Vehicle surrounding target pedestrian detection method, and parking method and device

Abstract

The application relates to a vehicle surrounding target pedestrian detection method, a car moving method and a device, which are applied to automobile electronic products, and the method comprises the following steps: monitoring a monitoring area of a current vehicle in real time, and obtaining a target track; judging whether a target pedestrian exists in the monitoring area according to the target track; if the judgment result is true, the monitoring area has the target pedestrian; otherwise, no processing is performed. Through the vehicle surrounding target pedestrian detection method, whether a target vehicle owner exists around the vehicle is judged, the owner is informed, and then the owner selects whether to connect a vehicle communication module to communicate; the first-layer communication initiative right is grasped on the electronic rearview mirror, and the problem that the owner is disturbed is avoided.

Description

Technical Field

[0001] This application relates to the field of automotive electronics technology, and more specifically, to a method and apparatus for detecting pedestrians around a vehicle, and for moving a vehicle. Background Technology

[0002] With the rapid development of automotive technology, more and more people prefer to use vehicles for transportation when running errands. In urban areas, drivers may need to temporarily park on the side of the road due to special circumstances such as paying bills at the bank or buying small items. However, this parking may result in illegal parking or obstruction of other vehicles. Patent CN203496774U, "Rearview Mirror with Integrated Communication Function," describes a rearview mirror with an integrated communication function. The mirror housing has a communication button, an external speaker, and a microphone. An electrical connection cable extends from the mirror to connect to a mobile communication module. The driver, standing outside the car, can press the communication button to communicate with the car owner via the mobile communication module. However, the communication button is exposed outside the car, making it easily accessible to anyone and potentially harassing the driver. Furthermore, adding a mobile communication module increases the vehicle's cost. Summary of the Invention

[0003] To overcome the problems in the prior art, this application provides a method for detecting pedestrians around a vehicle, a method for moving a vehicle, and an apparatus for doing so.

[0004] A method for detecting pedestrians around a vehicle, applied in automotive electronic products, the method comprising:

[0005] Real-time monitoring of the current vehicle's monitoring area and acquisition of the target trajectory;

[0006] Based on the target trajectory, determine whether there is a target pedestrian in the monitored area;

[0007] If the determination result is true, then there is a target pedestrian in the monitored area;

[0008] Otherwise, no action will be taken.

[0009] Optionally, the detection area of ​​the current vehicle is monitored in real time, and the target trajectory is obtained;

[0010] The monitoring area is monitored by the left and / or right cameras installed on the current vehicle. The image information of the monitoring area is acquired in real time, and the image signal is identified to determine whether the image signal contains the characteristics of the target pedestrian.

[0011] If the image signal contains features of the target pedestrian, the target trajectory is obtained based on the continuous image signal;

[0012] Otherwise, no action will be taken.

[0013] Optionally, determining whether a target pedestrian exists in the monitored area based on the target trajectory includes:

[0014] The similarity between the target trajectory and the preset trajectory is calculated, and the similarity value is obtained;

[0015] If the similarity value is less than the preset similarity value, it is determined that there is a target pedestrian in the current vehicle detection area; otherwise, no action is taken.

[0016] Optionally, the similarity calculation between the target trajectory and the preset trajectory is performed using the Fraser distance algorithm or the Hausdorff distance algorithm.

[0017] Optionally, the target trajectory includes a left trajectory captured by the left camera, and the preset trajectory includes a first preset trajectory; the step of calculating the similarity between the target trajectory and the preset trajectory and obtaining a similarity value includes:

[0018] The left trajectory is AB, the continuous function of the left trajectory in variable t is y(t), and the spatial position of the left trajectory is represented by f(y(t)).

[0019] The first preset trajectory is A'B', the continuous function of the first preset trajectory in variable t is y′(t), and the spatial position of the first preset trajectory is represented by f(y′(t)).

