Method and apparatus for determining dynamic target appearance size

Abstract

The present disclosure provides a dynamic target appearance size determination method and a dynamic target appearance size determination device, and relates to the technical field of intelligent transportation. The method comprises the following steps: determining a detection area, and acquiring a plurality of frames of point cloud data and a plurality of measured speeds of a dynamic target in the detection area; determining motion compensation according to the plurality of measured speeds based on the time points at which the plurality of measured speeds are acquired and the time points at which the plurality of frames of point cloud data are acquired; performing registration processing on the plurality of frames of point cloud data based on the motion compensation to obtain reconstructed point cloud of the moving target; and determining the measured appearance size of the dynamic target according to the reconstructed point cloud and the motion direction of the dynamic target. The technical scheme provides a dynamic target appearance size determination scheme, and can accurately determine the appearance size of the dynamic target.

Description

[0001] This application is a divisional application of Chinese application No. 202111169245.3, the foregoing contents of which are incorporated herein by reference. Technical Field

[0002] This disclosure relates to the field of intelligent transportation technology, and in particular to a method and apparatus for determining the appearance size of a dynamic target, a method and apparatus for determining the size, a computer-readable storage medium, and an electronic device. Background Technology

[0003] Current object appearance inspection technologies generally require the target to be inspected to be in a static state; that is, the object's appearance dimensions are measured while it is stationary. However, a method for determining the appearance dimensions of dynamic targets is lacking.

[0004] For example, on roads with height restrictions, determining the external dimensions of vehicles (dynamic targets) traveling on the road will help to quickly identify vehicles violating the rules, and timely warnings will effectively improve driving safety.

[0005] Therefore, the dynamic object target appearance determination method proposed in this paper can effectively solve this type of application scenario. By matching with the backend database, it can effectively reduce law enforcement costs and improve road transport safety.

[0006] It should be noted that the information disclosed in the background section above is only used to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0007] The purpose of this disclosure is to provide a method, apparatus, computer-readable storage medium, and electronic device for determining the appearance size of a dynamic target, which, while ensuring low latency, at least to some extent reduces bandwidth and the high power consumption problem caused by bandwidth issues.

[0008] Other features and advantages of this disclosure will become apparent from the following detailed description, or may be learned in part from practice of this disclosure.

[0009] According to one aspect of this disclosure, a method for determining the appearance size of a dynamic target is provided. The method includes: determining a detection area; acquiring multi-frame point cloud data and multiple measurement velocities of a dynamic target within the detection area; determining motion compensation based on the time points of acquiring the multiple measurement velocities and the time points of acquiring the multi-frame point cloud data; performing registration processing on the multi-frame point cloud data based on the motion compensation to obtain a reconstructed point cloud of the dynamic target; and determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0010] According to another aspect of this disclosure, a device for determining the appearance size of a dynamic target is provided, the device comprising: an acquisition module, a compensation determination module, a registration module, and a determination module.

[0011] The acquisition module is used to determine the detection area and acquire multi-frame point cloud data and multiple measurement velocities of the dynamic target within the detection area; the compensation determination module is used to determine motion compensation based on the time points of acquiring the multiple measurement velocities and the time points of acquiring the multiple frame point cloud data; the registration module is used to perform registration processing on the multiple frame point cloud data based on the motion compensation to obtain the reconstructed point cloud of the moving target; and the determination module is used to determine the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0012] Example 1. A method for determining the appearance size of a dynamic target, the method comprising: determining a detection area; acquiring multi-frame point cloud data and multiple measurement velocities of a dynamic target within the detection area; determining motion compensation based on the time points of acquiring the multiple measurement velocities and the time points of acquiring the multi-frame point cloud data; performing registration processing on the multi-frame point cloud data based on the motion compensation to obtain a reconstructed point cloud of the moving target; and determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0013] Example 2. According to the method described in Example 1, acquiring multi-frame point cloud data and multiple measured velocities of a dynamic target within the detection area includes: acquiring point cloud data C of the moving target using a lidar, where C = {(T C0 ,C0), (T C1 ,C1),…(T Ci C i ), …(T CN C N )}, where T Ci This indicates that the point cloud data C of the i-th frame is obtained. i The time point, N is a positive integer, and i is a positive integer not greater than N; in response to the lidar determining that the dynamic target has entered the detection area, the velocity measuring radar is triggered to acquire the measured velocity V of the moving target, where V = {(T V0 ,V0),(T V1 ,V1),…(T Vk V k ), …(T VM V M )}, where T Vk This indicates that the measured speed V is obtained. kAt a time point, M is a positive integer, and k is a positive integer not greater than M.

[0014] Example 3. According to the method described in Example 2, based on the time points for obtaining the multiple measured velocities and the time points for obtaining the multiple frames of point cloud data, determining motion compensation according to the multiple measured velocities includes: According to T Ci and T Vk to determine the compensation mode based on their magnitude relationship, and to determine the compensation velocity according to the measured velocity V k and the estimated velocity, and determining the compensation mode and the compensation velocity as the motion compensation.

[0015] Example 4. According to the method described in Example 3, determining the compensation mode according to the magnitude relationship between T Ci and T Vk , and determining the compensation velocity according to the measured velocity V k and the estimated velocity, includes: If T Ci < T V0 and T V0 -T Ci < T0, then perform motion compensation according to the compensation mode of uniform linear motion, and determine the compensation velocity as the measured velocity V0 obtained at the time point of T V0 , where T0 is the first preset duration; if T Ci > T Vl and T Ci < T Vp , then perform motion compensation according to the compensation mode of uniform linear motion, and use interpolation to obtain the compensation velocity according to V l and V p , where both l and p are positive integers less than M, and l is less than p, and V l and V p are the measured velocities obtained at the time points of T Vl and T Vp respectively; if T Ci > T Vk , then perform motion compensation according to the compensation mode of uniform linear motion, and the compensation velocity corresponding to the first stage is the measured velocity V Vk obtained at the time point of T k , and the compensation velocity corresponding to the second stage is the estimated velocity, where the estimated velocity is determined according to V k and the point cloud registration velocity.

[0016] Example 5. According to any one of Examples 1 to 4, the step of determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target includes: determining a minimum bounding rectangle in the reconstructed point cloud based on the direction perpendicular to the motion direction; determining the distance of the minimum bounding rectangle perpendicular to the motion direction as the width information W of the moving target; determining the distance of the minimum bounding rectangle in the motion direction as the length information L of the moving target; and determining the maximum value in the reconstructed point cloud as the height information H of the moving target.

[0017] Example 6. The method according to any one of Examples 1 to 4, the method further comprising: acquiring background height information of the detection area through the lidar; rasterizing the detection area to obtain X*Y grids, where X and Y are positive integers; acquiring background height information B(x,y) of each grid, where B(x,y) represents the height of the ground point of grid (x,y), where x takes values ​​of 1, 2, ..., X, and y takes values ​​of 1, 2, ..., Y; the step of determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target includes: acquiring the measured height of each grid based on the reconstructed point cloud. The measurement height information G(x,y) and the background height information B(x,y) are calculated, and the reconstructed point cloud is binarized based on the height difference. A minimum bounding rectangle is determined in the reconstructed point cloud based on the direction perpendicular to the motion direction. The distance of the minimum bounding rectangle perpendicular to the motion direction is determined as the width information W of the moving target, and the distance of the minimum bounding rectangle in the motion direction is determined as the length information L of the moving target. The maximum value of the height difference is determined as the height information H of the moving target.

