Roadside sensing equipment testing method, device, equipment and medium
By acquiring and matching detection information from roadside sensing devices, the number of device identifier changes and trajectory recall rate in the airport environment are evaluated, solving the problem of stability assessment of sensing devices in complex environments, realizing the quantification of device performance and deployment optimization, and improving the coverage and reliability of the sensing system.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-07
- Publication Date
- 2026-03-10
AI Technical Summary
Existing sensing devices lack stability assessment for target recognition tasks in airport environments, especially under complex electromagnetic environments and lighting changes, making it difficult to achieve high-precision sensing coverage across the entire area.
By acquiring detection information from roadside sensing devices, we can identify and match moving object markers and trajectories in the target area, calculate the number of marker changes, evaluate device performance by combining trajectory recall rate, identify areas with weak performance, and optimize device deployment.
It enables stability assessment of roadside sensing devices in complex scenarios, quantifies the tracking capabilities of the devices and identifies lost areas, guides the optimization of device installation and deployment, and improves the coverage and reliability of the sensing system.
Smart Images
Figure CN121640704A_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The present application relates to the technical field of vehicle-road cooperation, and in particular to a roadside perception device testing method, device, equipment and medium. BACKGROUND
[0002] In recent years, with the rapid development of unmanned driving technology, some airports have begun to explore the construction of few-people aprons, and by introducing unmanned vehicles to improve operational efficiency and reduce labor costs. However, there are significant differences between airport environments and public roads, with characteristics such as closed scenes, task-oriented, low-speed operation, and the existence of cross-activity areas of vehicles and aircraft, making the application of unmanned driving technology face unique challenges.
[0003] In the flight area, due to the limited installation conditions of the perception device (such as the high safety requirements of the aircraft activity area, the complex electromagnetic environment, etc.), it is difficult to achieve full-coverage high-precision perception. Therefore, airports usually adopt the technical route of "cloud planning, field cooperation, and vehicle autonomy", among which "field cooperation" deploys perception devices (such as cameras and radars) in key areas to build auxiliary perception systems to make up for the lack of vehicle perception capabilities.
[0004] However, the existing perception system lacks sufficient evaluation of the performance of the perception device when it is deployed. This is particularly reflected in the fact that it fails to fully evaluate its stability in target identification tasks. SUMMARY
[0005] Therefore, it is necessary to provide a roadside perception device testing method, device, equipment and medium to solve the problem of insufficient evaluation of the stability of the perception device in target identification tasks in the prior art.
[0006] To solve the above problems, in a first aspect, the present application provides a roadside perception device testing method, comprising: obtaining detection information of a target area by a roadside perception device; identifying the detection information to obtain the identification of a moving object in the target area and a detection trajectory; wherein when the moving object is re-identified, the moving object is assigned a new identification; matching an actual trajectory of an actual moving object in the target area with the detection trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory; determining the number of identification changes of the actual moving object in the detection process according to the identification of the target moving object corresponding to the target detection trajectory; determining the performance of the roadside perception device according to the number of identification changes.
[0007] In a possible implementation manner, the method further comprises: When the target detection trajectories are multiple, a region in which actual moving object recognition loss occurs in the target region is determined according to a matching result of the actual trajectory and the target detection trajectory.
[0008] In a possible implementation, the determining the performance of the roadside perception device according to the number of times of identity change comprises: determining a trajectory recall rate according to the matching result of the actual trajectory and the target detection trajectory; determining the performance of the roadside perception device according to the number of times of identity change and the trajectory recall rate.
[0009] In a possible implementation, the roadside perception device comprises a fisheye camera, and the detection information comprises a detection image; the method further comprises: dividing the detection image into multiple regions according to an image distortion degree; respectively determining distribution of the actual trajectory and the target detection trajectory in the multiple regions.
[0010] In a possible implementation, the method further comprises: displaying the target detection trajectory in the detection image.
[0011] In a possible implementation, the actual moving object comprises multiple vehicles; the method comprises: controlling the multiple vehicles to travel in the target region according to a planned test route in a following or parallel manner.