[0020] The similarity between the left trajectory and the first preset trajectory is calculated using the Frescher distance algorithm to obtain the first similarity value, which is given by the following formula:

[0021]

[0022] Where M is the length of the left trajectory, M′ is the length of the first preset trajectory, and d((f(y(t)),f′(y′(t))) is the distance between f(y(t)) and f′(y′(t)).

[0023] Optionally, the target trajectory includes a right trajectory captured by the right camera, and the preset trajectory includes a second preset trajectory; the step of calculating the similarity between the target trajectory and the preset trajectory and obtaining a similarity value includes:

[0024] The right trajectory is CD, the continuous function of the right trajectory in variable t is g(t), and the spatial position of the right trajectory is represented by f(g(t)).

[0025] The second preset trajectory is A'B'=N, the continuous function of the second preset trajectory in variable t is g′(t), and the spatial position of the second preset trajectory is represented by f(g′(t)).

[0026] The similarity between the right trajectory and the second preset trajectory is calculated using the Frescher distance algorithm to obtain the second similarity value, which is given by the following formula:

[0027]

[0028] Where N is the length of the right trajectory, N′ is the length of the second preset trajectory, and d((f(g(t)),f′(g′(t))) is the distance between f(g(t)) and f′(g′(t)).

[0029] Optionally, determining whether a target pedestrian exists in the monitored area based on the target trajectory further includes:

[0030] Calculate the average velocity V of the left trajectory AB and the right trajectory CD, where, V ab V represents the average data of the left trajectory. cd The average velocity of the right trajectory;

[0031] Let the lengths of endpoint B of the left trajectory and endpoint D of the right trajectory be L, then determine... Whether it is true or not, where T b Let T be the time it takes for the pedestrian to arrive at B. d Let Δt be the time when the pedestrian arrives at D, and let Δt be the preset time.

[0032] Optionally, determining whether a target pedestrian exists in the monitored area based on the target trajectory includes:

[0033] If the first similarity value is less than the first preset similarity value, the second similarity value is less than the second preset similarity value, or any one or more of the formula (4), then it is determined that there is a target pedestrian in the monitoring area.

[0034] In addition, this application also discloses a method for moving a car, the method comprising:

[0035] The above-mentioned method for detecting pedestrians around vehicles is used to determine whether there are target pedestrians in the monitored area.

[0036] If a target pedestrian is present, a notification is sent to the target terminal, along with the field-of-view images from the left and right cameras. The target terminal then confirms whether to connect to the vehicle's communication module for a call.

[0037] In addition, this application also discloses a device, which is an electronic rearview mirror installed on both sides of a vehicle, or a communication camera installed on the left and right rearview mirrors; the device is provided with a main control module, a communication module, a camera, a loudspeaker, and a microphone that are interconnected; the main control module operates the above-mentioned vehicle relocation method.

[0038] Compared with existing technologies, the beneficial effects of this application are as follows: This application uses a target pedestrian detection method around the vehicle to determine whether there is a target driver nearby, notifies the driver, and then allows the driver to choose whether to connect to the vehicle's communication module for communication; this puts the first layer of communication initiative in the electronic rearview mirror, avoiding the problem of driver harassment. Furthermore, the driver can further determine whether other vehicles are blocked based on the images transmitted from the electronic rearview mirror, putting the second layer of communication initiative in the driver's hands. This double protection enhances customer satisfaction with the product. Attached Figure Description

[0039] Figure 1 This is a flowchart of an embodiment of this application.

[0040] Figure 2 This is a field-of-view diagram of the device according to an embodiment of this application.

[0041] Figure 3 This is a device monitoring status diagram according to an embodiment of this application.

[0042] Figure 4 This is a schematic diagram of the device structure according to an embodiment of this application.

[0043] Figure 5 This is a schematic diagram of the target trajectory and the preset trajectory in an embodiment of this application. Detailed Implementation

[0044] The present application will be further described below with reference to specific embodiments.