[0018] Example 7. According to any one of Examples 1 to 4, after obtaining the reconstructed point cloud of the moving target, the method further includes: calculating the motion speed of the moving target by calculating the motion speed of the moving target by calculating the neighboring point clouds that have undergone the registration process, and obtaining the point cloud registration speed; and correcting the speed of the speed measuring radar by the point cloud registration speed.

[0019] Example 8. According to the method described in Example 6, the method includes: obtaining the actual appearance size of the moving target, including: the width information W', length information L', and height information H' of the moving target; comparing the measured appearance size of the dynamic target with the actual appearance size corresponding to the dynamic target; and determining the dynamic target as a target to be processed if the comparison result includes one or more of the following: W is greater than W', L is greater than L', and H is greater than H'.

[0020] Example 9. According to the method described in Example 8, obtaining the actual appearance size of the moving target includes: responding to the lidar to determine that the dynamic target has entered the detection area, triggering the camera component to acquire an image of the moving target; performing image pixel compensation based on the measured speed to remove motion blur in the image; performing identity recognition of the moving target based on the image after the removal process, and obtaining the actual appearance size of the moving target based on the recognized identity.

[0021] Example 10. A device for determining the appearance size of a dynamic target, the device comprising: an acquisition module for determining a detection area and acquiring multi-frame point cloud data and multiple measurement velocities of a dynamic target within the detection area; a compensation determination module for determining motion compensation based on the time points of acquiring the multiple measurement velocities and the time points of acquiring the multi-frame point cloud data; a registration module for performing registration processing on the multi-frame point cloud data based on the motion compensation to obtain a reconstructed point cloud of the moving target; and a determination module for determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0022] According to another aspect of this disclosure, an electronic device is provided, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor, when executing the computer program, implements the method for determining the dynamic target appearance size as described in the above embodiments.

[0023] According to another aspect of this disclosure, a computer-readable storage medium is provided having a computer program stored thereon, characterized in that, when the computer program is executed by a processor, it implements the method for determining the dynamic target appearance size as described in the above embodiments.

[0024] The method, apparatus, computer-readable storage medium, and electronic device for determining the appearance size of a dynamic target provided in the embodiments of this disclosure have the following technical effects:

[0025] This technical solution, after determining the detection area, acquires multi-frame point cloud data and multiple measured velocities of the dynamic target within that area, and determines motion compensation based on the measured velocities. Then, point cloud data registration is performed based on the motion compensation, resulting in a reconstructed point cloud containing motion information. Furthermore, the external dimensions of the dynamic target can be determined based on the reconstructed point cloud and its motion direction. This technical solution provides a method for determining the external dimensions of dynamic targets with high accuracy.

[0026] 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

[0027] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure. It is obvious that the drawings described below are merely some embodiments of this disclosure, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0028] Figure 1 This is a schematic diagram illustrating a scenario for determining the appearance size of a dynamic target in an exemplary embodiment of this disclosure.

[0029] Figure 2 This diagram illustrates a flowchart of a method for determining the appearance size of a dynamic target in an exemplary embodiment of this disclosure.

[0030] Figure 3 A flowchart illustrating a method for determining the appearance size of a dynamic target in another exemplary embodiment of this disclosure is shown.

[0031] Figure 4 The illustration shows a flowchart of determining a reconstructed point cloud about a moving target using point cloud data C and velocity data V, according to an exemplary embodiment of the present disclosure.

[0032] Figure 5 This diagram illustrates a flowchart of a method for determining motion compensation in an exemplary embodiment of this disclosure.

[0033] Figure 6 This diagram illustrates a flowchart of a method for determining the appearance size of a dynamic target in yet another exemplary embodiment of the present disclosure.

[0034] Figure 7 A schematic diagram of a grid reflecting background height information is shown in an exemplary embodiment of the present disclosure.

[0035] Figure 8 A schematic diagram of a measurement device compensation enhancement scheme provided according to an exemplary embodiment of the present disclosure is shown.

[0036] Figure 9 A schematic diagram of a device for determining the appearance size of a dynamic target that can be applied according to an embodiment of the present disclosure is shown.

[0037] Figure 10 A schematic diagram of the structure of a device for determining the appearance size of a dynamic target according to another embodiment of the present disclosure is shown.

[0038] Figure 11 A schematic diagram of the structure of a computer system suitable for implementing the embodiments of the present disclosure is shown. Detailed Implementation

[0039] To make the objectives, technical solutions, and advantages of this disclosure clearer, the embodiments of this disclosure will be described in further detail below with reference to the accompanying drawings.

[0040] In the following description, when referring to the accompanying drawings, the same numbers in different drawings denote the same or similar elements unless otherwise indicated. 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.

[0041] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in many forms and should not be construed as limited to the examples set forth herein; rather, these embodiments are provided to make this disclosure more comprehensive and complete, and to fully convey the concept of the example embodiments to those skilled in the art. The described features, structures, or characteristics can be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a full understanding of embodiments of this disclosure. However, those skilled in the art will recognize that the technical solutions of this disclosure can be practiced with one or more of the specific details omitted, or other methods, components, apparatus, steps, etc., can be employed. In other instances, well-known technical solutions are not shown or described in detail to avoid obscuring various aspects of this disclosure.

[0042] Furthermore, the accompanying drawings are merely illustrative of this disclosure and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities may be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor devices and / or microcontroller devices.

[0043] The following is a detailed description of an embodiment of the method for determining the dynamic target appearance dimensions provided in this disclosure:

[0044] in, Figure 1 This diagram illustrates a scenario illustrating a scheme for determining the appearance size of a dynamic target in an exemplary embodiment of this disclosure. (See reference...) Figure 1For a moving target 110 (e.g., a moving vehicle), after it enters the monitoring area 120, point cloud data and speed of the moving target are acquired by a measuring device 130 (e.g., lidar, speed radar, camera components, etc.). Further, the point cloud data and speed are processed by a computing device 150 to determine the external dimensions of the moving target 110. For example, the external dimensions of each moving target 110 can also be displayed by a display device 160.

[0045] Figure 2 This diagram illustrates a flowchart of a method for determining the appearance dimensions of a dynamic target in an exemplary embodiment of this disclosure. (See reference...) Figure 2 The method includes:

[0046] S210, determine the detection area, and acquire multi-frame point cloud data and multiple measurement velocities of dynamic targets within the detection area.

[0047] In an exemplary embodiment, reference is made to Figure 1 The monitoring area 120 can be a road with height restrictions. The range of the area to be measured is related to the point cloud density acquired by the lidar and the size of the moving target. Specifically, the point cloud density of the lidar scan result of the area to be measured 120 needs to reach a preset value, and the size of the area to be measured 120 needs to be basically larger than the size of the target object.

[0048] In an exemplary embodiment, point cloud data of the dynamic target 110 is acquired using a lidar system. Specifically, the lidar acquires point cloud data of the dynamic target 110 by scanning, for example, scanning measurement point data of attributes such as the length, width, and height of the vehicle. However, since the lidar measurement points are discrete points, and the measured object may be large, it is impossible to scan the entire scene containing the appearance of the measured object in one scan. Therefore, in this embodiment, the lidar needs to continuously scan the moving target to obtain multiple frames of point cloud data to reduce measurement errors and ensure the integrity of the measured object.

