to determine the distribution position of the target wheel stops in the target storage location if the number of wheel stops contained in the target storage location is more than one.

[0144] The ninth determining submodule is configured to determine the wheel stopper identifier corresponding to the wheel stopper that matches the distribution location from the wheel stopper identifiers associated with the wheel stopper in the associated information of the target storage location as the target wheel stopper identifier.

[0145] Optionally, the correction module 24 is configured to correct the identification information of the target wheel chock to the target wheel chock identifier.

[0146] Regarding the apparatus in the above embodiments, the specific manner in which each module performs its operation has been described in detail in the embodiments related to the method, and will not be elaborated upon here.

[0147] Figure 3 This is a block diagram illustrating a vehicle 600 according to an exemplary embodiment. For example, vehicle 600 can be a hybrid vehicle, a non-hybrid vehicle, an electric vehicle, a fuel cell vehicle, or other types of vehicle. Vehicle 600 can be an autonomous vehicle, a semi-autonomous vehicle, or a non-autonomous vehicle.

[0148] Reference Figure 3The vehicle 600 may include various subsystems, such as an infotainment system 610, a perception system 620, a decision control system 630, a drive system 640, and a computing platform 650. The vehicle 600 may also include more or fewer subsystems, and each subsystem may include multiple components. Furthermore, each subsystem and each component of the vehicle 600 can be interconnected via wired or wireless means.

[0149] In some embodiments, the infotainment system 610 may include a communication system, an entertainment system, and a navigation system, etc.

[0150] The perception system 620 may include several sensors for sensing information about the environment surrounding the vehicle 600. For example, the perception system 620 may include a global positioning system (which may be GPS, BeiDou, or other positioning systems), an inertial measurement unit (IMU), lidar, millimeter-wave radar, ultrasonic radar, and a camera device.

[0151] The decision control system 630 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0152] The drive system 640 may include components that provide powered motion to the vehicle 600. In one embodiment, the drive system 640 may include an engine, an energy source, a transmission system, and wheels. The engine may be one or a combination of internal combustion engines, electric motors, and compressed air engines. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0153] Some or all of the functions of vehicle 600 are controlled by computing platform 650. Computing platform 650 may include at least one first processor 651 and a first memory 652, the first processor 651 being able to execute instructions 653 stored in the first memory 652.

[0154] The first processor 651 can be any conventional processor, such as a commercially available CPU. The processor may also include, for example, a Graphics Processing Unit (GPU), a Field Programmable Gate Array (FPGA), a System on Chip (SOC), an Application Specific Integrated Circuit (ASIC), or a combination thereof.

[0155] The first memory 652 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0156] In addition to instruction 653, the first memory 652 can also store data, such as road maps, route information, vehicle position, direction, speed, and other data. The data stored in the first memory 652 can be used by the computing platform 650.

[0157] In this embodiment of the disclosure, the first processor 651 may execute instruction 653 to complete all or part of the steps of the wheel chock identification method described above.

[0158] This disclosure also provides a computer-readable storage medium having computer program instructions stored thereon, which, when executed by a second processor, implement the steps of the wheel chock identification method provided in this disclosure.

[0159] In another exemplary embodiment, a computer program product is also provided, the computer program product comprising a computer program executable by a programmable device, the computer program having a code portion for performing the wheel chock identification method described above when executed by the programmable device.

[0160] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of this disclosure. This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.

[0161] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.

Claims

1. A method for identifying wheel chocks, characterized in that, The method includes: The identification information of the target wheel stopper in the first image and the storage location information used to characterize the storage location of the target wheel stopper are determined. The first image is a frame image in an image sequence obtained through image acquisition, and the identification information is an ID obtained by tracking the target wheel stopper. Obtain association information to indicate the wheel stop identifier associated with each storage location, wherein the wheel stop identifier is the wheel stop ID; Determine whether the identification information and storage location information of the target wheel chock match the associated information; If the correlation information is not met, the identification information of the target wheel chock shall be corrected. The associated information is formed in the following ways: Based on the tracking results of the wheel stops and storage locations in the first image and previous images in the image sequence, the wheel stop identifier of each wheel stop and the storage location identifier of each storage location are determined. Based on the relative positional relationship between the wheel stopper and the storage location, determine the wheel stopper whose location is within the coverage area of ​​the storage location; For any storage location, the wheel stop identifier corresponding to the wheel stop within the coverage area of ​​that storage location is determined as the wheel stop identifier associated with that storage location.