[0045] In the accompanying drawings of the embodiments of this application, the same or similar reference numerals correspond to the same or similar components. In the description of this application, it should be understood that if terms such as "upper", "lower", "left", "right", "top", "bottom", "inner", "outer" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, they are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, the terms used to describe positional relationships in the drawings are only for illustrative purposes and should not be construed as limiting this patent.

[0046] Furthermore, if terms such as "first" or "second" are used for descriptive purposes only, they are mainly used to distinguish different devices, components or parts (the specific types and structures may be the same or different), and are not used to indicate or imply the relative importance or quantity of the indicated devices, components or parts, and should not be construed as indicating or implying relative importance.

[0047] In such Figure 1-5 In the illustrated embodiment, this application provides a method for detecting pedestrians around a vehicle, applied in automotive electronic products. The method includes:

[0048] 100. Real-time monitoring of the current vehicle's monitoring area and acquisition of the target trajectory; In step 100, real-time monitoring of the current vehicle's detection area and acquisition of the target trajectory; Monitoring is performed in the monitoring area by the left and / or right cameras installed on the current vehicle, real-time acquisition of image information of the monitoring area, and identification of the image signals to determine whether the image signals contain the characteristics of the target pedestrian; If the image signals contain the characteristics of the target pedestrian, the target trajectory is acquired based on the continuous image signals; otherwise, no processing is performed.

[0049] 200. Based on the target trajectory, determine whether there is a target pedestrian in the monitoring area; in step 200, determining whether there is a target pedestrian in the monitoring area based on the target trajectory includes: calculating the similarity between the target trajectory and the preset trajectory, and obtaining the similarity value; if the similarity value is less than the preset similarity value, it is determined that there is a target pedestrian in the current vehicle detection area, otherwise no processing is performed. Or, it can be determined by combining formula (4) to determine whether there is a target pedestrian in the current vehicle detection area.

[0050] 300. If the result is true, then there is a target pedestrian in the monitored area.

[0051] 400, otherwise no action will be taken.

[0052] In this embodiment, the application uses a vehicle-around-the-vehicle target pedestrian detection method to determine whether there is a target vehicle owner around the vehicle, notifies the owner, and then allows the owner to choose whether to connect to the vehicle's communication module for communication. This places the first layer of communication control in the electronic rearview mirror, avoiding the problem of the owner being harassed. Furthermore, the owner can further determine whether other vehicles are blocked based on the image transmitted from the electronic rearview mirror, placing the second layer of communication control in the owner's hands. This dual protection enhances customer satisfaction with the product.

[0053] In some embodiments, in step 100, the detection area of ​​the current vehicle is monitored in real time, and the target trajectory is acquired. The monitoring area is monitored using the left and / or right cameras located on the current vehicle, and image information of the monitored area is acquired in real time. The image signals are then identified to determine whether they contain the characteristics of a target pedestrian. If the image signals contain the characteristics of a target pedestrian, the target trajectory is acquired based on the continuous image signals; otherwise, no processing is performed. In this embodiment, the application identifies surrounding pedestrians using the left and / or right cameras. The identification method can be any known person identification method. Simultaneously, based on age or other characteristics, a preliminary judgment is made as to whether the person is likely to be stuck in traffic. If, after preliminary screening, it is determined that the person is likely to be stuck in traffic, the trajectory of that person is tracked.

[0054] In some embodiments, in step 200, determining whether a target pedestrian exists in the monitored area based on the target trajectory includes:

[0055] The similarity between the target trajectory and the preset trajectory is calculated, and the similarity value is obtained.

[0056] If the similarity value is less than the preset similarity value, it is determined that there is a target pedestrian in the current vehicle detection area; otherwise, no action is taken.

[0057] The similarity calculation is performed between the target trajectory and the preset trajectory using either the Fraser distance algorithm or the Hausdorff distance algorithm.

[0058] In one embodiment of the above examples, the target trajectory includes a left trajectory captured by the left camera, and the preset trajectory includes a first preset trajectory; calculating the similarity between the target trajectory and the preset trajectory and obtaining a similarity value includes:

[0059] The left trajectory is AB, the continuous function of the left trajectory at variable t is y(t), and the spatial position of the left trajectory is represented by f(y(t)).

