Light curtain control method and apparatus, vehicle and storage medium

Abstract

A light curtain control method comprises: acquiring an environmental image of a vehicle (S10); when there is a human body image region in the environmental image, determining a first lateral relative position of the human body image region in the environmental image (S20); on the basis of the first lateral relative position, determining a second lateral relative position on a light curtain of the vehicle (S30); and displaying a light curtain pattern on the basis of the second lateral relative position (S40). By means of the light curtain control method, the light curtain can reflect the real dynamics of a person in real time, thereby effectively improving the interactivity between a person outside a vehicle and the vehicle. Further provided are a light curtain control apparatus, a vehicle and a computer-readable storage medium.

Description

Light curtain control method, device, vehicle and storage medium

[0001] This application claims priority to Chinese patent application No. 202311740750.8 filed on December 18, 2023, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The embodiments of the present application relate to, but are not limited to, the field of vehicle light curtain control technology, and in particular to a light curtain control method, device, vehicle, and storage medium. Background Art

[0003] With the continuous advancement of automotive technology, headlight design is becoming increasingly personalized and intelligent. Some vehicles can even display text or images on their light curtains when they recognize a person in front of them. However, in these scenarios, the position of the displayed content on the light curtains is often fixed. Technical issues

[0004] The following is a summary of the subject matter described in detail herein. This summary is not intended to limit the scope of the claims.

[0005] The main purpose of this application is to provide a light curtain control method, device, vehicle and storage medium, aiming to solve the problem that the light curtain display content cannot reflect the real dynamics of people. Technical Solutions

[0006] To achieve the above objectives, the present application provides a light curtain control method, the light curtain control method comprising:

[0007] Acquire an image of the vehicle's environment;

[0008] In a case where a human body image region exists in the environment image, determining a first horizontal relative position of the human body image region in the environment image;

[0009] determining a second lateral relative position on the light curtain of the vehicle based on the first lateral relative position;

[0010] A light curtain pattern is displayed based on the second transverse relative position.

[0011] In one embodiment, after the step of acquiring the environmental image of the vehicle and before the step of determining the first lateral relative position of the human body image area in the environmental image, the method further includes:

[0012] Convert the environment image in color-coded YUV format to an environment image in GPU texture format;

[0013] Inputting the environment image in GPU texture format into a pre-trained human body recognition model to obtain a human body recognition result output by the human body recognition model, wherein the human body recognition result includes human body image area position data;

[0014] The step of determining a first lateral relative position of the human body image area in the environment image comprises:

[0015] The first horizontal relative position is calculated according to the human body image area position data and the pixel width of the environment image in the GPU texture format.

[0016] In one embodiment, the human body image area position data includes the horizontal coordinate of the center pixel of the human body image area, and the step of calculating the first horizontal relative position according to the human body image area position data and the pixel width of the environment image in the GPU texture format includes:

[0017] The first horizontal relative position is obtained by dividing the horizontal coordinate of the central pixel of the human body image area by the pixel width of the environment image in the GPU texture format.

[0018] In one embodiment, the step of displaying a light curtain pattern based on the second horizontal relative position includes:

[0019] Acquire the light curtain pattern, wherein a pixel width of the light curtain pattern is smaller than a pixel width of the light curtain, and a pixel height of the light curtain pattern is equal to a pixel height of the light curtain;

[0020] Based on the second horizontal relative position, pixels are supplemented outside the light curtain pattern to obtain a target pattern, wherein at the second horizontal relative position of the light curtain pattern in the target pattern, the pixel width and pixel height of the target pattern are respectively consistent with the pixel width and pixel height of the light curtain;

[0021] The light curtain is controlled to display the target pattern.

[0022] In one embodiment, the human body recognition result further includes human body posture data, and the step of obtaining the light curtain pattern includes:

[0023] According to the human body posture data, a corresponding light curtain pattern reflecting the human body posture is obtained.

[0024] In one embodiment, the step of obtaining the light curtain pattern includes:

[0025] According to the movement state of the human body, a corresponding light curtain pattern reflecting the movement state of the human body is obtained, wherein the movement state of the human body is predetermined based on the first lateral relative position.

[0026] In one embodiment, the human body recognition result further includes human body posture data, and the step of obtaining the light curtain pattern includes:

[0027] According to the human body posture data and the human body movement state, a corresponding light curtain pattern reflecting the human body posture and the human body movement state is obtained, wherein the human body movement state is predetermined based on the first lateral relative position.

[0028] In one embodiment, the number of the environment images in the GPU texture format is at least two frames, each frame of the environment image in the GPU texture format corresponds to a first horizontal relative position, and the horizontal coordinates of the pixels of the environment image in the GPU texture format increase from the first side to the second side. The process of determining the movement state of the human body includes:

[0029] storing at least two first lateral relative positions in a queue of a preset length;

[0030] performing a weighted average calculation on the at least two first horizontal relative positions in the queue to obtain a weighted average, wherein the weights corresponding to the at least two first horizontal relative positions in the queue are set in sequence according to the time sequence in which they were stored in the queue, with the first horizontal relative position stored most recently in the queue having the highest weight;

[0031] If the absolute difference between the first lateral relative position stored in the queue the latest and the weighted average value is less than a preset threshold, determining that the movement state of the human body is stationary;

[0032] If the absolute difference between the first lateral relative position stored latest in the queue and the weighted average value is greater than or equal to a preset threshold, and the first lateral relative position stored latest in the queue is greater than the weighted average value, determining that the movement state of the human body is moving toward the second side;

[0033] If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is less than the weighted average value, it is determined that the human body movement state is moving toward the first side.