[0049] In an exemplary embodiment, the speed of the dynamic target 110 is acquired by the speed measuring radar (referred to as "measured speed" here to distinguish it from the "actual speed" mentioned later). Specifically, for the speed measuring radar, when the movement is within the measurement area 120, the speed measuring radar acquires the speed of the vehicle under test. In order to improve the measurement accuracy of the system, the speed measuring radar also needs to continuously acquire the speed of the object under test.

[0050] S220, based on the time points for acquiring the multiple measured speeds and the time points for acquiring the multiple frames of point cloud data, determine motion compensation according to the multiple measured speeds; and S230, based on the motion compensation, perform registration processing on the multiple frames of point cloud data to obtain the reconstructed point cloud of the moving target.

[0051] In the registration process based on point cloud data, this embodiment of the disclosure considers motion compensation to make the resulting reconstructed point cloud more closely match the appearance of the moving target, thereby improving the accuracy of appearance information determination. Specifically, motion compensation is determined based on the time points at which the multiple measured velocities are acquired and the time points at which the multiple frames of point cloud data are acquired. By considering the velocity information at the time of acquiring the current frame of point cloud data, this technical solution is applicable not only to moving targets at a constant speed but also to moving targets at varying speeds, thus improving the accuracy of motion compensation determination and expanding the applicability of this solution.

[0052] S240, determine the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0053] In this Figure 2 In the embodiment shown, after determining the detection area, multiple frames of point cloud data and multiple measured velocities of the dynamic target within that detection area are acquired. Motion compensation is determined based on the measured velocities. Point cloud data registration is then performed based on the motion compensation, resulting in a reconstructed point cloud containing motion information. Furthermore, the external dimensions of the dynamic target can be determined based on the reconstructed point cloud and its motion direction. This technical solution provides a method for determining the external dimensions of a dynamic target, enabling efficient and accurate determination of the target's external dimensions.

[0054] In an exemplary embodiment, Figure 3 A flowchart illustrating a method for determining the appearance dimensions of a dynamic target in another exemplary embodiment of this disclosure is shown below. Figure 3 The specific implementation methods of each step in the illustrated embodiments are described in detail below:

[0055] refer to Figure 3 In S310, point cloud data C of the moving target is acquired by lidar.

[0056] Specifically, where C = {(T C0 ,C0), (T C1 ,C1),…(T Ci C i ), …(T CN C N )}, where T Ci This indicates that the point cloud data C of the i-th frame is obtained. i The time point is N, where N is a positive integer and i is a positive integer not greater than N.

[0057] refer to Figure 1The measuring device 130 shown includes a lidar, a speed radar, and a camera. Initially, only the lidar is active, while the speed radar and camera are in standby mode. When the lidar detects changes in depth, height, or color information within the detection area 120, it can determine that a target object (i.e., a moving target) is present in the detection area.

[0058] In this embodiment, to reduce system power consumption, the speed measuring radar and camera components are only activated after the lidar determines that a dynamic target has entered the detection area. Similarly, at the initial time point, only the camera device is on, while the speed measuring radar and lidar are in standby mode. Activating the speed measuring radar and lidar only after the camera determines that a dynamic target has entered the detection area minimizes the power consumption of the entire measurement system.

[0059] S320-S350 are used to determine the measured external dimensions of the moving target, while S320'-S340' are used to determine the actual external dimensions of the moving target.

[0060] Specifically, in the process of determining the measured appearance size of a moving target, motion compensation is first determined (S330). Then, based on the motion compensation, point cloud data is registered to obtain the reconstructed point cloud of the moving target (S340). Further, the measured appearance size of the moving target is determined according to the reconstructed point cloud and the motion direction of the moving target (S350).

[0061] The specific implementation methods of S320-S350 are described below:

[0062] S320, in response to the lidar determining that a dynamic target has entered the detection area, the velocity measuring radar is triggered to acquire the measured velocity V of the moving target.

[0063] Specifically, where V = {(T V0 ,V0),(T V1 ,V1),…(T Vk V k ), …(T VM V M )}, where T Vk This indicates that the measured speed V is obtained. k The time points are defined as follows: M is a positive integer, and k is a positive integer not greater than M. By acquiring velocity data at multiple time points, more comprehensive velocity information of the moving target within the detection area can be obtained, making this scheme applicable not only to targets moving at a constant speed but also to targets moving at varying speeds. Furthermore, continuously acquiring the velocity of the object under test also helps improve the system's measurement accuracy.

[0064] refer to Figure 4 The reconstructed point cloud for the moving target is determined using point cloud data C and velocity data V for the same moving target (see the embodiment corresponding to S340 for details). The reconstructed point cloud is a reconstruction of the appearance of the measured object. For example, the iterative nearest neighbor method (ICP) is used to register point cloud data C0 and point cloud data C1. To improve the efficiency and reconstruction accuracy of point cloud data matching, motion compensation is needed based on information such as the vehicle's speed; that is, motion compensation is considered during the registration process to obtain point cloud C. P1 Among them, motion compensation is based on T Ci and T Vk The magnitude relationship and the speed data V are determined (refer to the embodiment corresponding to S330 for details).

[0065] Continue to refer to Figure 3 In S330, according to T Ci and T Vk The magnitude relationship determines the corresponding compensation mode under different conditions, and the measured speed V k By estimating the velocity, the corresponding compensation velocity is determined for different situations, thereby determining the motion compensation for different situations.

[0066] In an exemplary embodiment, Figure 5 This diagram illustrates a flowchart of a motion compensation determination method in an exemplary embodiment of this disclosure. Specifically, Figure 5 This can be used as a specific implementation of S330, see reference. Figure 5 This includes S331-S337.

[0067] In S331, T V0 -T0 <T Ci That is, determine T. Ci Is it greater than T? V0 -T0. Where T0 is the first preset duration.

[0068] In this embodiment, the point cloud data C of the i-th frame is determined. i The corresponding acquisition time point T Ci Is it greater than the time point T corresponding to the first measured speed V0 acquired by the speed measuring radar? V0 .

[0069] If S331 is not true, it means that the measurement time T of the current frame point cloud data is... Ci Less than T V0 If the measurement time of the first measured velocity V0 differs too much (greater than T0), meaning the current frame point cloud data is not suitable for motion compensation, then execute S337: discard the point cloud data, thereby reducing the amount of computation while ensuring the accuracy of the calculation.

[0070] If S331 is true, it means that the current frame point cloud data C i Measurement time T Ci And the measurement time T of the first measured velocity V0 V0 If the difference is small (less than T0), then continue executing S332. In S332, T... V0 -T0 <T Ci <T V0 Determine the current frame's point cloud data C. i Measurement time T Ci Is it less than the measurement time T of the first measured velocity V0? V0 .

[0071] If S332 is true, it means that the current frame point cloud data C i Measurement time T Ci Measurement time T less than V0 V0 And the measurement time T of V0 V0 Similar, meaning that the current frame point cloud data C i This refers to the point cloud data scanned by the lidar before the speed radar begins normal operation, but due to the measurement time T of V0... V0 Since they are similar, S333 is executed: motion compensation is performed according to the compensation mode of uniform linear motion, and the compensation speed is determined to be T. V0 The measured velocity V0 is obtained at a specific time point.

[0072] If S332 is not true, it means that the current frame point cloud data C i Measurement time T Ci Measurement time T greater than V0 V0 Then the current frame point cloud data C i This refers to the point cloud data scanned by the lidar after the speed measuring radar begins normal operation. Then, further execution of S334:T Vl <T Ci <T Vp Determine the current frame's point cloud data C. i Measurement time T Ci Is it in T? Vl With T Vp between.