[0060] The first preset trajectory is A'B', the continuous function of the first preset trajectory in variable t is y′(t), and the spatial position of the first preset trajectory is represented by f(y′(t)).

[0061] The similarity between the left trajectory and the first preset trajectory is calculated using the Frescher distance algorithm to obtain the first similarity value, which is given by the following formula:

[0062]

[0063] Where M is the length of the left trajectory, M′ is the length of the first preset trajectory, and d((f(y(t)),f′(y′(t))) is the distance between f(y(t)) and f′(y′(t)).

[0064] The target trajectory includes the right trajectory captured by the right camera, and the preset trajectory includes a second preset trajectory; the similarity between the target trajectory and the preset trajectory is calculated, and the similarity value is obtained, including:

[0065] The right trajectory is CD, the continuous function of the right trajectory in variable t is g(t), and the spatial position of the right trajectory is represented by f(g(t)).

[0066] The second preset trajectory is A'B'=N, the continuous function of the second preset trajectory in variable t is g′(t), and the spatial position of the second preset trajectory is represented by f(g′(t)).

[0067] The similarity between the right trajectory and the second preset trajectory is calculated using the Frescher distance algorithm to obtain the second similarity value, which is given by the following formula:

[0068]

[0069] Where N is the length of the right trajectory, N′ is the length of the second preset trajectory, and d((f(g(t)),f′(g′(t))) is the distance between f(g(t)) and f′(g′(t)).

[0070] In this embodiment, as Figure 5 As shown, the tracked person's trajectory appears within the field of view of the left and right cameras. The left trajectory and the right trajectory are AB and CD, respectively, with corresponding speeds of Vab and Vcd. The times when the person appears at points A, B, C, and D are Ta, Tb, Tc, and Td, respectively.

[0071] Trajectories A′B′ and C′D′ are the first and second preset trajectories, respectively. These trajectories can be calibrated based on different vehicle models or are empirical trajectories. The corresponding speeds are Va′b′ and Vc′d′, respectively, and the times at points A′, B′, C′, and D′ are Ta′, Tb′, Tc′, and Td′, respectively.

[0072] In this embodiment, taking the left-side trajectory of the vehicle as an example, the lengths of the left trajectories AB and A′B′ are M and N, respectively. The position of the person's movement on both trajectories can be described by a continuous function of the same variable t, where y(t) and y′(t) represent the position description functions of the person's movement on the tracking trajectory of the electronic rearview mirror and the preset trajectory, respectively, where t∈(0,t) max For the tracking trajectory, y(0) = 0, y(t) max ) = M, for the first preset trajectory, y′(0) = 0, y′(t max )=M′.

[0073] With the center of the rear axle of the current vehicle as the origin of the coordinate system, the horizontal axis is the X-axis, and the vertical axis is the Y-axis, as follows: Figure 5 As shown in the figure. In this coordinate system, f(x) and f′(x) represent the tracking trajectory and the preset trajectory, respectively.

[0074] Then at time t, f(y(t)) represents the character based on Figure 5 The spatial position of the coordinate system on the tracking trajectory, f′(y′(t)) represents the character's position based on the coordinate system. Figure 5 The spatial position of the coordinate system on the preset trajectory, d((f(y(t)),f′(y′(t))) represents the coordinate system based on the spatial position of the preset trajectory. Figure 5 The distance between coordinate systems f(y(t)) and f′(y′(t)).

[0075] The Frescher distance is used to characterize the curve similarity between the tracking trajectory AB and the preset trajectory A′B′:

[0076]

[0077] When F(AB, A′B′) < K, AB is considered similar to A′B′; otherwise, they are not similar. The first preset similarity value K here can be a length determined based on different vehicle models, or it can be an empirical length.