[0034] The present application also provides a light curtain control device, which includes:

[0035] An acquisition module, used to acquire an environmental image of the vehicle;

[0036] a first determining module, configured to determine, when a human body image region exists in the environmental image, a first horizontal relative position of the human body image region in the environmental image;

[0037] a second determining module, configured to determine a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position;

[0038] A display module is configured to display a light curtain pattern based on the second horizontal relative position.

[0039] An embodiment of the present application also proposes a vehicle, which includes a memory, a processor, and a light curtain control program stored in the memory and executable on the processor. When the light curtain control program is executed by the processor, the steps of the light curtain control method described above are implemented.

[0040] An embodiment of the present application further provides a computer-readable storage medium, on which a light curtain control program is stored. When the light curtain control program is executed by a processor, the steps of the light curtain control method described above are implemented. Beneficial effects

[0041] The light curtain control method, device, vehicle and storage medium proposed in the embodiments of the present application obtain an environmental image of the vehicle; if a human image area exists in the environmental image, determine the first lateral relative position of the human image area in the environmental image; determine the second lateral relative position on the light curtain of the vehicle based on the first lateral relative position; and display the light curtain pattern based on the second lateral relative position. Based on the scheme of the present application, the environmental image of the vehicle is first obtained, and the first lateral relative position of the human image area in the environmental image is determined. Subsequently, the second lateral relative position is determined on the light curtain of the vehicle based on the first relative position, thereby achieving dynamic tracking of the person's position. The light curtain pattern is displayed on the light curtain based on the second lateral relative position, so that the light curtain can reflect the real dynamics of the person in real time, effectively improving the interactivity between people outside the vehicle and the vehicle.

[0042] Still other aspects will become apparent upon reading and understanding the accompanying drawings and detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] FIG1 is a schematic diagram of the functional modules of a vehicle to which the light curtain control device of the present application belongs;

[0044] FIG2 is a flow chart of a first exemplary embodiment of a light curtain control method of the present application;

[0045] FIG3 is a schematic diagram of a light curtain display effect involved in the light curtain control method of the present application;

[0046] FIG4 is a flow chart of a second exemplary embodiment of a light curtain control method of the present application;

[0047] FIG5 is a flowchart of a third exemplary embodiment of a light curtain control method of the present application;

[0048] FIG6 is a flowchart of a fourth exemplary embodiment of a light curtain control method of the present application;

[0049] FIG7 is a schematic diagram of a pattern involved in the light curtain control method of the present application;

[0050] FIG8 is a flowchart of a fifth exemplary embodiment of a light curtain control method of the present application;

[0051] FIG9 is a flow chart of a sixth exemplary embodiment of a light curtain control method of the present application;

[0052] FIG10 is a flow chart of a seventh exemplary embodiment of a light curtain control method of the present application;

[0053] FIG11 is a flow chart of an eighth exemplary embodiment of the light curtain control method of the present application.

[0054] The realization of the objectives, functional features and advantages of this application will be further explained in conjunction with embodiments and with reference to the accompanying drawings. Modes for Carrying Out the Invention

[0055] It should be understood that the specific embodiments described herein are only used to explain the present application and are not intended to limit the present application.

[0056] The main solution of the embodiment of the present application is: acquiring an environmental image of the vehicle; in the case where a human image area exists in the environmental image, determining a first lateral relative position of the human image area in the environmental image; determining a second lateral relative position on the light curtain of the vehicle based on the first lateral relative position; and displaying a light curtain pattern based on the second lateral relative position. Based on the solution of the present application, the environmental image of the vehicle is first acquired, and the first lateral relative position of the human image area in the environmental image is determined. Subsequently, the second lateral relative position is determined on the light curtain of the vehicle based on the first relative position, thereby achieving dynamic tracking of the person's position. The light curtain pattern is displayed on the light curtain based on the second lateral relative position, so that the light curtain can reflect the real dynamics of the person in real time, effectively improving the interactivity between people outside the vehicle and the vehicle.

[0057] Specifically, referring to Figure 1, which is a schematic diagram of the functional modules of the vehicle to which the light curtain control device of the present application belongs, the light curtain control device can be a device independent of the vehicle that can control the light curtain and can be carried on the vehicle in the form of hardware or software.

[0058] In this embodiment, the vehicle to which the light curtain control device belongs includes at least an output module 110 , a processor 120 , a memory 130 and a communication module 140 .

[0059] The memory 130 stores an operating system and a light curtain control program. The light curtain control device can store information such as the vehicle's environmental image, the first lateral relative position of the human image area within the environmental image, the second lateral relative position on the light curtain, and the light curtain pattern in the memory 130. The output module 110 can be a display screen, etc. The communication module 140 can include a Wi-Fi module, a mobile communication module, and a Bluetooth module, etc., and communicates with external devices or servers through the communication module 140.