[0012] In a possible implementation, the roadside perception device comprises multiple, different roadside perception devices are used to detect different positions in the target region; the method further comprises: displaying a detection region of each of the roadside perception devices in a plan view of the target region, and displaying the target detection trajectories located in different detection regions in the plan view in different colors In a second aspect, the present application further provides a roadside perception device testing apparatus, comprising: a detection information acquisition module, configured to acquire detection information of a target region by a roadside perception device; an identification module, configured to identify the detection information to obtain an identity of a moving object in the target region and a detection trajectory; when the moving object is re-identified, the moving object is assigned a new identity; a matching module, configured to match an actual trajectory of an actual moving object in the target region with the detection trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory; An identification change number determination module is configured to determine the identification change number of the actual moving object in the detection process according to the identification of the target moving object corresponding to the target detection trajectory. A performance determination module is configured to determine the performance of the roadside perception device according to the identification change number.
[0013] In a third aspect, the present application further provides an electronic device, comprising a memory and a processor, wherein the memory is configured to store a program; the processor is coupled to the memory and is configured to execute the program stored in the memory to implement the steps of the roadside perception device testing method.
[0014] In a fourth aspect, the present application further provides a computer readable storage medium configured to store a computer readable program, wherein the program or instruction is executed by a processor to implement the steps of the roadside perception device testing method.
[0015] The present application has the following beneficial effects: The present application can quantitatively and accurately evaluate the continuous tracking capability of the roadside perception device in a complex scene. BRIEF DESCRIPTION OF DRAWINGS
[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.
[0017] Figure 1 A flowchart of an embodiment of the roadside perception device testing method provided by the present application; Figure 2 A flowchart of an embodiment of S102 in the present application; Figure 1 A flowchart of an embodiment of S102 in the present application;Figure 3 A schematic diagram of a structure of an embodiment of the roadside sensing device testing apparatus provided by the present invention; Figure 4 A schematic diagram of an embodiment of the electronic device provided by the present invention. Detailed Implementation
[0018] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0019] In the description of the embodiments of this invention, unless otherwise stated, "multiple" means two or more. "And / or" describes the relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following associated objects have an "or" relationship.
[0020] In the embodiments of this invention, the terms "first," "second," etc., are used to distinguish similar objects, and are not used to describe a specific order or sequence, nor to indicate or imply their relative importance or implicitly specify the number of technical features indicated. It should be understood that such data can be interchanged where appropriate so that embodiments of this application can be implemented in orders other than those illustrated or described herein, and the objects distinguished by "first," "second," etc., are generally of the same class, and the number of objects is not limited; for example, the first object can be one or more.
[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.
[0022] Reference Figure 1 The diagram illustrates a flowchart of an embodiment of the roadside sensing device testing method provided by the present invention, the method comprising: S101, Obtain detection information of the target area from the roadside sensing device.
[0023] Roadside sensing devices may include one or more, such as individual fisheye cameras, bullet cameras, lidar, millimeter-wave radar, etc., or combinations of fisheye cameras, bullet cameras, lidar, millimeter-wave radar, etc.
[0024] The target area can be the road sections within the airport that require vehicle-road cooperation.
[0025] The detection information can be dynamic data of the target area collected by roadside sensing devices, specifically video streams, radar point clouds, or laser scan data, used to extract the spatial location information of moving objects. The number of tag changes refers to the number of times the same object is assigned different identification tags during the detection process, which can be statistically calculated using trajectory matching algorithms to quantify the tracking stability of the equipment.
[0026] S102, the detection information is identified to obtain the identifier and detection trajectory of the moving object in the target area; wherein, when the moving object is re-identified, the moving object is assigned a new identifier.
[0027] The roadside sensing device testing system identifies moving objects at a certain sensing frequency (e.g., 10Hz) and assigns an ID to each identified moving object. For example, when a moving object appears, the system automatically assigns it the ID: xxxxxxx…0001. This moving object may be obscured by other vehicles / buildings during its journey, or the system may lose sight of it due to a sensing problem. Specifically, it might not be detected in frame XX. When the moving object with this ID reappears, the system will assign it a new ID: xxxx…0002. In other words, each time a moving object is re-identified after being lost, a new ID will be assigned to it.
[0028] Simultaneously, the roadside sensing device testing system will also identify the trajectories of moving objects with the same ID based on the detection information, thus obtaining the detection trajectory. The detection trajectory refers to the motion path formed by the device's continuous detection of the same moving object, which can be specifically realized through time series analysis of multi-frame data.
[0029] S103, match the actual trajectory of the moving object in the target area with the detected trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory.