[0078] Similarly, the process for determining whether the trajectory on the right side of the current vehicle is similar is the same as described above. When F(CD,C′D′)<K′, the electronic rearview mirror determines whether CD and C′D′ are similar.

[0079] In some embodiments, determining whether a target pedestrian exists in the monitored area based on the target trajectory further includes: calculating the average velocity V of the left trajectory AB and the right trajectory CD, wherein... V ab V represents the average data of the left trajectory. cd Let L be the average velocity of the right trajectory; let L be the length of one endpoint B of the left trajectory and one endpoint D of the right trajectory, and determine... Whether it is true or not, where T b Let T be the time it takes for the pedestrian to arrive at B. d Let Δt be the time when the pedestrian arrives at D, and Δt be a preset time. In this embodiment, the application further determines V using the following method: V represents the average speed of the tracked person on the left trajectory AB and the right trajectory CD, i.e. Let L represent based on Figure 5 The length from point B to point D in the coordinate system. If This indicates that the tracked person's trajectory AB and CD are connected, meaning the tracked person continuously passes through A, B, D, C or C, D, B, A. Here, Δt can be the time length determined based on different vehicle models, or it can be an empirical time length.

[0080] In some embodiments, determining whether a target pedestrian exists in the monitored area based on the target trajectory includes: determining the presence of a target pedestrian in the monitored area when the first similarity value is less than a first preset similarity value, the second similarity value is less than a second preset similarity value, or any one or more of formula (4). In this embodiment, the presence of a target pedestrian is determined when the following three conditions are met: ① F(AB, A′B′) < K; ② F(CD, C′D′) < K′; ③ |T b -T d The electronic rearview mirror determines that the tracking trajectory AB is connected to CD, and the tracking trajectory ABCD is similar to the preset trajectory A′B′C′D′.

[0081] In some embodiments, this application also discloses a method for moving a vehicle, the method comprising: determining whether a target pedestrian exists in the monitored area using the aforementioned vehicle-surrounding target pedestrian detection method; if a target pedestrian exists, sending a notification to the target terminal, and simultaneously sending the field-of-view images from the left and right cameras to the target terminal, and having the target terminal confirm whether to connect the vehicle's communication module for a call. In this embodiment, this application uses the vehicle-surrounding target pedestrian detection method to determine whether the vehicle owner should connect the vehicle's electronic rearview mirror, and simultaneously sending the field-of-view image from the electronic rearview mirror. If the vehicle owner connects the vehicle's electronic rearview mirror, they can communicate with the target pedestrian through a loudspeaker and microphone to determine whether the vehicle needs to be moved.

[0082] In some embodiments, this application also discloses a device, which is an electronic rearview mirror installed on both sides of a vehicle, or a communication camera installed on the left and right rearview mirrors; the device is provided with a main control module, a communication module, a camera, a loudspeaker, and a microphone interconnected with each other; the main control module operates the above-described vehicle relocation method. In this embodiment,

[0083] In this embodiment, the device can be an electronic rearview mirror, including a left electronic rearview mirror and a right electronic rearview mirror. The right electronic rearview mirror 110 and the left electronic rearview mirror 140 are respectively arranged on the left and right sides of the vehicle 100, replacing the traditional side rearview mirrors. The right electronic rearview mirror 110 has a right camera 111, and the field of view of the right camera 111 is shown as the field of view of the right camera 112. The left electronic rearview mirror 140 has a left camera 141, and the field of view of the left camera 141 is shown as the field of view of the left camera 142. Here, the field of view of the left and right cameras is only for illustration, and the actual field of view is subject to the actual vehicle. The left display screen 130 and the right display screen 120 are arranged inside the vehicle. The left display screen 130 is connected to the left electronic rearview mirror 140 through a wire, and the right display screen 120 is connected to the right electronic rearview mirror 110 through a wire.