[0060] When the light curtain control program in the memory 130 is executed by the processor, the following steps are implemented:

[0061] Acquire an image of the vehicle's environment;

[0062] In a case where a human body image region exists in the environment image, determining a first horizontal relative position of the human body image region in the environment image;

[0063] determining a second lateral relative position on the light curtain of the vehicle based on the first lateral relative position;

[0064] A light curtain pattern is displayed based on the second transverse relative position.

[0065] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0066] Convert the environment image in color-coded YUV format to an environment image in GPU texture format;

[0067] Inputting the environment image in GPU texture format into a pre-trained human body recognition model to obtain a human body recognition result output by the human body recognition model, wherein the human body recognition result includes human body image area position data;

[0068] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0069] The first horizontal relative position is calculated according to the human body image area position data and the pixel width of the environment image in the GPU texture format.

[0070] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0071] The first horizontal relative position is obtained by dividing the horizontal coordinate of the central pixel of the human body image area by the pixel width of the environment image in the GPU texture format.

[0072] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0073] Acquire the light curtain pattern, wherein a pixel width of the light curtain pattern is smaller than a pixel width of the light curtain, and a pixel height of the light curtain pattern is equal to a pixel height of the light curtain;

[0074] Based on the second horizontal relative position, pixels are supplemented outside the light curtain pattern to obtain a target pattern, wherein at the second horizontal relative position of the light curtain pattern in the target pattern, the pixel width and pixel height of the target pattern are respectively consistent with the pixel width and pixel height of the light curtain;

[0075] The light curtain is controlled to display the target pattern.

[0076] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0077] According to the human body posture data, a corresponding light curtain pattern reflecting the human body posture is obtained.

[0078] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0079] According to the movement state of the human body, a corresponding light curtain pattern reflecting the movement state of the human body is obtained, wherein the movement state of the human body is predetermined based on the first lateral relative position.

[0080] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0081] According to the human body posture data and the human body movement state, a corresponding light curtain pattern reflecting the human body posture and the human body movement state is obtained, wherein the human body movement state is predetermined based on the first lateral relative position.

[0082] Furthermore, when the light curtain control program in the memory 130 is executed by the processor, the following steps are also implemented:

[0083] storing at least two first lateral relative positions in a queue of a preset length;

[0084] performing a weighted average calculation on the at least two first horizontal relative positions in the queue to obtain a weighted average, wherein the weights corresponding to the at least two first horizontal relative positions in the queue are set in sequence according to the time sequence in which they were stored in the queue, with the first horizontal relative position stored most recently in the queue having the highest weight;

[0085] If the absolute difference between the first lateral relative position stored in the queue the latest and the weighted average value is less than a preset threshold, determining that the movement state of the human body is stationary;

[0086] If the absolute difference between the first lateral relative position stored latest in the queue and the weighted average value is greater than or equal to a preset threshold, and the first lateral relative position stored latest in the queue is greater than the weighted average value, determining that the movement state of the human body is moving toward the second side;

[0087] If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is less than the weighted average value, it is determined that the human body movement state is moving toward the first side.

[0088] In this embodiment, an image of the vehicle's surroundings is first acquired, and a first lateral relative position of the human image region within the surrounding image is determined. Subsequently, a second lateral relative position is determined on the vehicle's light curtain based on the first relative position, thereby dynamically tracking the person's position. A light curtain pattern is displayed on the light curtain based on the second lateral relative position, enabling the light curtain to reflect the person's real-life movements in real time, effectively enhancing interaction between the person outside the vehicle and the vehicle.

[0089] 2 , a first embodiment of a light curtain control method of the present application provides a flow chart, wherein the light curtain control method includes:

[0090] Step S10: Acquire an environmental image of the vehicle.

[0091] Specifically, with the continuous advancement of automotive technology, headlight design is becoming increasingly personalized and intelligent. Some vehicles can even display text or images on their light curtains when they recognize a person in front of them. However, in these scenarios, the position of the displayed content on the light curtains is often fixed, failing to reflect the person's actual movements.

[0092] This embodiment proposes an image-based light curtain control method, enabling the light curtain display to reflect a person's real-life movements. More specifically, this embodiment involves a vehicle equipped with a light curtain featuring multiple independently controllable LEDs. By adjusting the color, brightness, and flashing pattern of the LEDs, the light curtain can present different lighting expressions to convey information, highlight the vehicle's personality, or enhance its visual appeal.

[0093] First, the vehicle's surroundings must be captured using an onboard camera. Onboard cameras are one of the most common hardware devices, typically mounted on the front, rear, side, or other locations. They capture images of the vehicle's surroundings, providing visual information such as front and rear views, panoramic views, and more.

[0094] Step S20 : ​​When a human body image region exists in the environment image, determining a first horizontal relative position of the human body image region in the environment image.

[0095] Specifically, after acquiring an environmental image, the environmental image can be processed using a computer vision algorithm to detect whether a human image region exists in the environmental image. A human image region refers to a specific area occupied by a human body detected in the environmental image. The human image region may be detected and identified using an image processing algorithm and is typically represented as a specific portion or outline of a human body within the image, or as a detection frame containing a human body.