[0030] The actual moving object can be a pedestrian or a vehicle. The actual moving object may be equipped with a positioning device, which allows the acquisition of its actual trajectory. There may be one or more actual moving objects. Matching the actual trajectories of the moving objects with the detected trajectories yields at least one target detection trajectory corresponding to each actual trajectory. Because the recognition process may be interrupted, one actual trajectory may correspond to multiple target detection trajectories; if the recognition process is not interrupted, one actual trajectory corresponds to only one target detection trajectory.
[0031] S104, Based on the identifier of the moving object corresponding to the target detection trajectory, determine the number of times the identifier of the actual moving object changes during the detection process.
[0032] For example, if the same actual trajectory corresponds to two target detection trajectories, and the two target detection trajectories each correspond to the identifiers of two moving objects, then the number of times the identifier of the actual moving object changes during the detection process is 1.
[0033] S105, determine the performance of roadside sensing devices based on the number of times the signage changes.
[0034] The more frequently the signage changes, the worse the performance of the roadside sensing equipment becomes.
[0035] The roadside sensing device testing method provided in this embodiment can be applied to a roadside sensing device testing system, which can be a software system running on a terminal device. The terminal device can be a tablet computer, in-vehicle device, augmented reality (AR) / virtual reality (VR) device, laptop computer, ultra-mobile personal computer (UMPC), netbook, personal digital assistant (PDA), mobile phone, etc. This embodiment does not impose any restrictions on the specific type of terminal device.
[0036] In summary, by introducing the number of identifier changes as an evaluation dimension, this embodiment can effectively identify the identity jump problem caused by roadside sensing devices during continuous target tracking, and is particularly suitable for evaluating the robustness of roadside sensing devices in complex scenarios such as occlusion and changes in lighting.
[0037] Furthermore, this embodiment can accurately quantify the stability indicators of roadside sensing devices in target tracking tasks, providing data support for optimizing the deployment location of roadside sensing devices. For example, if an abnormally high number of changes to a device's markings is detected at an airport runway entrance, the installation angle can be adjusted or auxiliary sensors can be added to improve the overall collaborative working capability of the sensing system.
[0038] In some embodiments of the present invention, the roadside sensing device testing method further includes: When there are multiple target detection trajectories, the area in the target region where the actual moving object was lost is determined based on the comparison results between the actual trajectory and the target detection trajectory.
[0039] Among them, multiple target detection trajectories refer to the fact that the actual moving object is assigned multiple different labels during the detection process, resulting in the same moving object being identified as multiple independent targets. For example, when a moving object briefly leaves the perception range and then re-enters, it may be assigned a new label.
[0040] The comparison result between the actual trajectory and the target detection trajectory refers to associating the actual trajectory with the target detection trajectory through a spatiotemporal matching algorithm, such as trajectory similarity calculation or time window alignment method. Identifying the lost region refers to the spatial range that exists in the actual trajectory but is not covered by the detected trajectory, which can be achieved through trajectory point missing statistics or region gridding analysis.
[0041] Specifically, within the target area, when the same actual moving object is detected as multiple target detection trajectories, the actual trajectory is spatiotemporally matched with all associated target detection trajectories to identify continuous time periods or spatial segments within the actual trajectory that are not covered by any target detection trajectory. For example, if the actual trajectory does not overlap with any target detection trajectory within a certain time period, the spatial location corresponding to that time period is marked as a region of lost detection. Furthermore, by overlaying the distribution of lost detection regions of multiple actual moving objects, high-incidence areas where sensing devices within the target area have detection blind spots or insufficient performance can be statistically identified.
[0042] This embodiment, through trajectory comparison and region mapping, can accurately identify the weak performance areas of roadside sensing devices in the target area, providing spatial data support for device deployment optimization.
[0043] This embodiment can quickly locate the specific locations in the target area where the roadside sensing device is prone to interruption of object recognition or trajectory breakage, such as in the electromagnetic interference area of the airport or the area of sudden change in lighting conditions. This can guide the adjustment of the device installation angle, optimization of the sensing range, or the deployment of additional sensors, effectively improving the coverage integrity and reliability of the auxiliary sensing system.
[0044] In some embodiments of the present invention, such as Figure 2 As shown, step S105 includes: S201. Determine the trajectory recall rate based on the comparison results between the actual trajectory and the target detection trajectory.
[0045] S202, determine the performance of roadside sensing devices based on the number of signage changes and trajectory recall rate.