[0084] The left electronic rearview mirror 140 acquires the image within the field of view of the left camera 142 via the left camera 141, and transmits the image to the left display screen 130 through electrical circuits. The left display screen 130 then displays the image within the field of view of the left camera 142, thus achieving the same effect as a traditional side rearview mirror. The principle on the right side is the same as that on the left side, and will not be repeated here.

[0085] Figure 2 This demonstrates the traditional side rearview mirror function of the patented solution. When the vehicle is parked and the owner is away from the vehicle, this patented solution implements its monitoring function, such as... Figure 3 As shown.

[0086] Figure 3The left electronic rearview mirror 140 rotates around its rotation axis, so that the left camera 141 can monitor the space to the left of the current vehicle 1, and the right electronic rearview mirror 110 rotates around its rotation axis, so that the right camera 111 can monitor the space to the right of the current vehicle 1.

[0087] like Figure 4 As shown, the left electronic rearview mirror 140 not only has a left camera 141, but also a left loudspeaker 144, a left microphone 145, and a housing for the left electronic rearview mirror 143. The left camera 141, the left loudspeaker 144, and the left microphone 145 are all mounted on the housing for the left electronic rearview mirror 143. The structure of the right rearview mirror is symmetrical to that of the left rearview mirror.

[0088] When the vehicle is Figure 3 The monitoring status shown indicates that the electronic rearview mirror monitors the trajectory of pedestrians around the vehicle. When the electronic rearview mirror determines that a pedestrian is a driver whose car is blocked, it communicates with the driver's mobile phone via the TBOX module, transmitting the view from the electronic rearview mirror to the driver's mobile phone. The driver can then choose whether to activate the electronic rearview mirror. If the driver chooses to activate the electronic rearview mirror, they can communicate with the driver whose car is blocked through the loudspeaker and microphone on the electronic rearview mirror, and can also view the image within the electronic rearview mirror's field of view in real time. The communication module can be a TBOX module. This device monitors the trajectory of pedestrians around the vehicle. When this device determines that a pedestrian is a driver whose car is blocked, it communicates with the driver's mobile phone via the TBOX module, transmitting the view from the device to the driver's mobile phone. The driver can then choose whether to activate the electronic rearview mirror. If the driver chooses to activate the electronic rearview mirror, they can communicate with the driver whose car is blocked through the loudspeaker and microphone on the device, and can also view the image within the device's field of view in real time. In this embodiment, the notification method to the car owner is actively determined by the electronic rearview mirror. The first layer of communication initiative rests with the electronic rearview mirror, preventing the car owner from being harassed. Simultaneously, after the electronic rearview mirror's determination, the car owner can further determine whether other vehicles are blocked based on the image transmitted by the mirror. The second layer of communication initiative rests with the car owner, providing double protection and enhancing customer satisfaction. Structurally, no hardware modules are added; existing electronic rearview mirrors and TBOX modules from vehicles are used. Functionality is implemented solely at the software algorithm level, ensuring high replicability. The communication method is actively determined by the vehicle's electronic rearview mirror, unlike other methods that require active operation by other car owners, preventing damage to the vehicle. The communication method utilizes the existing common communication method between TBOX and the car owner's mobile phone, reducing the car owner's usage costs.

[0089] Obviously, the above embodiments of this application are merely examples for clear illustration and are not intended to limit the implementation of this application. Those skilled in the art can make other variations or modifications based on the above description. It is neither necessary nor possible to exhaustively describe all possible implementations here. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of the claims of this application.

