Hierarchical recognition method and device for video sensor perception capability for traffic perception

Abstract

The present invention relates to the field of smart city geographic information service technology, and in particular to a hierarchical recognition method and device for video sensor perception capabilities for traffic perception. In view of the problem that the perception capabilities of video sensors are difficult to accurately evaluate, the present invention constructs a perception capability standard for video sensors for traffic element perception, and proposes a hierarchical recognition method and device for the theoretical maximum perception capability and instantaneous perception capability of video sensors. Compared with the existing video sensor capability evaluation models, the method of the present invention realizes a refined evaluation of the perception capability of video sensors through the idea of ​​stratification, and provides a new idea and method for the hierarchical recognition of the perception capability of video sensors for traffic perception, which is conducive to the accurate evaluation of the perception capability of video sensors under different observation elements and different perception accuracies, and can provide decision support for the capability optimization and collaboration of video sensors in traffic scenarios.

Description

Technical Field

[0001] The present invention relates to the technical field of smart city geographic information services, and in particular to a method and device for layered recognition of video sensor perception capabilities for traffic perception. Background Art

[0002] Video sensors are an important part of smart city perception infrastructure. They have the characteristics of all-weather, real-time, and continuous monitoring. They can obtain the target motion status within the sensor's field of view and are widely used in urban traffic perception and emergency perception of public emergencies. According to a report released by research firm IHS Markit, the number of video surveillance cameras worldwide will exceed 1 billion by 2021, with 560 million in China alone. Video sensors have become the most common event perception and recording tool. With the continuous development of video sensor technology, people's demand for video sensors is no longer limited to simple video monitoring and information extraction. The demand for fine cognition and quantitative calculation of video sensor perception capabilities is also increasingly urgent (such as the multi-target, multi-element, three-dimensional coverage optimization scheduling method for video sensor networks published by author Gao Fei at Nanjing Normal University in 2018).

[0003] Most of the current surveillance camera coverage models abstract the camera's instantaneous field of view into a fan, trapezoid or triangle (such as the three-dimensional coverage optimization method for indoor surveillance cameras published by author Wang Youwen in Nanjing Normal University in 2020). For PTZ cameras, their perception range can be described by an omnidirectional perception model, and for gun-type video sensors, they can be described by a directed coverage model (such as the directed sensor network coverage control algorithm published by authors Tao Dan and Ma Huadong in the Journal of Software in 2011, issue number: 22(10):2317-2334). In most studies, it is assumed that the visibility of every point in the field of view of the video sensor is the same, and the main focus is on covering the point, line and surface physical elements in the field of view. It cannot answer whether the monitoring elements can be seen clearly in the field of view and at what distance they can be seen clearly. There is a lack of accurate definition of the perception capability standard of video sensors (Cohen N, Gattuso J, MacLennan-Brown K. CCTV Operational Requirements Manual [M]. UK: Home Office Scientific Development Branch, 2009: 14-15), and it is impossible to distinguish the viewing effects of different elements in the field of view. Therefore, it is necessary to define the observation clarity standard for different target elements. In 1958, Johnson proposed the Johnson Criterion (J. Johnson, Analysis of image forming systems, in Proc. Image Intensifier Symp., pp. 249-273, Warfare Vision Branch, Electrical Engineering Department, USA Army Engineering Branch and Development Laboratories, Ft. Belvoir, VA, 1958.). Without considering the nature of the target and image defects, the sensor imaging system can determine the target recognition ability based on the resolution standard and the correlation between the pixels on the sensor. The equivalent fringes can be further converted into the number of effective pixels. The Johnson Criterion defines the specific standards for the detection, recognition and identification of target elements by video sensors. It is based on rigorous engineering experiments and has been widely used for decades (Gerald C Holst. Electro-Optical Imaging System Performance [M]. 4th, Beijing: National Defense Industry Press, 2015: 30-40).

[0004] As the problem of urban traffic congestion becomes increasingly serious, the perception needs for the main elements of urban traffic, namely people and vehicles, are gradually increasing. For urban people and vehicles, the diversity of actual needs is taken into account. For example, at key traffic nodes such as toll stations, detailed records of license plates, faces and other information are required, while in large areas such as regional sections and toll plazas, vehicle and personnel counting and basic vehicle model and target action recognition are sufficient. Therefore, the observation needs of different elements are different, and the camera capability range is also different. The perception capabilities of video sensors for people and vehicles are layered and quantitatively calculated, and the perception capability range of video sensors is accurately characterized, which is conducive to the comprehensive perception and timely response of various traffic elements and events; by analyzing areas with insufficient perception capabilities, decision support can also be provided for the optimal allocation of limited observation resources.