2. The method according to claim 1, characterized in that, The identification information of the target wheel chock is obtained in the following way: Determine the detection position information of each wheel stopper in the first image; Obtain the predicted position information of each wheel stopper generated based on the second image, where the second image is the frame preceding the first image in the image sequence; Determine the target predicted position information that matches the detected position information of the target wheel chock; The wheel chock identifier corresponding to the predicted target location information is determined as the identification information of the target wheel chock.

3. The method according to claim 2, characterized in that, The predicted location information is obtained through the following methods: Wheel stops are detected in the second image to obtain detection results, which are used to indicate the position of each wheel stop in the second image; Using the Kalman filter algorithm, the position of each stopper in the second image is predicted in the next frame of the second image, thus obtaining the predicted position information.

4. The method according to claim 2, characterized in that, The determination of the target predicted position information that matches the detection position information of the target wheel chock includes: For each of the predicted position information, the relative distance between the target wheel stop and the wheel stop corresponding to the predicted position information is determined based on the detected position information of the target wheel stop. The predicted position information corresponding to the smallest relative distance is determined as the target predicted position information that matches the detection position information of the target wheel chock.

5. The method according to claim 1, characterized in that, Determining whether the identification information and storage location information of the target wheel chock match the associated information includes: Determine the target wheel stop identifier associated with the target storage location indicated by the target wheel stop storage location information in the associated information; Determine whether the identification information of the target wheel chock matches the target wheel chock identifier; If the identification information of the target wheel stopper matches the identification of the target wheel stopper, it is determined that the identification information of the target wheel stopper and the storage location information conform to the associated information; If the identification information of the target wheel stop does not match the identification of the target wheel stop, it is determined that the identification information of the target wheel stop and the storage location information do not conform to the associated information.

6. The method according to claim 5, characterized in that, The determination of the target wheel chock location information indicating the target storage location is associated with the target wheel chock identifier in the associated information, including: Determine the number of wheel stops contained in the target storage location; If the target storage location contains more than one wheel chock, determine the distribution location of the target wheel chock in the target storage location; Among the wheel stopper identifiers associated with the target storage location in the associated information, the wheel stopper identifier corresponding to the wheel stopper that matches the distribution location is determined as the target wheel stopper identifier.

7. The method according to claim 5, characterized in that, The step of correcting the identification information of the target wheel chock includes: The identification information of the target wheel chock is corrected to the target wheel chock identifier.

8. A wheel chock identification device, characterized in that, The device includes: The first determining module is configured to determine the identification information of the target wheel stopper in the first image and the storage location information used to characterize the storage location of the target wheel stopper. The first image is a frame image in an image sequence obtained through image acquisition, and the identification information is an ID obtained by tracking the target wheel stopper. The first acquisition module is configured to acquire association information for indicating the wheel stop identifier associated with each storage location, wherein the wheel stop identifier is a wheel stop ID; The second determining module is configured to determine whether the identification information and storage location information of the target wheel stopper match the associated information; The correction module is configured to correct the identification information of the target wheel chock if it does not conform to the associated information. The associated information is formed in the following ways: Based on the tracking results of the wheel stops and storage locations in the first image and previous images in the image sequence, the wheel stop identifier of each wheel stop and the storage location identifier of each storage location are determined. Based on the relative positional relationship between the wheel stopper and the storage location, determine the wheel stopper whose location is within the coverage area of ​​the storage location; For any storage location, the wheel stop identifier corresponding to the wheel stop within the coverage area of ​​that storage location is determined as the wheel stop identifier associated with that storage location.

9. A vehicle, characterized in that, include: First processor; A first memory for storing instructions executable by a first processor; The first processor is configured to execute instructions in the first memory to implement the steps of the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the program instructions are executed by the second processor, they implement the steps of the method according to any one of claims 1 to 7.

Images