[0073] If S334 is true, it means that the current frame point cloud data C i The point cloud data scanned by the lidar after the speed-measuring radar starts working normally, and is at a known measurement speed V. l Measurement time T Vl With the known measured velocity V p Measurement time T Vp In between, execute S335: based on the measured speed V l With the measured speed Vp The difference determines the compensation mode, and when the compensation mode is uniform linear motion, an interpolation method is used based on V. l and V p Interpolation is performed to obtain the compensation speed (as in formula (1)), where l and p are both positive integers less than M, and l is less than p, V l and V p T respectively Vl and T Vp The measured velocity is obtained at a specific time point. The formula for calculating the compensated velocity V in this case is as follows:

[0074] V” = (T Vp - T Ci ) / (T Vp - T Vl ) * V l + (T Ci - T Vl ) / (T Vp - T Vl ) * V p (1)

[0075] For example, measuring speed V l With the measured speed V p When the difference is small, the compensation mode is uniform linear motion. (Measurement speed V) l With the measured speed V p When the difference is large (greater than the preset value), a uniform acceleration motion model can be used for motion modeling, or a higher-order interpolation method can be used to obtain velocity interpolation, thereby improving the accuracy of motion compensation.

[0076] If S334 is not true, and because the previous frame point cloud data C i Measurement time T Ci Measurement time T greater than V0 V0 This indicates that the current frame point cloud data C i This should refer to the point cloud data scanned by the lidar after the speed measuring radar has finished its measurement work, i.e., belonging to T. Vk <T Ci The situation.

[0077] Then execute S336: Perform motion compensation according to the uniform linear motion compensation mode, and the compensation speed corresponding to the first stage is T. Vk Measurement of velocity V obtained at time points k The compensation speed corresponding to the second stage is the estimated speed (formula (2)), where the estimated speed is based on the measured speed V. k The registration speed with point clouds is determined.

[0078] For example, in TVk <T Ci In this situation, to further improve the accuracy of motion compensation, the compensation speed is determined in two stages. The first stage mentioned above specifically refers to the distance T... Vk Within a relatively recent timeframe, within this timeframe (i.e., the first stage mentioned above), the compensation velocity can be determined as T. Vk Measurement of velocity V obtained at time points k The second stage mentioned above can specifically refer to the distance T. Vk Over a longer time period, within that time segment (i.e., the second stage mentioned above), if the compensation velocity is again determined to be T... Vk Measurement of velocity V obtained at time points k This may affect the accuracy of the compensation speed; therefore, the compensation speed corresponding to the second stage will be based on the measurement speed V. k Registration speed V of point cloud cti Make an estimate.

[0079] For example, the calculation formula for the compensation velocity V (i.e., the estimated velocity mentioned above) corresponding to the second stage is as follows:

[0080] V” = k * V k + (1 - k) * V cti (2)

[0081] Where, k = (T Ci -T Vk ) / T1, where T1 is the second preset duration and V is the point cloud registration speed. cti The method for determining the speed is as follows: after matching two frames of point cloud data, the distance change value of a certain point will be obtained. Based on the time difference between the two frames of point cloud data, the speed of the point can be determined, which is the point cloud registration speed mentioned above.

[0082] For example, in order to reduce motion estimation errors in point cloud matching, the estimated velocity can be smoothed by applying Kalman filtering.

[0083] pass Figure 5 The illustrated embodiment, according to T Ci and T Vk The magnitude relationship determines the corresponding compensation mode under different conditions, and the measured speed V k The estimated velocity determines the corresponding compensation velocity under different conditions, thus determining the corresponding motion compensation under different conditions. Motion compensation is then used in the point cloud matching process (e.g.,...). Figure 4 S340 is executed to register multi-frame point cloud data based on motion compensation under different conditions (including compensation mode and compensation speed) to obtain the reconstructed point cloud of the moving target.

[0084] For example, refer to Figure 4 When performing registration processing between point clouds C0 and C1, the measurement time T of point cloud C1 is determined. C1 belong Figure 5 In the illustrated embodiment, which scenario is considered, and motion compensation is performed based on the corresponding compensation mode and speed, thereby determining the registered point cloud C0 and point cloud C1. P1 .

[0085] Furthermore, in performing point cloud C... P1 During the registration process with point cloud C2, the measurement time T of point cloud C2 is determined. C2 belong Figure 5 In the illustrated embodiment, which scenario is considered, and motion compensation is performed based on the corresponding compensation mode and compensation speed to determine the point cloud C? P1 Point cloud C after registration with point cloud C2 P2 .

[0086] Finally, in the point cloud C P(N-1) With point cloud C N During the registration process, the point cloud C is determined. N Measurement time T CN belong Figure 5 In the illustrated embodiment, which scenario is considered, and motion compensation is performed based on the corresponding compensation mode and compensation speed to determine the point cloud C? P(N-1) With point cloud C N Point cloud C after registration processing PN .

[0087] This achieves registration processing of all point cloud data related to the aforementioned moving targets, resulting in a reconstructed point cloud that accurately reflects the appearance of the moving targets. Further reference is then made. Figure 4 During the registration process, the motion estimator V P1 V P2 ...V PN The calculation method is as follows:

[0088] M(x, y, z, roll, pitch, yaw) = [Vx * t, Vy * t, 0, 0, 0, 0] (3)

[0089] Where Vx and Vy are the components of the measured velocity V and y respectively, t is the time interval between two point cloud scans; x, y, z are the translational amounts of the motion estimation; roll, pitch, yaw are the rotational amounts of the motion estimation.

[0090] Continue to refer to Figure 3In step S350: The reconstructed point cloud is projected onto a target plane parallel to the direction of motion. Based on the direction of motion and the distance perpendicular to the direction of motion within the target plane, a minimum bounding rectangle is determined in the reconstructed point cloud. The distance of the minimum bounding rectangle perpendicular to the direction of motion is determined as the width information W of the moving target, and the distance of the minimum bounding rectangle in the direction of motion is determined as the length information L of the moving target. The maximum Z-coordinate value in the reconstructed point cloud is determined as the height information H of the moving target.

[0091] For example, due to possible matching errors, the reconstructed point cloud (denoted as C0) is noise-processed before executing S350, for example, by using an isolated point removal method, thereby reducing the matching noise introduced during point cloud matching, and the resulting noise-removed point cloud is denoted as C1.

[0092] In the embodiment of S350, when the moving target is a moving vehicle, the target plane / xy plane can be considered as the cross-section of the vehicle. The point cloud C1 is then projected onto the xy plane, and the minimum inscribed quadrilateral is determined with the vehicle's direction of motion in the xy plane as the first side direction and the direction of the vehicle's movement in the xy plane as the second side direction.

[0093] In this context, the length of the first side can be considered as the length L of the vehicle, and the length of the second side can be considered as the width W of the vehicle. Furthermore, the maximum Z-coordinate in the reconstructed point cloud is determined as the height information H of the moving target.

[0094] In an exemplary embodiment, to further improve the accuracy of determining the height information of the moving target, this embodiment also considers the background height. That is, the Z-axis coordinate in the point cloud is the measured height, and the difference between the measured height and the background height is taken as the height of the moving target.

[0095] Specifically, Figure 6 This diagram illustrates a flowchart of a method for determining the dynamic target appearance size in another exemplary embodiment of this disclosure. Specifically, it can be implemented as one embodiment of S350. Specifically:

[0096] S351, acquire the background height information of the detection area using a lidar, and rasterize the detection area to obtain X*Y grids, where X and Y are positive integers. S352, acquire the background height information B(x,y) of each grid, where B(x,y) represents the height of the ground point of grid (x,y).