[0005] In summary, the main issues in understanding the perception capabilities of video sensors for traffic perception are:

[0006] Existing studies generally believe that target elements are covered if they are within the field of view of the video sensor, but do not care about the specific perception capabilities of specific target elements within the field of view, that is, whether they can be seen clearly and at what distance. Therefore, there is still a lack of an evaluation standard for the perception capabilities of specific target elements at different positions within the coverage range of the video sensor, making it impossible to accurately evaluate the effective coverage range of the video sensor in space. Summary of the invention

[0007] The technical problem to be solved by the present invention is mainly to provide an evaluation standard for the perception ability of specific target elements at different positions within the coverage range of the video sensor, so as to accurately evaluate the effective coverage range of the video sensor in space.

[0008] In order to achieve the above objectives, the present invention is based on the Johnson standard and adopts the idea of ​​capability stratification. By defining the distance standard of the sensor's perception capability for different elements within the coverage range, the capability levels are divided according to the effective observation range of the target elements, and the perception capability of the video sensor is hierarchically recognized and quantitatively calculated.

[0009] According to one aspect of the present invention, the present invention provides a hierarchical recognition method of video sensor perception capability for traffic perception, comprising the following steps:

[0010] S1: define the observation task of traffic perception, and determine the observation element type, perception accuracy and observation space range of the traffic scene according to the observation task;

[0011] S2: according to the observation element type, perception accuracy and observation space range, a perception capability standard of the video sensor is established based on the Johnson criterion, and the number of effective pixels corresponding to different observation tasks is obtained according to the perception capability standard;

[0012] S3: Calculating the theoretical maximum perception capability of the video sensor under a specific observation task based on the number of effective pixels corresponding to the different observation tasks;

[0013] S4: Based on the number of effective pixels corresponding to the different observation tasks, the instantaneous perception capability of the video sensor under the specific observation task is calculated according to the specific observation angle.

[0014] Preferably, step S1 includes the following sub-steps:

[0015] S1-1: define an observation task for traffic perception and obtain a road traffic scene corresponding to the observation task;

[0016] S1-2: Determine the observation element type of the traffic scene according to the road traffic scene, where the observation element type includes people and vehicles;

[0017] S1-3: According to different observation requirements of detection, recognition and identification of the observation element types, the perception accuracy of traffic observation elements is divided into three categories: low accuracy, medium accuracy and high accuracy; when the video sensor perceives with high accuracy, the identification information of the perception element, such as face information and license plate information, etc.; when the video sensor perceives with medium accuracy, the category identification information of the perception element, such as the distinction between people and cars; when the video sensor perceives with low accuracy, only counting is performed;

[0018] S1-4: Determine the spatial scope of observation according to the observation task, and perceive the observation elements within the spatial scope. The spatial scope is defined as the study area. Only after the spatial scope is determined can the perception capability of the video sensor be calculated.

[0019] Preferably, step S2 includes the following sub-steps:

[0020] S2-1: Determine the specific observation content of the observation element in a specific traffic scene based on the perception accuracy level of the observation element; specific traffic scenes include road intersections, highway toll stations, etc., and the observation content may include face information, license plate information, personnel movement information, vehicle model information, etc.;

[0021] S2-2: According to the target recognition level specified by Johnson's criterion, determine the corresponding relationship between different target recognition levels such as detection, recognition and discrimination and the observation accuracy of observation elements;

[0022] S2-3: According to the correspondence between the different target recognition levels and the observation accuracy of the observation elements, a video sensor perception capability standard is established to determine the number of effective pixels corresponding to different observation tasks.

[0023] Preferably, step S3 includes the following sub-steps:

[0024] S3-1: Based on the number of effective pixels corresponding to the different observation tasks, the video sensor perception capability distance of different elements is calculated according to the video sensor capability distance calculation formula; the video sensor perception capability distance calculation formula is:

[0025]

[0026] Where D is the video sensor's sensing distance, PPT is the number of target pixels, f is the camera's focal length, h is the object height, and Pixel size is the pixel size, which is the target surface size divided by the effective pixels;

[0027] S3-2: Calculate the theoretical maximum perception coverage of the video sensor in the traffic perception scenario according to the perception capability distance of the video sensor of the different elements;

[0028] S3-3: A buffer zone is made according to the perception capability distance of the video sensor of the different elements to represent the theoretical maximum perception coverage of the video sensor under a specific task; the theoretical maximum perception capability of the video sensor includes the perception capability distance and the theoretical maximum perception coverage.