[0046] The trajectory recall rate refers to the matching ratio between the actual trajectory of a moving object and the detected trajectory, or the ratio of the number of detected sensing trajectory points to the number of actual trajectory points. Specifically, it can be calculated using algorithms such as the least squares method, and is used to measure the completeness of the sensing device's continuous tracking of moving objects.
[0047] Specifically, within the target area, as the vehicle travels along a preset route, the sensing device continuously generates detection trajectories. The trajectory recall rate is calculated by comparing the actual driving trajectory with the detection trajectory over time. Simultaneously, the number of identification changes caused by occlusion or misidentification during the detection process is also recorded. Finally, the trajectory recall rate and the number of identification changes are weighted together to obtain a comprehensive performance score, thus avoiding bias caused by evaluating a single indicator.
[0048] This embodiment evaluates the continuous stability performance of the recognition process by combining two indicators: trajectory recall rate and number of identifier changes, which can improve the accuracy and comprehensiveness of the evaluation.
[0049] In some embodiments of the present invention, the roadside sensing device includes a fisheye camera, and the detection information includes a detection image; the roadside sensing device test further includes: The detection image is divided into multiple regions based on the degree of image distortion. The distribution of the actual trajectory and the target detection trajectory in multiple regions is determined separately.
[0050] Image distortion refers to the degree of image deformation caused by the optical characteristics of a fisheye camera lens. This can be quantified using geometric correction algorithms or mathematical models based on distortion coefficients, such as obtaining radial distortion parameters through a calibration plate. Dividing the detected image into multiple regions means segmenting the image into different sub-regions based on differences in distortion levels. This can be achieved using predefined distortion threshold segmentation methods or uniform segmentation methods based on grid partitioning, such as dividing the image's central region into a low-distortion region and the edge regions into high-distortion regions.
[0051] The distribution of actual trajectories and target detection trajectories refers to the coverage or density statistics of the actual trajectories and target detection trajectories in different regions of the detection image. Specifically, it can be achieved by mapping the coordinates of trajectory points to the corresponding regions and then counting and statistically analyzing them, or by generating heatmaps.
[0052] Specifically, in airport scenarios, fisheye cameras may suffer from severe distortion in edge regions due to installation limitations. By calculating the distortion of each pixel in the detected image, the image is divided into a low-distortion central region, a medium-distortion transition region, and a high-distortion edge region. After mapping the coordinates of the actual trajectory / target detection trajectory to the corresponding regions, the proportion of trajectory points and continuity indicators in each region are statistically analyzed. For example, in the high-distortion edge region, the proportion of actual trajectory points is significantly greater than that of target detection trajectory points, indicating a low recognition rate for the detected trajectory in this region. Furthermore, if the trajectory breakage rate in the high-distortion edge region is significantly higher than that in the central region, it indicates a deficiency in the perception stability of this region.
[0053] This embodiment can identify the differences in trajectory detection capabilities of fisheye cameras in different distortion areas, accurately locate the problem areas of trajectory breakage or deviation caused by image distortion, thereby guiding the adjustment of the installation position of sensing equipment or the supplementary deployment of auxiliary equipment, effectively improving the reliability of moving object detection in complex airport environments.
[0054] In some embodiments of the present invention, the roadside sensing device testing method further includes: The target detection trajectory is displayed in the detection image.
[0055] Specifically, in the detection images captured by the fisheye camera, image processing algorithms can be used to render and display the target detection trajectory in real time as an overlay layer. For example, in each frame of the detection image, continuous polylines or dynamic marker points are drawn based on the historical coordinates of the moving target, forming a trajectory visualization effect with time attributes. This allows for intuitive observation of the distribution characteristics of the target detection trajectory within the fisheye camera's field of view. Especially in areas with varying degrees of image distortion, the continuity of the trajectory and the stability of the markers can be analyzed through visual comparison.
[0056] In some embodiments of the present invention, the actual moving object includes multiple vehicles; the roadside sensing device testing method further includes: Control multiple vehicles to travel in a target area based on a planned test route, either following or in parallel.
[0057] Among them, multiple vehicles refer to multiple test vehicles used to simulate real traffic scenarios. Specifically, different models or sizes of vehicles can be used to cover diverse test needs. By introducing multi-vehicle cooperative driving, the recognition capabilities of roadside perception equipment in complex traffic flows can be more realistically reflected.
[0058] Following or parallel driving refers to a driving mode in which vehicles follow each other in front and behind or drive side by side. This can be achieved through preset formation control algorithms or vehicle-to-everything (V2X) cooperative systems to simulate interactive behaviors in complex traffic flows. This mode can effectively test the trajectory tracking stability of sensing devices in dense or intersecting traffic scenarios.