[0096] If a human image region exists in the environmental image, the first horizontal relative position of the human image region in the environmental image can be further determined. More specifically, the horizontal coordinates of the pixels in the human image region and the pixel width of the environmental image can be obtained. Then, based on the horizontal coordinates and the pixel width of the human image region, the first horizontal relative position of the human image region in the environmental image can be calculated. The first horizontal relative position refers to the relative position of the human image region in the horizontal direction of the entire environmental image and can be expressed as a percentage or decimal. For example, the first horizontal relative position of the human image region in the environmental image is 0.65, which means that a point in the human image region is 65% of the horizontal position of the entire environmental image. This representation typically normalizes the width of the environmental image to a range of 0 to 1, facilitating uniform processing across environmental images of varying resolutions. If the horizontal coordinates of the pixels in the environmental image increase from a first side (e.g., the left side) to a second side (e.g., the right side), then a first horizontal relative position of 0 indicates that a point in the human image region is at the first edge (e.g., the leftmost side) of the environmental image, and a first horizontal relative position of 1 indicates that a point in the human image region is at the second edge (e.g., the rightmost side) of the environmental image.

[0097] Step S30: determining a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position.

[0098] Specifically, the second horizontal relative position refers to the position where the light curtain pattern reflecting human motion should be displayed, and is also the relative position of the light curtain pattern in the horizontal direction of the entire light curtain. To determine the second horizontal relative position on the vehicle's light curtain, the installation position of the vehicle camera must be considered first. There are two specific situations:

[0099] Case 1: The visual center of the on-board camera and the center of the light curtain are both located on the central axis of the vehicle. The field of view width of the on-board camera can be scaled proportionally to the pixel width of the light curtain, and the relative position of people or objects in the field of view remains unchanged. At this time, the first lateral relative position can be directly used as the second lateral relative position. Take Figure 3 as an example. Figure 3 is a schematic diagram of the light curtain display effect involved in the light curtain control method of this application. The on-board camera that obtains the environmental image is the driving recorder shown in Figure 3. The visual center of the driving recorder and the center of the light curtain are both located on the central axis of the vehicle. That is to say, based on the environmental image obtained by the driving recorder, the first lateral relative position can be calculated, and the first lateral relative position can be used as the second lateral relative position.

[0100] Case 2: The on-board camera's visual center is not located on the vehicle's central axis. In other words, the on-board camera may be installed on the left or right side of the vehicle. In this case, the camera's field of view cannot be directly scaled to the pixel width of the light curtain, and the relative position of people or objects in the field of view will also change. Therefore, it is necessary to convert the first lateral relative position into the second lateral relative position based on pre-calibrated parameters.

[0101] In this embodiment, the vehicle-mounted camera installation position described in the above-mentioned situation 1 may be selected, and the first horizontal relative position and the second horizontal relative position do not need to be converted.

[0102] Step S40: displaying a light curtain pattern based on the second horizontal relative position.

[0103] Specifically, after determining the second horizontal relative position, a light curtain pattern can be further obtained and displayed at the second horizontal relative position. The light curtain pattern not only reflects the position of the human body, but also reflects the posture and / or movement state (direction) of the human body.

[0104] In one possible implementation, a vehicle-mounted camera is located at the rear of the vehicle, capturing images from behind, while a light curtain is located at the front of the vehicle. When someone moves near the rear of the vehicle, their movement is reflected in real time on the light curtain at the front of the vehicle. This allows users to monitor the movement of people at the rear of the vehicle through the display on the light curtain at the front of the vehicle.

[0105] In this embodiment, an image of the vehicle's surroundings is first acquired, and a first lateral relative position of the human image region within the surrounding image is determined. Subsequently, a second lateral relative position is determined on the vehicle's light curtain based on the first relative position, thereby dynamically tracking the person's position. A light curtain pattern is displayed on the light curtain based on the second lateral relative position, enabling the light curtain to reflect the person's real-life movements in real time, effectively enhancing interaction between the person outside the vehicle and the vehicle.

[0106] Further, referring to FIG4 , a second embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG2 , after obtaining the vehicle's environmental image in step S10 and before "determining the first lateral relative position of the human body image area in the environmental image" in step S20, the method further includes:

[0107] Step S011, converting the environment image in color-coded YUV format into an environment image in a graphics processor GPU texture format;

[0108] Step S012, inputting the environment image in the GPU texture format into a pre-trained human body recognition model to obtain a human body recognition result output by the human body recognition model, wherein the human body recognition result includes human body image area position data;

[0109] Further refining the step S20 of “determining a first horizontal relative position of the human body image region in the environment image” includes:

[0110] Step S201 : Calculate the first horizontal relative position according to the human body image area position data and the pixel width of the environment image in the GPU texture format.

[0111] Specifically, the environmental image recorded in real time by the on-board camera is an environmental image in YUV format. The YUV format is an image color encoding format that divides the image into two components: brightness (Y) and chrominance (UV) to effectively compress color information, thereby reducing the size of image data. It is widely used in image compression and transmission.

[0112] YUV-formatted environmental images cannot be used directly for human body recognition. They must be converted to GPU texture format. This conversion can be implemented using a pre-defined algorithm framework, such as the Camera2 framework in Android. A GPU (graphics processing unit) texture is a data structure used in graphics programming to store and process image data.