[0029] Preferably, step S4 includes the following sub-steps:

[0030] S4-1: Based on the number of effective pixels corresponding to the different observation tasks, the projection parameters of the video sensor geometric model on the plane are calculated according to the instantaneous field of view model of the video sensor; the instantaneous field of view model adopts the directed coverage model of the video sensor, and the most commonly used fan-shaped, triangle and trapezoidal shapes are selected as the basis for calculating the instantaneous field of view projection coordinates or angles; when a fan-shaped instantaneous field of view is used, only the horizontal FOV (Field of View) angle is calculated, and when a trapezoidal or triangular instantaneous field of view is used, the projection coordinates of the FOV and the trapezoidal vertex need to be calculated;

[0031] S4-2: Obtaining the instantaneous field of view coverage of the video sensor according to the projection parameters, and further calculating the area of ​​the instantaneous field of view;

[0032] S4-3: Based on the perception capability distance of the video sensor and the projection parameters, a visualization diagram is made to represent the instantaneous perception coverage of the video sensor under a specific task; the instantaneous perception capability of the video sensor includes the perception capability distance and the instantaneous perception coverage.

[0033] According to another aspect of the present invention, a video sensor perception capability layered recognition device for traffic perception is provided, comprising the following modules:

[0034] An observation task definition module is used to define the observation task of traffic perception and determine the observation element type, perception accuracy and observation space range of the traffic scene according to the observation task;

[0035] A perception capability standard establishment module is used to establish a perception capability standard of the video sensor based on the Johnson criterion according to the observation element type, perception accuracy and observation space range, and obtain the number of effective pixels corresponding to different observation tasks according to the perception capability standard;

[0036] A maximum perception capability calculation module, used to calculate the theoretical maximum perception capability of the video sensor under a specific observation task based on the number of effective pixels corresponding to the different observation tasks;

[0037] The instantaneous perception capability calculation module is used to calculate the instantaneous perception capability of the video sensor under a specific observation task according to the specific observation angle based on the number of effective pixels corresponding to the different observation tasks.

[0038] The beneficial effects brought by the technical solution provided by the present invention are:

[0039] (1) A video sensor perception capability standard has been formed. A hierarchical cognitive method for video sensor perception capability suitable for traffic scene perception has been established, which is conducive to the effective evaluation of the coverage quality of different locations within the video sensor perception range and can provide decision support for the capability evaluation and collaboration of video sensors in traffic scenes.

[0040] (2) The quantitative calculation and visual expression of the theoretical maximum perception capability and instantaneous perception capability of video sensors are realized. The quantitative calculation of the theoretical maximum perception capability and instantaneous perception capability of sensors in traffic scenarios is realized through a hierarchical cognition method, which is conducive to the refined cognition of the perception capability of video sensors and can visualize the theoretical maximum coverage capability and instantaneous coverage of different sensor types. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] The present invention will be further described below with reference to the accompanying drawings and embodiments, in which:

[0042] Figure 1 It is an overall flow chart of a hierarchical recognition method of video sensor perception capability for traffic perception in an embodiment of the present invention;

[0043] Figure 2 is a specific flow chart of step S1 in an embodiment of the present invention;

[0044] Figure 3 is a specific flow chart of step S2 in an embodiment of the present invention;

[0045] Figure 4 is a specific flow chart of step S3 in an embodiment of the present invention;

[0046] Figure 5 is a specific flow chart of step S4 in an embodiment of the present invention;

[0047] Figure 6 This is a visualization example of the theoretical maximum coverage capability in an embodiment of the present invention;

[0048] Figure 7 is an example of visualization of instantaneous coverage capability in an embodiment of the present invention;

[0049] Figure 8 This is an example of visualization of personnel counting of instantaneous coverage capability in an embodiment of the present invention;

[0050] Fig. 9 It is a structural diagram of a video sensor perception capability hierarchical recognition device for traffic perception in an embodiment of the present invention. DETAILED DESCRIPTION

[0051] The proposed method of video sensor perception capability hierarchical recognition for traffic perception fully considers the different coverage of video sensors at different locations in space, provides hierarchical division standards for video sensor monitoring capabilities, and constructs observation indicators for human, vehicle, and object elements in general scenarios. By calculating the capability distance to determine the coverage rate at different levels and quantifying the capability level of video sensors, it is beneficial to improve the monitoring efficiency of limited observation resources.