[0059] The planned test route refers to the pre-set vehicle driving path, which can be achieved by using high-precision maps combined with dynamic path generation technology.
[0060] In some embodiments of the present invention, the roadside sensing devices include multiple devices, and different roadside sensing devices are used to detect different locations in the target area; the method further includes: The plan view of the target area displays the detection areas of each roadside sensing device, and the target detection trajectories located in different detection areas are displayed in different colors on the plan view.
[0061] Among them, multiple roadside sensing devices refer to independent sensing devices deployed in or near different locations within the target area. Specifically, they can be implemented using devices such as fisheye cameras, lidar, or millimeter-wave radar to cover multiple sub-areas that need to be detected within the target area.
[0062] The detection area refers to the spatial range in which a single roadside sensing device can effectively collect data. Specifically, it can be determined by the device's field of view, detection distance, or installation location parameters. For example, the detection area of a fisheye camera can be calculated based on its installation height and lens distortion parameters.
[0063] Different color display refers to using color coding to distinguish the source of the trajectory based on the detection area to which the target detection trajectory belongs. Specifically, this can be achieved by using a preset color mapping table or dynamically generating a contrasting color scheme.
[0064] Reference Figure 3 The diagram illustrates a structural schematic of an embodiment of the roadside sensing device testing apparatus provided by the present invention. The apparatus includes: The detection information acquisition module is used to acquire the detection information of the target area by the roadside sensing device; The recognition module is used to identify the detection information and obtain the identifier and detection trajectory of the moving object in the target area; when the moving object is re-identified, the moving object is assigned a new identifier. The matching module is used to match the actual trajectory of the moving object in the target area with the detected trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory. The identifier change count determination module is used to determine the number of identifier changes of the actual moving object during the detection process based on the identifier of the moving object corresponding to the target detection trajectory. The performance determination module is used to determine the performance of roadside sensing devices based on the number of times the markings change.
[0065] It should be noted that the implementation principles or processes of the above modules can be referred to the aforementioned test methods for roadside sensing devices, and will not be elaborated here.
[0066] Reference Figure 4 The present invention illustrates an electronic device 400. The electronic device 400 includes a processor 401, a memory 402, and a display 403. Figure 4 Only some components of the electronic device 400 are shown, but it should be understood that it is not required to implement all the components shown, and more or fewer components may be implemented instead.
[0067] In some embodiments, processor 401 may be a central processing unit (CPU), microprocessor, or other data processing chip, used to run program code stored in memory 402 or process data, such as the roadside sensing device testing method of the present invention.
[0068] In some embodiments, processor 401 may be a single server or a group of servers. The server group may be centralized or distributed. In some embodiments, processor 401 may be local or remote. In some embodiments, processor 401 may be implemented on a cloud platform. In one embodiment, the cloud platform may include a private cloud, public cloud, hybrid cloud, community cloud, distributed cloud, internal cloud, multi-cloud, etc., or any combination thereof.
[0069] In some embodiments, memory 402 may be an internal storage unit of electronic device 400, such as a hard disk or memory of electronic device 400. In other embodiments, memory 402 may also be an external storage device of electronic device 400, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on electronic device 400.
[0070] Furthermore, the memory 402 may include both internal storage units of the electronic device 400 and external storage devices. The memory 402 is used to store application software and various types of data installed on the electronic device 400.
[0071] In some embodiments, display 403 may be an LED display, a liquid crystal display, a touch-sensitive liquid crystal display, or an OLED (Organic Light-Emitting Diode) touchscreen. Display 403 is used to display information from electronic device 400 and to display a visual user interface. Components 401-403 of electronic device 400 communicate with each other via a system bus.
[0072] In one embodiment, when the processor 401 executes the roadside sensing device test program in the memory 402, the following steps can be implemented: Acquire detection information of the target area from roadside sensing devices; The detection information is identified to obtain the identifiers and detection trajectories of moving objects in the target area; when a moving object is re-identified, it is assigned a new identifier. Match the actual trajectory of the moving object in the target area with the detected trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory; Based on the identifier of the moving object corresponding to the target detection trajectory, determine the number of times the identifier of the actual moving object changes during the detection process; The performance of roadside sensing devices is determined based on the number of times the signage changes.