[0097] For example, background modeling is performed on the measurement area 120: such as Figure 7 As shown, background height information of the measurement area is obtained using a lidar system. Figure 7The grayscale value of the grid is determined based on the grid's background height information B(x,y). For example, the larger the B(x,y) value, the larger the grid grayscale value. Simultaneously, to facilitate data processing, the measurement area is rasterized, and the rasterized background modeling information is represented as B(x,y), where x takes values ​​of 1, 2, ..., X, and y takes values ​​of 1, 2, ..., Y.

[0098] Where B(x,y) is the height of the ground point of the grid (x,y).

[0099] S353, obtain the measured height information G(x,y) of each grid cell based on the reconstructed point cloud.

[0100] For example, the measured height information G(x,y) of each grid cell is determined based on the Z-axis coordinates in the point cloud C1 after the above denoising process.

[0101] S354 calculates the height difference between the measured height information G(x,y) and the background height information B(x,y), and performs binarization processing on the reconstructed point cloud based on the height difference.

[0102] For example, the height difference is: H(x,y) = G(x,y) - B(x,y)

[0103] H(x,y) is compared with the preset length threshold T2 to obtain the foreground grid occupied by the dynamic target. The foreground grid judgment formula (4) is as follows:

[0104]

[0105] S355, Based on the direction perpendicular to the motion direction and the direction of motion, determine the minimum bounding rectangle in the reconstructed point cloud after binarization. And, S356, determine the distance of the minimum bounding rectangle perpendicular to the motion direction as the width information W of the moving target, determine the distance of the minimum bounding rectangle in the motion direction as the length information L of the moving target, and determine the maximum value of the height difference H(x,y) as the height information H of the moving target.

[0106] pass Figure 6 The embodiment shown obtains the measured external dimensions S = {L, W, H} of the dynamic target.

[0107] The above describes how to determine the measured external dimensions S={L,W,H} of a moving target using S320-S350. The following describes how to determine the actual external dimensions D={L',W',H'} of a moving target using S320'-S340'. Specifically:

[0108] refer to Figure 3In S320', in response to the lidar determining that a dynamic target has entered the detection area, the camera component is triggered to acquire an image of the moving target; in S330', image pixel compensation is performed based on the measured speed to remove motion blur in the image; and in S340', the identity of the moving target is identified based on the image after the removal process, and the actual appearance size of the moving target is obtained based on the identified identity.

[0109] For example, when the lidar detects an object entering the measurement area based on background cancellation, it triggers the camera component and speed radar to operate. In the case where the moving target is a moving vehicle, the camera component acquires an image of the vehicle within the detection area. Further, image recognition technology is used to identify the license plate in the image. For example, the identified license plate information is denoted as P.

[0110] refer to Figure 1 License plate recognition can be performed on either the measuring device 130 or the computing device 150. If the image is recognized on the measuring device 130, the amount of data uploaded to the computing device 150 is smaller. However, this increases the processing unit capacity requirements of the measuring device 130.

[0111] For example, when the moving target is moving at a high speed, motion blur removal is required to obtain image quality that meets measurement requirements. Specifically, image pixel compensation is performed based on the vehicle speed information acquired by the speed measuring radar, thereby removing motion blur from the image generated by the camera component.

[0112] For example, to improve recognition accuracy, this embodiment identifies moving targets (such as vehicles in motion) based on images processed by motion blur removal. Further, the actual external dimensions D = {L', W', H'} of the moving target are obtained based on the identified identity (e.g., license plate information). For instance, the obtained license plate information P is sent to the traffic enforcement department's database to query the vehicle's actual external dimensions. This allows for illegal vehicle appearance detection based on a threshold, as described in the following embodiment. This effectively utilizes the database information of relevant departments.

[0113] Continue to refer to Figure 3 After determining the measured external dimensions S={L,W,H} of the moving target through S320-S350, and the actual external dimensions D={L',W',H'} of the moving target through S320'-S340', S360 is executed: comparing the measured external dimensions of the dynamic target with the corresponding actual external dimensions of the dynamic target; and S370 is executed: if the comparison result includes one or more of the following: W is greater than W', L is greater than L', and H is greater than H', the dynamic target is determined to be the target to be processed.

[0114] For example, the calculation method for the target to be processed is determined based on the measured appearance size S and the actual appearance size D, as shown in formulas (5) and (6).

[0115] D(L, W, H) = S - D = {|L' - L|, |W' - W|, |H' - H|} (5)

[0116] Here, |*| represents taking the absolute value of *.

[0117]

[0118] In the case where the moving target is a moving vehicle, the target to be processed is a modified vehicle. "1" indicates that the moving target is a modified vehicle, and "0" indicates that the moving target is not a modified vehicle. The preset length threshold T3 = {TL, TW, TH}, where TL, TW, and TH represent the thresholds for changes in the length, width, and height of the moving target, respectively.

[0119] This technical solution proposes a method that utilizes multiple sensors to measure the appearance of objects in real time and combines this with a traffic enforcement database. By fully leveraging existing intelligent transportation infrastructure and data, it can quickly detect illegal vehicle modifications, improving enforcement efficiency. Simultaneously, it can effectively prevent dangerous transportation behaviors. For example, by combining road height restriction information, it can determine whether a vehicle can safely travel on a given road section, reducing traffic accidents or impassable roads caused by height restrictions. In other application scenarios, such as on closed highway sections, measuring the appearance at intervals can prevent drivers from discovering undetected deformations, thereby reducing accident risks.

[0120] In an exemplary embodiment, to improve the overall stability and robustness of the system, Figure 8 A schematic diagram of a compensation enhancement scheme between measurement devices provided according to an exemplary embodiment of this disclosure is shown. Specifically, Figure 8 The logic for mutual compensation and correction of data from multiple sensors is shown.

[0121] (a) Regarding compensation 1: During point cloud registration, the measured velocity obtained by the velocimetry radar is used to provide motion compensation. This improves the point cloud matching accuracy and is beneficial for obtaining a reconstructed point cloud with higher accuracy.

[0122] (b) Regarding Compensation 2: Image pixel compensation is performed on the image of the camera component based on the measured speed obtained by the speed measuring radar to remove motion blur in the image. Therefore, the identity of moving targets can be more accurately identified based on the image after the blurring process.

[0123] (c) Regarding compensation 3: Since lidar has a strong positioning capability for highly reflective objects (such as license plates of vehicles in motion), lidar can improve the positioning accuracy of the camera component for the target (such as license plates).

[0124] (d) Regarding compensation 4: After point cloud matching is completed, motion velocity can also be estimated based on point cloud matching to obtain point cloud registration velocity. Thus, the velocity of the speed measuring radar can be corrected by the point cloud matching velocity, which can improve the accuracy and robustness of the speed measuring radar measurement.

[0125] It should be noted that the above figures are merely illustrative of the processes included in the method according to exemplary embodiments of the present invention, and are not intended to be limiting. It is readily understood that the processes shown in the above figures do not indicate or limit the temporal order of these processes. Furthermore, it is readily understood that these processes may, for example, be executed synchronously or asynchronously in multiple modules.

[0126] The following are embodiments of the apparatus disclosed herein, which can be used to execute embodiments of the method disclosed herein. For details not disclosed in the apparatus embodiments of this disclosure, please refer to the embodiments of the method disclosed herein.