[0113] By inputting an environment image in GPU texture format into a pre-trained human body recognition model, a human body recognition result output by the human body recognition model can be obtained, wherein the human body recognition result includes human body image area position data. In one embodiment, the human body recognition result may also include human body posture data. The reason why the human body recognition model can output human body image area position data and human body posture data is that it learns the correspondence between input data and expected output during the training process. For example, when performing human body posture recognition, the human body recognition model will learn how to identify the key points of the human skeleton from the image, understand and determine the corresponding human body posture.

[0114] The human image region position data may include the pixel coordinates of a point in the human image region. The point may be the center or a non-center point of the human image region. The coordinates of the point are used to represent the position of the human image region. The first horizontal relative position is obtained by dividing the horizontal pixel coordinate of the point by the pixel width of the environment image in GPU texture format.

[0115] In this embodiment, by converting an environmental image in YUV format into an environmental image in GPU texture format and combining it with a pre-trained human body recognition model, the image can be more efficiently processed and analyzed, thereby obtaining a human body recognition result output by the human body recognition model. The human body recognition result includes human body image region position data. Based on the human body image region position data and the pixel width of the environmental image in GPU texture format, the first horizontal relative position can be calculated. This makes the calculated first horizontal relative position more accurate.

[0116] Further, with reference to FIG5 , a third embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG4 , the human image area position data includes the horizontal coordinate of the center pixel of the human image area. The “calculating the first horizontal relative position according to the human image area position data and the pixel width of the environment image in the GPU texture format” in step S201 is further refined, including:

[0117] Step S2011: Divide the horizontal coordinate of the center pixel of the human body image area by the pixel width of the environment image in the GPU texture format to obtain the first horizontal relative position.

[0118] Specifically, in this embodiment, the center point of the human body image area can be selected to represent the position of the human body image area. The human body image area position data includes the horizontal coordinate of the center pixel of the human body image area. The first horizontal relative position can be obtained by dividing the horizontal coordinate of the center pixel of the human body image area by the pixel width of the environment image in the GPU texture format.

[0119] For example, the horizontal coordinate of the center pixel of the human body image area is 1248, and the pixel width of the environment image in GPU texture format (1080P resolution) is 1920, then the first horizontal relative position is 1248 / 1920=0.65.

[0120] In this embodiment, the center point of the human image region is used to represent the position of the human image region. The first horizontal relative position is calculated using the horizontal coordinate of the central pixel of the human image region and the pixel width of the environment image in GPU texture format. In this way, the first horizontal relative position more accurately reflects the actual position of the person.

[0121] Further, with reference to FIG6 , a fourth embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG4 , the “displaying the light curtain pattern based on the second horizontal relative position” in step S40 is further refined, including:

[0122] Step S401, acquiring the light curtain pattern, wherein the pixel width of the light curtain pattern is smaller than the pixel width of the light curtain, and the pixel height of the light curtain pattern is equal to the pixel height of the light curtain;

[0123] Step S402: Based on the second horizontal relative position, pixels are supplemented outside the light curtain pattern to obtain a target pattern, wherein at the second horizontal relative position of the light curtain pattern in the target pattern, the pixel width and pixel height of the target pattern are respectively consistent with the pixel width and pixel height of the light curtain;

[0124] Step S403: controlling the light curtain to display the target pattern.

[0125] Specifically, after determining the second horizontal relative position, a light curtain pattern needs to be obtained. The pixel height of the light curtain pattern is equal to the pixel height of the light curtain, and the pixel width of the light curtain pattern is less than the pixel width of the light curtain. In other words, the light curtain pattern only occupies a portion of the light curtain, and the portion outside the light curtain pattern is a blank portion that needs to be filled.

[0126] Furthermore, based on the second horizontal relative position, the horizontal coordinates of the pixels of the lamp curtain pattern in the lamp curtain are determined, and the blank parts outside the lamp curtain pattern are supplemented with pixels (for example, "0" is supplemented, and "0" indicates that the corresponding lamp beads are not lit), and finally the target pattern that completely occupies all pixels of the lamp curtain is obtained.

[0127] Taking Figure 7 as an example, Figure 7 is a schematic diagram of the pattern involved in the light curtain control method of this application. The arrow pattern with a size of 35*8 is the light curtain pattern, and the part outside the light curtain pattern is padded with "0" to obtain a target pattern with a size of 200*8.

[0128] In this embodiment, the light curtain pattern only occupies a portion of the light curtain. This means that compared to a full-size pattern, this embodiment has the advantages of smaller storage space and higher transmission efficiency. To display the light curtain pattern, one simply fills the pixels on top of the light curtain pattern to obtain the full-size target pattern, and then controls the light curtain to display the target pattern. Furthermore, the pattern display effect achieved by this embodiment is consistent with that achieved by a full-size pattern.

[0129] Further, with reference to FIG8 , a fifth embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG6 , the human body recognition result also includes human body posture data, and the “obtaining the light curtain pattern” in step S401 is further refined, including:

[0130] Step S4011: acquiring a corresponding light curtain pattern reflecting the human posture according to the human posture data.

[0131] Specifically, the human body recognition results in this embodiment also include human posture data, which reflects a specific human posture. Based on the human posture data, a corresponding light curtain pattern reflecting the human posture can be obtained. The light curtain pattern reflecting the human posture includes the corresponding human posture content, such as walking, standing still, waving, "wide-spreading", and leaning over.