[0073] It should be understood that when the processor 401 executes the roadside sensing device test program in the memory 402, in addition to the functions mentioned above, it can also perform other functions, as can be found in the description of the corresponding method embodiments above.
[0074] Furthermore, this embodiment of the invention does not specifically limit the type of electronic device 400 mentioned. Electronic device 400 can be a mobile phone, tablet computer, personal digital assistant (PDA), wearable device, laptop computer, or other portable electronic device. Exemplary embodiments of portable electronic devices include, but are not limited to, portable electronic devices running iOS, Android, Microsoft, or other operating systems. The aforementioned portable electronic device can also be other portable electronic devices, such as a laptop computer with a touch-sensitive surface (e.g., a touch panel). It should also be understood that in some other embodiments of the invention, electronic device 400 may not be a portable electronic device, but rather a desktop computer with a touch-sensitive surface (e.g., a touch panel).
[0075] In one embodiment, the present invention also provides a computer-readable storage medium storing a computer program that, when executed by the processor, implements the steps of any of the roadside sensing device testing methods described above.
[0076] Those skilled in the art will understand that all or part of the processes of the methods described in the above embodiments can be implemented by a computer program instructing related hardware, and the program can be stored in a computer-readable storage medium. The computer-readable storage medium may be a disk, optical disk, read-only memory, or random access memory, etc.
[0077] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.
Claims
1. A method of testing a roadside perception device, the method comprising: The method comprises: obtaining detection information of a target area by a roadside perception device; identifying the detection information to obtain an identification of a moving object in the target area and a detection trajectory; wherein when the moving object is re-identified, the moving object is assigned a new identification; matching an actual trajectory of an actual moving object in the target area with the detection trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory; determining a number of identification changes of the actual moving object in a detection process according to an identification of a target moving object corresponding to the target detection trajectory; determining a performance of the roadside perception device according to the number of identification changes.
2. The roadside perception device testing method of claim 1, wherein, The method further comprises: when the target detection trajectory is multiple, determining a region in which actual moving object identification loss occurs in the target area according to a comparison result of the actual trajectory and the target detection trajectory.
3. The method of testing a road-side perception device of claim 1, wherein, The determining of the performance of the roadside perception device according to the number of identification changes comprises: determining a trajectory recall rate according to the comparison result of the actual trajectory and the target detection trajectory; determining the performance of the roadside perception device according to the number of identification changes and the trajectory recall rate.
4. The method of testing roadside perception devices of claim 1, wherein, The roadside perception device comprises a fisheye camera, and the detection information comprises a detection image; the method further comprises: dividing the detection image into multiple regions according to an image distortion degree; respectively determining a distribution of the actual trajectory and the target detection trajectory in the multiple regions.
5. The method of testing a road-side perception device of claim 4, wherein, The method further comprises: displaying the target detection trajectory in the detection image.
6. The method of testing roadside perception devices of claim 1, wherein, The actual moving object comprises multiple vehicles; the method comprises: controlling the multiple vehicles to travel in the target area according to a planned test route in a following or parallel manner.
7. The method of testing roadside perception devices of claim 1, wherein, The roadside perception device comprises multiple, and different roadside perception devices are used to detect different positions in the target area; the method further comprises: displaying a detection region of each roadside perception device in a plan view of the target area, and displaying the target detection trajectories located in different detection regions in the plan view in different colors.
8. A roadside perception device testing apparatus, comprising: The method comprises: a detection information obtaining module configured to obtain detection information of a target area by a roadside perception device; an identifying module configured to identify the detection information to obtain an identification of a moving object in the target area and a detection trajectory; wherein when the moving object is re-identified, the moving object is assigned a new identification; a matching module configured to match an actual trajectory of an actual moving object in the target area with the detection trajectory to obtain at least one target detection trajectory corresponding to the actual trajectory; an identification change number determining module configured to determine a number of identification changes of the actual moving object in a detection process according to an identification of a target moving object corresponding to the target detection trajectory; a performance determining module configured to determine a performance of the roadside perception device according to the number of identification changes.
9. An electronic device, comprising: The device comprises a memory and a processor, wherein: the memory is configured to store a program; The processor is coupled with the memory and used to execute the program stored in the memory to realize the steps in the roadside perception device testing method of any one of claims 1 to 7.
10. A computer-readable storage medium, characterized in that, A computer readable program or instruction is stored, and the program or instruction is executed by the processor to realize the steps in the roadside perception device testing method of any one of claims 1 to 7.