Claims

1. A vehicle surrounding target pedestrian detection method characterized by, When applied to automotive electronic products, the method includes: Real-time monitoring of the current vehicle's monitoring area and acquisition of the target trajectory; Based on the target trajectory, determine whether there is a target pedestrian in the monitored area; If the determination result is true, then there is a target pedestrian in the monitored area; Otherwise, no action will be taken; The step of determining whether a target pedestrian exists in the monitored area based on the target trajectory includes: calculating the similarity between the target trajectory and a preset trajectory, and obtaining a similarity value; if the similarity value is less than the preset similarity value, it is determined that a target pedestrian exists in the current vehicle detection area; otherwise, no processing is performed. The target trajectory includes the left trajectory captured by the left camera, and the preset trajectory includes a first preset trajectory; the step of calculating the similarity between the target trajectory and the preset trajectory and obtaining a similarity value includes: The left trajectory is AB, and the continuous function of the left trajectory with respect to variable t is: The spatial position of the left trajectory is represented as ; The first preset trajectory is A'B', and the continuity function of the first preset trajectory with respect to variable t is: The spatial position of the first preset trajectory is represented as ; The similarity between the left trajectory and the first preset trajectory is calculated using the Frescher distance algorithm to obtain the first similarity value.

2. The method for detecting pedestrians around a vehicle according to claim 1, characterized in that, The detection area of ​​the current vehicle is monitored in real time, and the target trajectory is obtained; The monitoring area is monitored by the left and / or right cameras installed on the current vehicle. The image information of the monitoring area is acquired in real time, and the image signal is identified to determine whether the image signal contains the characteristics of the target pedestrian. If the image signal contains features of the target pedestrian, the target trajectory is obtained based on the continuous image signal; Otherwise, no action will be taken.

3. The method for detecting pedestrians around a vehicle according to claim 1, characterized in that, The similarity between the left trajectory and the first preset trajectory is calculated using the Fraser distance algorithm to obtain the first similarity value, which is expressed by the following formula: Where M is the length of the left trajectory. The first preset trajectory length, for and The distance between them, where t is the value of the continuous function. max This represents the value of the variable corresponding to the endpoint of a continuous function.

4. The method for detecting pedestrians around a vehicle according to claim 3, characterized in that, The target trajectory includes the right trajectory captured by the right camera, and the preset trajectory includes a second preset trajectory; the step of calculating the similarity between the target trajectory and the preset trajectory and obtaining a similarity value includes: The right trajectory is CD, and the continuous function of the right trajectory with respect to variable t is... The spatial position of the right trajectory is represented as ; The second preset trajectory is =N, the second preset trajectory is a continuous function of variable t. The spatial position of the second preset trajectory is represented as ; The similarity between the right trajectory and the second preset trajectory is calculated using the Frescher distance algorithm to obtain the second similarity value, which is given by the following formula: Where N is the length of the right trajectory. The second preset trajectory length, for and The distance between them.

5. The method for detecting pedestrians around a vehicle according to claim 4, characterized in that, The step of determining whether a target pedestrian exists in the monitored area based on the target trajectory also includes: Calculate the average velocity V of the left trajectory AB and the right trajectory CD, where, …(3), The average data for the left trajectory. The average velocity of the right trajectory; Let L be the length from endpoint B of the left trajectory to endpoint D of the right trajectory. Determine... …(4) Whether it is true, where, Let B be the time it takes for the pedestrian to arrive at point B. The time when the pedestrian arrives at D, This is the preset time.

6. The method for detecting pedestrians around a vehicle according to claim 5, characterized in that, The step of determining whether a target pedestrian exists in the monitored area based on the target trajectory includes: If the first similarity value is less than the first preset similarity value, the second similarity value is less than the second preset similarity value, or any one or more of the formula (4), then it is determined that there is a target pedestrian in the monitoring area.

7. A method for moving a car, characterized in that, The method includes: The method for detecting pedestrians around a vehicle as described in any one of claims 1-6 is used to determine whether a target pedestrian exists in the monitored area. If a target pedestrian is present, a notification is sent to the target terminal, along with the field-of-view images from the left and right cameras. The target terminal then confirms whether to connect to the vehicle's communication module for a call.

8. An apparatus, characterized in that, The device is an electronic rearview mirror installed on both sides of the vehicle, or a communication camera installed on the left and right rearview mirrors; the device is equipped with a main control module, a communication module, a camera, a loudspeaker, and a microphone that are interconnected; the main control module operates the vehicle relocation method as described in claim 7.

Images