[0127] in, Figure 9 A schematic diagram of a device for determining the dynamic appearance dimensions of a target, according to an embodiment of this disclosure, is shown. Please refer to... Figure 9 The device for determining the dynamic target appearance size shown in the figure can be implemented as a whole or part of an electronic device through software, hardware, or a combination of both, or it can be integrated as an independent module into an electronic device or server.

[0128] The dynamic target appearance size determination device 900 in this embodiment includes: an acquisition module 910, a compensation determination module 920, a registration module 930, and a size determination module 940, wherein:

[0129] The acquisition module 910 is used to determine the detection area and acquire multi-frame point cloud data and multiple measurement velocities of the dynamic target within the detection area; the compensation determination module 920 is used to determine motion compensation based on the time points of acquiring the multiple measurement velocities and the time points of acquiring the multiple frame point cloud data; the registration module 930 is used to perform registration processing on the multiple frame point cloud data based on the motion compensation to obtain the reconstructed point cloud of the moving target; and the size determination module 940 is used to determine the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target.

[0130] In an exemplary embodiment, Figure 10Schematically shows a structural diagram of a device for determining the dynamic target appearance size in another exemplary embodiment according to the present disclosure. Please refer to Figure 10 :

[0131] In an exemplary embodiment, based on the foregoing solution, the obtaining module 910 is specifically configured to: obtain the point cloud data C of the moving target through a lidar, where C = {(T C0 , C0), (T C1 , C1), … (T Ci , C i ), … (T CN , C N )}, where T Ci represents the time point when the i-th frame of point cloud data C i is obtained, N is a positive integer, and i is a positive integer not greater than N; and, in response to the lidar determining that the dynamic target enters the detection area, trigger a speed measurement radar to obtain the measured speed V of the moving target, where V = {(T V0 , V0), (T V1 , V1), … (T Vk , V k ), … (T VM , V M )}, where T Vk represents the time point when the measured speed V k is obtained, M is a positive integer, and k is a positive integer not greater than M.

[0132] In an exemplary embodiment, based on the foregoing solution, the compensation determination module 920 is specifically configured to: determine a compensation mode according to the magnitude relationship between T Ci and T Vk , and determine a compensation speed according to the measured speed V k and the estimated speed, and determine the compensation mode and the compensation speed as the motion compensation.

[0133] In an exemplary embodiment, based on the foregoing solution, the compensation determination module 920 is specifically configured to: if T Ci < T V0 and T V0 - T Ci < T0, then perform motion compensation according to the compensation mode of uniform linear motion, and determine the compensation speed as the measured speed V0 obtained at the time point T V0 , where T0 is a first preset duration; if T Ci > T Vl and T Ci < T Vp , then perform motion compensation according to the compensation mode of uniform linear motion, and use interpolation to determine the compensation speed according to V l and Vp Interpolation is performed to obtain the compensated speed, where l and p are both positive integers less than M, and l is less than p, V l and V p T respectively Vl and T Vp The measured velocity at a given time point; if T Ci >T Vk Then, motion compensation is performed according to the compensation mode of uniform linear motion, and the compensation speed corresponding to the first stage is T. Vk Measurement of velocity V obtained at time points k The compensation speed corresponding to the second stage is the estimated speed, wherein the estimated speed is based on V. k The registration speed with point clouds is determined.

[0134] In an exemplary embodiment, based on the foregoing scheme, the size determination module 940 is specifically configured to: determine the minimum bounding rectangle in the reconstructed point cloud based on the direction perpendicular to the direction of motion; determine the distance of the minimum bounding rectangle perpendicular to the direction of motion as the width information W of the moving target; determine the distance of the minimum bounding rectangle in the direction of motion as the length information L of the moving target; and determine the maximum value in the reconstructed point cloud as the height information H of the moving target.

[0135] In an exemplary embodiment, based on the foregoing solution, the above-mentioned device further includes: a background information determination module 950.

[0136] The background information determination module 950 is used to: obtain background height information of the detection area through the lidar; rasterize the detection area to obtain X*Y grids, where X and Y are positive integers; and obtain background height information B(x,y) of each grid, where B(x,y) represents the height of the ground point of the grid (x,y), and x takes the value 1, 2, ..., X, and y takes the value 1, 2, ..., Y.

[0137] Specifically, the size determination module 940 is used to: obtain the measured height information G(x,y) of each grid cell based on the reconstructed point cloud; calculate the height difference between the measured height information G(x,y) and the background height information B(x,y), and perform binarization processing on the reconstructed point cloud based on the height difference; determine the minimum bounding rectangle in the reconstructed point cloud after binarization based on the direction perpendicular to the motion direction; determine the distance of the minimum bounding rectangle perpendicular to the motion direction as the width information W of the moving target, and determine the distance of the minimum bounding rectangle in the motion direction as the length information L of the moving target; and determine the maximum value of the height difference as the height information H of the moving target.

[0138] In an exemplary embodiment, based on the foregoing scheme, the above-mentioned device further includes: a correction module 960.

[0139] The correction module 960 is configured to: after the size determination module 940 obtains the reconstructed point cloud of the moving target, calculate the moving speed of the moving target by calculating the adjacent point cloud of the moving target through the registration process, and obtain the point cloud registration speed; and correct the speed of the speed measuring radar by the point cloud registration speed.

[0140] In an exemplary embodiment, based on the foregoing scheme, the size determination module 940 is further specifically used to: obtain the actual appearance size of the moving target, including: the width information W', length information L', and height information H' of the moving target.

[0141] The aforementioned device further includes a comparison module 970 and a target determination module 980.

[0142] The comparison module 970 is used to compare the measured external dimensions of the dynamic target with the corresponding actual external dimensions of the dynamic target. The target determination module 980 is used to determine the dynamic target as a target to be processed if the comparison result includes one or more of the following: W > W', L > L', and H > H'.

[0143] In an exemplary embodiment, based on the foregoing scheme, the size determination module 940 is further specifically configured to: respond to the lidar determining that the dynamic target has entered the detection area, trigger the camera component to acquire an image of the moving target; perform image pixel compensation based on the measured speed to remove motion blur in the image; and perform identity recognition of the moving target based on the image after the removal process, and obtain the actual appearance size of the moving target based on the recognized identity.

[0144] It should be noted that the above embodiments of the dynamic target appearance size determination device are only illustrated by the division of the above functional modules when executing the dynamic target appearance size determination method. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the dynamic target appearance size determination device and the dynamic target appearance size determination method embodiments provided in the above embodiments belong to the same concept. Therefore, for details not disclosed in the device embodiments of this disclosure, please refer to the above embodiments of the dynamic target appearance size determination method, which will not be repeated here.

[0145] The sequence numbers of the embodiments disclosed above are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments.

[0146] This disclosure also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any of the methods described in the foregoing embodiments. The computer-readable storage medium may include, but is not limited to, any type of disk, including floppy disks, optical disks, DVDs, CD-ROMs, microdrives, as well as magneto-optical disks, ROMs, RAMs, EPROMs, EEPROMs, DRAMs, VRAMs, flash memory devices, magnetic cards or optical cards, nanosystems (including molecular memory ICs), or any type of medium or device suitable for storing instructions and / or data.

[0147] This disclosure also 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 program to implement the steps of any of the methods described above.

[0148] Figure 11 A schematic diagram illustrating the structure of an electronic device according to an exemplary embodiment of this disclosure is provided. Please refer to... Figure 11 As shown, the electronic device 1100 includes a processor 1101 and a memory 1102.