[0052] The present invention will be further described below with reference to specific embodiments.

[0053] Wuhan West Toll Station, as the western gate of Wuhan City, is a key transportation node in Wuhan City.

[0054] The following describes in detail a method for calculating the hierarchical capability of a video sensor proposed by the present invention in combination with the above application scenarios and the accompanying drawings. The overall process is shown in FIG. Figure 1 ,A hierarchical recognition method for video sensor perception ,capability for traffic perception includes the following steps:

[0055] Step S1: Define the observation task of traffic perception, and determine the observation element type, perception accuracy, and observation space range according to the observation task. The specific process is as follows: Figure 2 As shown;

[0056] S1-1: The observation task of traffic perception is defined as the traffic scene video sensor capability perception of Wuhan West Toll Station, and the traffic scene of Wuhan West Toll Station is obtained;

[0057] S1-2: The observation element types are determined as people and vehicles, that is, vehicles and people within the scope of Wuhan West Toll Station.

[0058] S1-3: The human and vehicle elements of Wuhan West Toll Station have the needs of perception, recognition and identification. According to different perception needs, the perception accuracy is divided into three categories: low precision, medium precision and high precision.

[0059] S1-4: For the perception of traffic scene video sensor capabilities, the spatial scope is defined as the entire spatial scope of the Wuhan West Toll Station Square from the entrance to the exit, that is, the human and vehicle elements within the entire toll station spatial scope are perceived.

[0060] Step S2: According to the type of observation elements, perception accuracy, and observation space range, the capability standard of specific observation tasks is established based on the Johnson criterion to determine the number of effective pixels that meet the requirements of different observation tasks. The specific process is as follows: Figure 3 shown.

[0061] S2-1: According to the perception accuracy level of traffic elements, the specific observation contents of people and vehicles at Wuhan West Toll Station are defined as follows: low precision: people counting, vehicle counting; medium precision: people movement information, vehicle model information; high precision: face information, license plate information.

[0062] S2-2: According to the target recognition level specified by the Johnson criterion, determine the corresponding relationship between different levels and the perception accuracy of traffic elements. The three levels of detection, recognition, and identification in the Johnson criterion correspond to the low precision, medium precision, and high precision of traffic scenes, respectively, and determine the number of effective pixels in traffic scenes without considering the signal-to-noise ratio of the target and weather conditions.

[0063] S2-3: Establish the perception capability standard of video sensors in traffic perception scenarios and determine the number of effective pixels corresponding to different observation tasks (observation contents). Based on the target recognition level specified by the Johnson criterion, combined with the requirements of human imaging and vehicle recognition, define the number of effective pixels at different accuracies to form the perception capability standard of video sensors for human and vehicle elements, as shown in Table 1.

[0064] Table 1. Video sensor perception capability standards for human and vehicle elements

[0065]

[0066] Step S3: Calculate the theoretical maximum perception capability of the video sensor under a specific observation task. The specific process is as follows: Figure 4 shown.

[0067] S3-1: Based on the number of effective pixels corresponding to traffic elements at different accuracies, the perception capability distance of the corresponding video sensor is obtained through the perception capability distance calculation formula of the video sensor, as shown in Table 2. Among them, the existing resource number is the observable video sensor resources set up at Wuhan West Toll Station.

[0068] Table 2. Video sensor perception distance

[0069]

[0070] S3-2: According to the sensor's sensing distance, the theoretical maximum coverage of human and vehicle elements within the Wuhan West Toll Station is calculated, as shown in Table 3. Figure 6 shown.

[0071] Table 3. Calculation results of the theoretical maximum coverage of video sensors

[0072]

[0073] S3-3: A buffer is created based on the perception capability distance obtained in S3-1 to represent the theoretical maximum coverage of the video sensor under a specific observation task.

[0074] Step S4: Calculate the instantaneous perception capability of the video sensor under a specific observation task according to the specific observation angle. The specific process is as follows: Figure 5 shown.

[0075] S4-1: Calculate the projection parameters of the geometric model on the plane according to the instantaneous field of view model of the video sensor. In this example, the sector instantaneous field of view model is selected. When the horizontal field of view angle FOV is fixed in focus, the projection parameters of the three existing resource video sensors are respectively the video sensor A plane coordinates (30.4581N, 114.0608E), field of view angle AFOV = 25°, sensor B plane coordinates (30.4583N, 114.0594E), field of view angle BFOV = 25°, sensor C plane coordinates (30.4589N, 114.0594E), field of view angle CFOV = 30°.