[0132] In this embodiment, a corresponding light curtain pattern reflecting the human body posture is obtained according to the human body posture data, which can display the content of the human body posture and effectively enhance the display effect of the light curtain pattern.

[0133] Further, referring to FIG9 , a sixth embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG6 , the step S401 of “obtaining the light curtain pattern” is further refined, including:

[0134] Step S4012: acquiring a corresponding light curtain pattern reflecting the movement state of the human body according to the movement state of the human body, wherein the movement state of the human body is predetermined based on the first lateral relative position.

[0135] Specifically, since the first lateral relative position can reflect a person's location, if there are a sufficient number of first lateral relative positions, the person's movement state can be determined based on the first lateral relative position analysis. There are three types of movement states: movement to a first side (e.g., the left side), standing still, and movement to a second side (e.g., the right side).

[0136] After determining the movement state of the human body, a corresponding light curtain pattern reflecting the movement state of the human body may be obtained according to the movement state of the human body. The light curtain pattern reflecting the movement state of the human body includes the corresponding movement state content of the human body.

[0137] In this embodiment, a corresponding light curtain pattern reflecting the movement state of the human body is obtained according to the movement state of the human body, which can display the content of the movement state of the human body and effectively enhance the display effect of the light curtain pattern.

[0138] Further, with reference to FIG10 , a seventh embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG6 , the human body recognition result also includes human body posture data, and the “obtaining the light curtain pattern” in step S401 is further refined, including:

[0139] Step S4013: acquiring a corresponding light curtain pattern reflecting the human body posture and the human body movement state according to the human body posture data and the human body movement state.

[0140] Specifically, the human body recognition results in this embodiment also include human posture data, which reflects a specific human posture. Based on the human posture data and the human body's movement state, a corresponding light curtain pattern reflecting the human posture and movement state can be obtained. The light curtain pattern reflecting the human posture and movement state includes the corresponding human posture content and movement state content, such as walking left, walking right, etc.

[0141] In this embodiment, a corresponding light curtain pattern reflecting the human body posture and human body movement status is obtained according to the human body posture data and human body movement status, which can display the content of the human body posture and human body movement status, and effectively enhance the display effect of the light curtain pattern.

[0142] Further, with reference to FIG11 , an eighth embodiment of the light curtain control method of the present application provides a flow chart. Based on the embodiment shown in FIG9 or FIG10 , the number of the environment images in the GPU texture format is at least two frames, each frame of the environment image in the GPU texture format corresponds to a first horizontal relative position, and the horizontal coordinates of the pixels of the environment image in the GPU texture format increase from the first side to the second side. The process of determining the movement state of the human body includes:

[0143] Step S021, storing at least two first horizontal relative positions into a queue of a preset length;

[0144] Step S022: performing a weighted average calculation on the at least two first horizontal relative positions in the queue to obtain a weighted average value, wherein the weights corresponding to the at least two first horizontal relative positions in the queue are set in sequence according to the time sequence in which they were stored in the queue, with the first horizontal relative position stored in the queue the latest having the highest weight;

[0145] Step S023: if the absolute difference between the first horizontal relative position stored latest in the queue and the weighted average value is less than a preset threshold, determining that the movement state of the human body is stationary;

[0146] Step S024: if the absolute difference between the first lateral relative position stored latest in the queue and the weighted average value is greater than or equal to a preset threshold, and the first lateral relative position stored latest in the queue is greater than the weighted average value, then determining that the movement state of the human body is moving toward the second side;

[0147] Step S025: If the absolute difference between the first lateral relative position stored latest in the queue and the weighted average value is greater than or equal to a preset threshold, and the first lateral relative position stored latest in the queue is less than the weighted average value, determining that the movement state of the human body is moving toward the first side;

[0148] Specifically, to enhance the display effect of the light curtain pattern, this embodiment introduces human movement status as a basis for obtaining the light curtain pattern. First, at least two first horizontal relative positions are stored in a queue of preset length. Based on the chronological order of storage in the queue, weights are set for the at least two first horizontal relative positions in the queue. The first horizontal relative position stored in the queue the latest has the highest weight, and the first horizontal relative position stored in the queue the earliest has the lowest weight. Then, a weighted average is calculated for the at least two first horizontal relative positions in the queue to obtain a weighted average.

[0149] If the absolute difference between the first lateral relative position stored in the queue latest and the weighted average value is less than a preset threshold, the movement state of the human body is determined to be stationary; if the absolute difference between the first lateral relative position stored in the queue latest and the weighted average value is greater than or equal to the preset threshold, and the first lateral relative position stored in the queue latest is greater than the weighted average value, the movement state of the human body is determined to be moving to the second side; if the absolute difference between the first lateral relative position stored in the queue latest and the weighted average value is greater than or equal to the preset threshold, and the first lateral relative position stored in the queue latest is less than the weighted average value, the movement state of the human body is determined to be moving to the first side.

[0150] For example, the queue is preset to have a length of 16 and is initially empty. Assume there are 16 consecutive frames of first horizontal relative positions. These 16 frames are sequentially stored in the queue as [p1, p2, p3, ..., p16]. The first horizontal relative positions in this queue are considered the movement path of the human body, representing the position change of the human body over the past 16 frames.