[0149] In this embodiment, the processor 1101 is the control center of the computer system, and can be a processor of a physical machine or a processor of a virtual machine. The processor 1101 may include one or more processing cores, such as a 4-core processor or an 8-core processor. The processor 1101 can be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). The processor 1101 may also include a main processor and a coprocessor. The main processor, also known as the CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state.

[0150] In this embodiment of the disclosure, the processor 1101 is specifically used for:

[0151] The detection area is determined, and multiple frames of point cloud data and multiple measured velocities of the dynamic target within the detection area are acquired; based on the time points of acquiring the multiple measured velocities and the time points of acquiring the multiple frame point cloud data, motion compensation is determined according to the multiple measured velocities; based on the motion compensation, the multiple frame point cloud data is registered to obtain the reconstructed point cloud of the dynamic target; and the measured appearance size of the dynamic target is determined according to the reconstructed point cloud and the motion direction of the dynamic target.

[0152] Furthermore, the acquisition of multiple frames of point cloud data and multiple measured velocities of the dynamic target within the detection area includes: acquiring point cloud data C of the moving target using a lidar, where C = {(T C0 ,C0), (T C1 ,C1),…(T Ci C i ), …(T CN C N )}, where T Ci This indicates that the point cloud data C of the i-th frame is obtained. i The time point, N is a positive integer, i is a positive integer not greater than N; and, in response to the above-mentioned lidar determining that the above-mentioned dynamic target has entered the above-mentioned detection area, the velocity measuring radar is triggered to acquire the measured velocity V of the above-mentioned moving target, where V={(T V0 ,V0),(T V1 ,V1),…(T Vk V k ), …(T VM V M )}, where T Vk This indicates that the measured speed V is obtained. k The time point is M, where M is a positive integer and k is a positive integer not greater than M.

[0153] Furthermore, the motion compensation determination based on the time points for acquiring the multiple measured velocities and the time points for acquiring the multiple frames of point cloud data includes: according to T Ci and T Vk The magnitude relationship determines the compensation mode, and the measured velocity V is used as described above. k The estimated speed is used to determine the compensation speed, and the above compensation mode and the above compensation speed are determined as the above motion compensation.

[0154] Furthermore, the above is based on T Ci and T Vk The magnitude relationship determines the compensation mode, and the measured velocity V is used as described above. k And the estimated speed determines the compensation speed, including: if T Ci <T V0 And T V0 -T CiIf \(T < T_0\), then perform motion compensation according to the uniform linear motion compensation mode, and determine the compensation speed as \(T\), the measured speed \(V_0\) obtained at the time point, where \(T_0\) is the first preset duration; if \(T\) V0 > \(T_0\), and \(T\) Ci < \(T\) Vl and \(T\) Ci < \(T\), then perform motion compensation according to the uniform linear motion compensation mode, and use interpolation to obtain the compensation speed according to \(V\) Vp and \(V\) l where \(l\) and \(p\) are positive integers less than \(M\), and \(l < p\), \(V\) p and \(V\) l are the measured speeds obtained at time points \(T\) p and \(T\) Vl respectively; and if \(T\) Vp [[ID=2}}\(> T\) Ci > \(T\) Vk , then perform motion compensation according to the uniform linear motion compensation mode, and the compensation speed corresponding to the first stage is the measured speed \(V\) Vk obtained at time point \(T\) k , and the compensation speed corresponding to the second stage is the estimated speed, where the estimated speed is determined according to \(V\) k and the point cloud registration speed.

[0155] Further, determining the measured appearance size of the dynamic target according to the reconstructed point cloud and the motion direction of the dynamic target includes: determining a minimum bounding rectangle in the reconstructed point cloud based on the direction perpendicular to the motion direction and the motion direction; determining the distance of the minimum bounding rectangle in the direction perpendicular to the motion direction as the width information \(W\) of the moving target, and determining the distance of the minimum bounding rectangle in the motion direction as the length information \(L\) of the moving target; and determining the maximum value in the reconstructed point cloud as the height information \(H\) of the moving target.

[0156] Further, the method further includes: obtaining the background height information of the detection area through the lidar; rasterizing the detection area to obtain \(X\times Y\) grids, where \(X\) and \(Y\) are positive integers; obtaining the background height information \(B(x,y)\) of each grid, where \(B(x,y)\) represents the height of the ground point of the grid \((x,y)\), \(x\) takes values of \(1, 2,\cdots, X\), and \(y\) takes values of \(1, 2,\cdots, Y\);

[0157] The method of determining the measured appearance size of the dynamic target based on the reconstructed point cloud and the motion direction of the dynamic target includes: obtaining the measured height information G(x,y) of each grid cell based on the reconstructed point cloud; calculating the height difference between the measured height information G(x,y) and the background height information B(x,y), and performing binarization processing on the reconstructed point cloud based on the height difference; determining the minimum bounding rectangle in the reconstructed point cloud after binarization based on the direction perpendicular to the motion direction; determining the distance of the minimum bounding rectangle perpendicular to the motion direction as the width information W of the moving target, determining the distance of the minimum bounding rectangle in the motion direction as the length information L of the moving target; and determining the maximum value of the height difference as the height information H of the moving target.

[0158] Furthermore, after obtaining the reconstructed point cloud of the moving target, the method further includes: calculating the moving speed of the moving target by calculating the moving speed of the neighboring point clouds that have undergone the above registration process, and obtaining the point cloud registration speed; and correcting the speed of the speed measuring radar by the point cloud registration speed.

[0159] Furthermore, the above method includes: obtaining the actual appearance dimensions of the moving target, including the width information W', length information L', and height information H' of the moving target; comparing the measured appearance dimensions of the dynamic target with the actual appearance dimensions corresponding to the dynamic target; and determining the dynamic target as a target to be processed if the comparison result includes one or more of the following: W is greater than W', L is greater than L', and H is greater than H'.

[0160] Furthermore, the acquisition of the actual appearance size of the moving target includes: responding to the lidar to determine that the dynamic target has entered the detection area, triggering the camera component to acquire an image of the moving target; performing image pixel compensation based on the measured speed to remove motion blur in the image; and performing identity recognition of the moving target based on the image after the removal process, and acquiring the actual appearance size of the moving target based on the recognized identity.

[0161] Memory 1102 may include one or more computer-readable storage media, which may be non-transitory. Memory 1102 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments of this disclosure, the non-transitory computer-readable storage media in memory 1102 are used to store at least one instruction, which is executed by processor 1101 to implement the methods in the embodiments of this disclosure.

[0162] In some embodiments, the electronic device 1100 further includes a peripheral device interface 1103 and at least one peripheral device. The processor 1101, memory 1102, and peripheral device interface 1103 can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface 1103 via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of a display screen 1104, a camera 1105, and an audio circuit 1106.

[0163] Peripheral device interface 1103 can be used to connect at least one I / O (Input / Output) related peripheral device to processor 1101 and memory 1102. In some embodiments of this disclosure, processor 1101, memory 1102, and peripheral device interface 1103 are integrated on the same chip or circuit board; in other embodiments of this disclosure, any one or two of processor 1101, memory 1102, and peripheral device interface 1103 can be implemented on separate chips or circuit boards. This disclosure does not specifically limit the scope of the embodiments.