[0076] S4-2: Calculate the area of ​​the instantaneous field of view based on FOV. The instantaneous field of view range depends not only on the sensor itself, but also on the shape of the study area. The instantaneous field of view coverage of human and vehicle elements at high, medium and low precision is shown in Table 4. Figure 7 shown.

[0077] Table 4. Calculation results of instantaneous coverage of video sensors

[0078]

[0079] S4-3: Based on the perception distance obtained in S3-1 and the projection parameters obtained in S4-1, characterize the instantaneous perception range of the video sensor under a specific task, such as Figure 8 shown.

[0080] As an optional implementation, this embodiment uses ArcGIS 10.2 software to calculate the instantaneous field of view area of ​​the video sensor. The specific operation process is as follows:

[0081] Data entry;

[0082] Select the appropriate projection;

[0083] Draw the corresponding sector according to the sector projection parameters. Here we use the ArcGIS secondary development interface. You can draw by inputting the sector coordinates, FOV and corresponding angle.

[0084] Use ArcGIS area calculation function and select the unit as "square kilometers" to calculate the current projection area;

[0085] Get the intersection of the sector layer and the study area layer for clipping;

[0086] Continue to call the area calculation function to calculate the instantaneous field of view area within the study area;

[0087] Use the Field Calculator to calculate instantaneous coverage.

[0088] refer to Fig. 9 This embodiment also provides a video sensor perception capability layered recognition device for traffic perception, including the following modules:

[0089] An observation task definition module 1 is used to define the observation task of traffic perception, and determine the observation element type, perception accuracy and observation space range of the traffic scene according to the observation task;

[0090] A perception capability standard establishing module 2 is used to establish a perception capability standard of the video sensor based on the Johnson criterion according to the observation element type, perception accuracy and observation space range, and obtain the number of effective pixels corresponding to different observation tasks according to the perception capability standard;

[0091] The maximum perception capability calculation module 3 is used to calculate the theoretical maximum perception capability of the video sensor under a specific observation task based on the number of effective pixels corresponding to the different observation tasks;

[0092] The instantaneous perception capability calculation module 4 is used to calculate the instantaneous perception capability of the video sensor under a specific observation task according to the specific observation angle based on the number of effective pixels corresponding to the different observation tasks.

[0093] The beneficial effects after the implementation of the present invention are:

[0094] (1) The traditional video sensor perception capability perception regards the visibility of all points within the field of view as consistent, while the present invention establishes a video sensor perception capability perception standard to provide a basis for perceiving the detailed visible content within the observation range of the video sensor.

[0095] (2) Traditional video sensor perception capability perception, theoretical maximum coverage and instantaneous coverage often only consider one of the two situations, while this paper considers both of the above situations at the same time, performs perception capability calculation and visualization, and makes the sensor perception capability more specific and intuitive. It can improve the efficiency of auxiliary decision-making and improve the efficiency of event processing in traffic scenarios.

[0096] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0097] The serial numbers of the embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. In a unit claim that lists several means, several of these means may be embodied by the same hardware item. The use of the words first, second, and third, etc. does not indicate any order and these words may be interpreted as identifiers.