[0151] Perform weighted average calculation on the first horizontal relative position in the queue. Assume that the latest frame is at the end of the queue and the earliest frame is at the head of the queue. You can define a weight array, such as linearly increasing weights [1, 2, 3, ..., 16]

[0152] Then calculate the weighted average wmean = (1*p1+2*p2+3*p3+…+16*p16) / (1+2+3+…+16). It can be understood that the weighted average takes into account the position information of the most recent 16 frames, where the most recent frame has the largest weight.

[0153] The preset threshold is t. If |p16-wmean|<t, the human body's movement state is determined to be stationary;

[0154] If |p16-wmean|≥t, and p16>wmean, then the movement state of the human body is determined to be moving toward the second side;

[0155] If |p16-wmean|≥t, and p16<wmean, it is determined that the movement state of the human body is moving toward the first side.

[0156] In this embodiment, by using a queue to store and weighted average multiple first lateral relative positions, the movement trend of the human body can be smoothly tracked. This method reduces the impact of fluctuations in single-frame data and improves the stability of the judgment of the human body's movement state. The time series recording of the queue helps to more accurately determine whether the person is stationary, moving left, or right, and thus precisely adjust the lighting pattern to provide a more consistent and natural interactive effect.

[0157] In addition, an embodiment of the present application further provides a light curtain control device, the light curtain control device comprising:

[0158] An acquisition module, used to acquire an environmental image of the vehicle;

[0159] a first determining module, configured to determine, when a human body image region exists in the environmental image, a first horizontal relative position of the human body image region in the environmental image;

[0160] a second determining module, configured to determine a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position;

[0161] A display module is configured to display a light curtain pattern based on the second horizontal relative position.

[0162] For the principle and implementation process of realizing the light curtain control in this embodiment, please refer to the above embodiments and will not be described in detail here.

[0163] In addition, an embodiment of the present application also proposes a vehicle, which includes a memory, a processor, and a light curtain control program stored in the memory and runnable on the processor. When the light curtain control program is executed by the processor, the steps of the light curtain control method described above are implemented.

[0164] Since the light curtain control program adopts all the technical solutions of all the aforementioned embodiments when executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0165] In addition, an embodiment of the present application further proposes a computer-readable storage medium, on which a light curtain control program is stored. When the light curtain control program is executed by a processor, the steps of the light curtain control method described above are implemented.

[0166] Since the light curtain control program adopts all the technical solutions of all the aforementioned embodiments when executed by the processor, it has at least all the beneficial effects brought by all the technical solutions of all the aforementioned embodiments, which will not be described one by one here.

[0167] It should be noted that, in this document, the terms "comprises," "includes," or any other variations thereof are intended to encompass non-exclusive inclusion, such that a process, method, article, or system comprising a series of elements includes not only those elements but also other elements not explicitly listed, or elements inherent to such process, method, article, or system. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or system comprising the element.

[0168] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0169] Through the above description of the embodiments, those skilled in the art will clearly understand that the above-mentioned embodiments and methods can be implemented using software plus the necessary general-purpose hardware platform. Of course, hardware can also be used, but in many cases the former is a more preferred implementation method. Based on this understanding, the technical solution of this application, or the portion that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, or optical disk) as described above and includes at least two instructions for causing a vehicle to execute the method of each embodiment of this application.

[0170] The above are merely optional embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A light curtain control method, wherein: The light curtain control method comprises: Acquire an environmental image of the vehicle; In the case where a human body image region exists in the environment image, determining a first lateral relative position of the human body image region in the environment image; determining a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position; A light curtain pattern is displayed based on the second lateral relative position.

2. The light curtain control method according to claim 1, wherein: After the step of acquiring the environment image of the vehicle and before the step of determining the first lateral relative position of the human body image area in the environment image, the method further includes: Converting a color-coded YUV environment image to a GPU texture environment image. Inputting the environment image in GPU texture format into a pre-trained human body recognition model to obtain a human body recognition result output by the human body recognition model, wherein the human body recognition result includes human body image area position data; The step of determining a first lateral relative position of the human body image area in the environment image comprises: The first lateral relative position is calculated according to the human body image area position data and the pixel width of the environment image in the GPU texture format.

3. The light curtain control method according to claim 2, wherein: The human image area position data includes the horizontal coordinate of the center pixel of the human image area, and the step of calculating the first horizontal relative position according to the human image area position data and the pixel width of the environment image in the GPU texture format includes: The first horizontal relative position is obtained by dividing the horizontal coordinate of the central pixel of the human body image area by the pixel width of the environment image in the GPU texture format.

4. The light curtain control method according to claim 1, wherein: The step of displaying a light curtain pattern based on the second lateral relative position comprises: Acquire the light curtain pattern, wherein the pixel width of the light curtain pattern is smaller than the pixel width of the light curtain, and the pixel height of the light curtain pattern is equal to the pixel height of the light curtain; Based on the second horizontal relative position, pixels are supplemented outside the light curtain pattern to obtain a target pattern, wherein at the second horizontal relative position of the light curtain pattern in the target pattern, the pixel width and pixel height of the target pattern are respectively consistent with the pixel width and pixel height of the light curtain; The light curtain is controlled to display the target pattern.