[0164] Display screen 1104 is used to display a UI (User Interface). This UI may include graphics, text, icons, videos, and any combination thereof. When display screen 1104 is a touch display screen, it also has the ability to collect touch signals on or above its surface. These touch signals can be input as control signals to processor 1101 for processing. In this case, display screen 1104 can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments of this disclosure, there may be one display screen 1104, which serves as the front panel of electronic device 1100; in other embodiments, there may be at least two display screens 1104, respectively disposed on different surfaces of electronic device 1100 or in a folded design; in still other embodiments, display screen 1104 may be a flexible display screen, disposed on a curved or folded surface of electronic device 1100. Furthermore, display screen 1104 may also be configured as a non-rectangular irregular shape, i.e., a non-rectangular screen. The display screen 1104 can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0165] Camera 1105 is used to capture images or videos. Optionally, camera 1105 includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the electronic device, and the rear-facing camera is located on the back of the electronic device. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusing the main camera and the depth-sensing camera, panoramic shooting by fusing the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments of this disclosure, camera 1105 may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm light flash and a cool light flash, which can be used for light compensation at different color temperatures.

[0166] The audio circuit 1106 may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, and convert the sound waves into electrical signals that are input to the processor 1101 for processing. For stereo sound acquisition or noise reduction purposes, there may be multiple microphones, each located in a different part of the electronic device 1100. The microphone may also be an array microphone or an omnidirectional microphone.

[0167] Power supply 1107 is used to supply power to various components in electronic device 1100. Power supply 1107 can be alternating current, direct current, a disposable battery, or a rechargeable battery. When power supply 1107 includes a rechargeable battery, the rechargeable battery can be a wired rechargeable battery or a wireless rechargeable battery. A wired rechargeable battery is a battery that is charged via a wired line, and a wireless rechargeable battery is a battery that is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0168] The block diagram of the electronic device shown in this embodiment does not constitute a limitation on the electronic device 1100. The electronic device 1100 may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0169] In the description of this disclosure, it should be understood that the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance. Those skilled in the art can understand the specific meaning of the above terms in this disclosure based on the specific circumstances. Furthermore, in the description of this disclosure, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0170] The above description is merely a specific embodiment of this disclosure, but the scope of protection of this disclosure is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this disclosure should be included within the scope of protection of this disclosure. Therefore, equivalent variations made in accordance with the claims of this disclosure are still within the scope of this disclosure.

Claims

1. A device for determining the appearance dimensions of a dynamic target, characterized in that, The device includes: The acquisition module is used to determine the detection area and acquire multi-frame point cloud data and multiple measurement velocities of dynamic targets within the detection area; The compensation determination module is used to determine motion compensation based on the multiple measured velocities and the time points of acquiring the multiple frames of point cloud data. The registration module is used to perform registration processing on the multi-frame point cloud data based on the motion compensation to obtain the reconstructed point cloud of the dynamic target; A determination module is used to determine the measured external dimensions of the dynamic target based on the reconstructed point cloud and the direction of motion of the dynamic target; The compensation determination module is specifically used to determine T. Ci and T Vk The magnitude relationship determines the compensation mode, and the measured speed V is used as a basis. k The estimated speed is used to determine the compensation speed, and the compensation mode and the compensation speed are determined as the motion compensation. Where T Ci This indicates that the point cloud data C of the i-th frame is obtained. i The time point, N is a positive integer, i is a positive integer not greater than N, T Vk This indicates that the measured speed V is obtained. k The time point, M is a positive integer, and k is a positive integer not greater than M; The compensation determination module is specifically used for, When T Ci <T V0 and T V0 -T Ci <T0, perform motion compensation according to the compensation mode of uniform linear motion, and determine the compensation speed as T V0 The measured speed V0 obtained at the time point, where T0 is the first preset duration; When T Ci >T Vl And T Ci <T Vp Motion compensation is performed according to the compensation mode of uniform linear motion, and interpolation is used based on V. l and V p Interpolation is performed to obtain the compensated speed, where l and p are both positive integers less than M, and l is less than p, V l and V p T respectively Vl and T Vp Measurement speed acquired at specific time points; When T Ci >T Vk Motion compensation is performed according to the compensation mode of uniform linear motion, and the compensation speed corresponding to the first stage is T. Vk Measurement of velocity V obtained at time points k The compensation speed corresponding to the second stage is the estimated speed, wherein the estimated speed is based on V. k The point cloud registration speed is determined based on the time difference between two frames of point cloud data.

2. The determining device according to claim 1, characterized in that, The compensation mode is uniform linear motion or uniformly accelerated motion.

3. The determining device according to claim 1, characterized in that, The compensation determination module is also used to perform motion estimation velocity balancing processing on the estimated velocity using Kalman filtering.

4. The determining device according to claim 1, characterized in that, The acquisition module is also used to acquire the actual external dimensions of the dynamic target, including: the width information W', length information L', and height information H' of the dynamic target; The determining module is used to compare the measured appearance size of the dynamic target with the actual appearance size corresponding to the dynamic target; if the comparison result includes one or more of the following: W is greater than W', L is greater than L', and H is greater than H', the dynamic target is determined to be the target to be processed.

5. The determining device according to claim 4, characterized in that, The step of obtaining the actual appearance size of the dynamic target includes: responding to the lidar to determine that the dynamic target has entered the detection area, triggering the camera component to acquire an image of the dynamic target; performing image pixel compensation based on the measured speed to remove motion blur in the image; performing identity recognition of the dynamic target based on the image after the removal process, and obtaining the actual appearance size of the dynamic target based on the recognized identity.

6. The determining device according to claim 5, characterized in that, The acquisition module is also used to identify license plates in the image of the dynamic target.

7. A method for determining the appearance dimensions of a dynamic target, characterized in that, The method includes: The detection area is determined, and multi-frame point cloud data and multiple measurement velocities of dynamic targets within the detection area are acquired. Motion compensation is determined based on the time points at which the multiple measured velocities are acquired and the time points at which the multiple frames of point cloud data are acquired. Based on the motion compensation, the multi-frame point cloud data is registered to obtain the reconstructed point cloud of the dynamic target; The measured external dimensions of the dynamic target are determined based on the reconstructed point cloud and the direction of motion of the dynamic target; The measured external dimensions of the dynamic target are compared with the actual external dimensions corresponding to the dynamic target; The step of determining motion compensation based on the time points for acquiring the multiple measured velocities and the time points for acquiring the multiple frames of point cloud data includes: determining motion compensation based on T... Ci and T Vk The magnitude relationship determines the compensation mode, and the measured speed V is used as a basis. k The estimated speed is used to determine the compensation speed, and the compensation mode and the compensation speed are determined as the motion compensation. According to T Ci and T Vk The magnitude relationship determines the compensation mode, and the measured speed V is used as a basis. k And the estimated speed determines the compensation speed, including: If T Ci <T V0 and T V0 -T Ci <T0, the motion compensation is performed according to the uniform linear motion compensation mode, and the compensation speed is determined as the measurement speed V0 obtained at the time point T V0 where T0 is the first preset duration; If T Ci >T Vl And T Ci <T Vp Then, motion compensation is performed according to the compensation mode of uniform linear motion, and interpolation is used based on V. l and V p Interpolation is performed to obtain the compensated speed, where l and p are both positive integers less than M, and l is less than p, V l and V p T respectively Vl and T Vp Measurement speed acquired at specific time points; If T Ci >T Vk Then, motion compensation is performed according to the uniform linear motion compensation mode, and the compensation velocity corresponding to the first stage is T. Vk Measurement of velocity V obtained at time points k The compensation speed corresponding to the second stage is the estimated speed, wherein the estimated speed is based on V. k The point cloud registration speed is determined based on the time difference between two frames of point cloud data.

Images