[0098] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A hierarchical recognition method for video sensor perception capabilities for traffic perception, characterized in that: The following steps are involved: S1: define the observation task of traffic perception, and determine the observation element type, perception accuracy and observation space range of the traffic scene according to the observation task; S2: according to the observation element type, perception accuracy and observation space range, a perception capability standard of the video sensor is established based on the Johnson criterion, and the number of effective pixels corresponding to different observation tasks is obtained according to the perception capability standard; Step S2 includes the following sub-steps: S2-1: Determine the specific observation content of the observation elements in a specific traffic scenario based on the perception accuracy level of the observation elements; S2-2: According to the target recognition level specified by Johnson's criterion, determine the corresponding relationship between different target recognition levels and the observation accuracy of the observation elements; S2-3: According to the correspondence between the different target recognition levels and the observation accuracy of the observation elements, the perception capability standard of the video sensor is established to determine the number of effective pixels corresponding to different observation tasks; S3: Calculating the theoretical maximum perception capability of the video sensor under a specific observation task based on the number of effective pixels corresponding to the different observation tasks; Step S3 includes the following sub-steps: S3-1: Based on the number of effective pixels corresponding to the different observation tasks, the video sensor perception capability distance of different elements is calculated according to the video sensor capability distance calculation formula; the video sensor perception capability distance calculation formula is: Where D is the video sensor's sensing distance, PPT is the number of target pixels, f is the camera's focal length, h is the object height, and Pixel size is the pixel size, which is the target surface size divided by the effective pixels; S3-2: Calculate the theoretical maximum perception coverage of the video sensor in the traffic perception scenario according to the perception capability distance of the video sensor of the different elements; S3-3: Buffering is done based on the perception capability distance of the video sensor of the different elements, representing the theoretical maximum perception coverage of the video sensor under a specific observation task; S4: Based on the number of effective pixels corresponding to the different observation tasks, the instantaneous perception capability of the video sensor under the specific observation task is calculated according to the specific observation angle; Step S4 includes the following sub-steps: S4-1: Based on the number of effective pixels corresponding to the different observation tasks, the projection parameters of the video sensor geometric model on the plane are calculated according to the instantaneous field of view model of the video sensor; S4-2: Obtaining the instantaneous field of view coverage of the video sensor according to the projection parameters, and further calculating the area of ​​the instantaneous field of view; S4-3: Based on the perception capability distance of the video sensor and the projection parameters, a visualization diagram is made to represent the instantaneous perception coverage of the video sensor under a specific observation task.

2. The method for layered recognition of video sensor perception capabilities for traffic perception as claimed in claim 1, characterized in that: Step S1 includes the following sub-steps: S1-1: define an observation task for traffic perception and obtain a road traffic scene corresponding to the observation task; S1-2: Determine the observation element type of the traffic scene according to the road traffic scene, where the observation element type includes people and vehicles; S1-3: According to different observation requirements of the observation element types, the perception accuracy of the traffic observation elements is divided into three categories: low accuracy, medium accuracy and high accuracy; S1-4: Determine the spatial scope of observation according to the observation task, and perceive the observation elements within the spatial scope.

3. A video sensor perception capability layered recognition device for traffic perception, characterized in that: Includes the following modules: An observation task definition module is used to define the observation task of traffic perception and determine the observation element type, perception accuracy and observation space range of the traffic scene according to the observation task; The perception capability standard establishment module is used to establish the perception capability standard of the video sensor according to the observation element type, perception accuracy and observation space range based on the Johnson criterion, and obtain the number of effective pixels corresponding to different observation tasks according to the perception capability standard; the specific process is: S2-1: Determine the specific observation content of the observation elements in a specific traffic scenario based on the perception accuracy level of the observation elements; S2-2: According to the target recognition level specified by Johnson's criterion, determine the corresponding relationship between different target recognition levels and the observation accuracy of the observation elements; S2-3: According to the correspondence between the different target recognition levels and the observation accuracy of the observation elements, the perception capability standard of the video sensor is established to determine the number of effective pixels corresponding to different observation tasks; The maximum perception capability calculation module is used to calculate the theoretical maximum perception capability of the video sensor under a specific observation task based on the number of effective pixels corresponding to the different observation tasks; the specific process is: S3-1: Based on the number of effective pixels corresponding to the different observation tasks, the video sensor perception capability distance of different elements is calculated according to the video sensor capability distance calculation formula; the video sensor perception capability distance calculation formula is: Where D is the video sensor's sensing distance, PPT is the number of target pixels, f is the camera's focal length, h is the object height, and Pixel size is the pixel size, which is the target surface size divided by the effective pixels; S3-2: Calculate the theoretical maximum perception coverage of the video sensor in the traffic perception scenario according to the perception capability distance of the video sensor of the different elements; S3-3: Buffering is done based on the perception capability distance of the video sensor of the different elements, representing the theoretical maximum perception coverage of the video sensor under a specific observation task; The instantaneous perception capability calculation module is used to calculate the instantaneous perception capability of the video sensor under a specific observation task according to the specific observation angle based on the number of effective pixels corresponding to the different observation tasks; the specific process is: S4-1: Based on the number of effective pixels corresponding to the different observation tasks, the projection parameters of the video sensor geometric model on the plane are calculated according to the instantaneous field of view model of the video sensor; S4-2: Obtaining the instantaneous field of view coverage of the video sensor according to the projection parameters, and further calculating the area of ​​the instantaneous field of view; S4-3: Based on the perception capability distance of the video sensor and the projection parameters, a visualization diagram is made to represent the instantaneous perception coverage of the video sensor under a specific observation task.

Images