5. The light curtain control method according to claim 4, wherein: The human body recognition result also includes human body posture data, and the step of obtaining the light curtain pattern includes: According to the human body posture data, a corresponding light curtain pattern reflecting the human body posture is obtained.

6. The light curtain control method according to claim 4, wherein: The step of obtaining the light curtain pattern comprises: According to the movement state of the human body, a corresponding light curtain pattern reflecting the movement state of the human body is obtained, wherein the movement state of the human body is predetermined based on the first lateral relative position.

7. The light curtain control method according to claim 4, wherein: The human body recognition result also includes human body posture data, and the step of obtaining the light curtain pattern includes: According to the human body posture data and the human body movement state, a corresponding light curtain pattern reflecting the human body posture and the human body movement state is obtained, wherein the human body movement state is predetermined based on the first lateral relative position.

8. The light curtain control method according to any one of claims 6 or 7, wherein: The number of the environment images in the GPU texture format is at least two frames, each frame of the environment image in the GPU texture format corresponds to a first horizontal relative position, and the pixel horizontal coordinates of the environment image in the GPU texture format increase from the first side to the second side. The process of determining the movement state of the human body includes: storing at least two first lateral relative positions in a queue of a preset length; Performing a weighted average calculation on at least two first horizontal relative positions in the queue to obtain a weighted average value, wherein the weights corresponding to the at least two first horizontal relative positions in the queue are set in sequence according to the time sequence of being stored in the queue, and the weight of the first horizontal relative position stored latest in the queue is the highest; If the absolute difference between the first lateral relative position stored in the queue the latest and the weighted average value is less than a preset threshold, determining that the movement state of the human body is stationary; If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is greater than the weighted average value, then determining that the movement state of the human body is moving toward the second side; If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is less than the weighted average value, it is determined that the movement state of the human body is moving toward the first side.

9. A light curtain control device, wherein: The light curtain control device comprises: An acquisition module, used for acquiring an environmental image of the vehicle; A first determining module, configured to determine a first lateral relative position of the human image region in the environment image when a human image region exists in the environment image; a second determining module, configured to determine a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position; A display module is used to display a light curtain pattern based on the second horizontal relative position.

10. A vehicle, wherein: The vehicle includes a memory, a processor, and a light curtain control program stored in the memory and executable on the processor, wherein the light curtain control program implements the following steps when executed by the processor: Acquire an environmental image of the vehicle; In the case where a human body image region exists in the environment image, determining a first lateral relative position of the human body image region in the environment image; determining a second lateral relative position on the light curtain of the vehicle according to the first lateral relative position; A light curtain pattern is displayed based on the second lateral relative position.

11. The vehicle of claim 10, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: Converting a color-coded YUV environment image to a GPU texture environment image. The environment image in the GPU texture format is input into a pre-trained human body recognition model to obtain a human body recognition result output by the human body recognition model, wherein the human body recognition result includes human body image area position data.

12. The vehicle of claim 11, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: The first lateral relative position is calculated according to the human body image area position data and the pixel width of the environment image in the GPU texture format.

13. The vehicle of claim 11, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: The first horizontal relative position is obtained by dividing the horizontal coordinate of the central pixel of the human body image area by the pixel width of the environment image in the GPU texture format.

14. The vehicle of claim 10, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: Acquire the light curtain pattern, wherein the pixel width of the light curtain pattern is smaller than the pixel width of the light curtain, and the pixel height of the light curtain pattern is equal to the pixel height of the light curtain; Based on the second horizontal relative position, pixels are supplemented outside the light curtain pattern to obtain a target pattern, wherein at the second horizontal relative position of the light curtain pattern in the target pattern, the pixel width and pixel height of the target pattern are respectively consistent with the pixel width and pixel height of the light curtain; The light curtain is controlled to display the target pattern.

15. The vehicle of claim 14, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: According to the human body posture data, a corresponding light curtain pattern reflecting the human body posture is obtained.

16. The vehicle of claim 14, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: According to the movement state of the human body, a corresponding light curtain pattern reflecting the movement state of the human body is obtained, wherein the movement state of the human body is predetermined based on the first lateral relative position.

17. The vehicle of claim 14, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: According to the human body posture data and the human body movement state, a corresponding light curtain pattern reflecting the human body posture and the human body movement state is obtained, wherein the human body movement state is predetermined based on the first lateral relative position.

18. A vehicle as claimed in claim 16 or 17, wherein: When the light curtain control program in the memory is executed by the processor, the following steps are also implemented: storing at least two first lateral relative positions in a queue of a preset length; Performing a weighted average calculation on at least two first horizontal relative positions in the queue to obtain a weighted average value, wherein the weights corresponding to the at least two first horizontal relative positions in the queue are set in sequence according to the time sequence of being stored in the queue, and the weight of the first horizontal relative position stored latest in the queue is the highest; If the absolute difference between the first lateral relative position stored in the queue the latest and the weighted average value is less than a preset threshold, determining that the movement state of the human body is stationary; If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is greater than the weighted average value, then determining that the movement state of the human body is moving toward the second side; If the absolute difference between the latest first lateral relative position stored in the queue and the weighted average value is greater than or equal to a preset threshold, and the latest first lateral relative position stored in the queue is less than the weighted average value, it is determined that the movement state of the human body is moving toward the first side.

19. A computer-readable storage medium, wherein: The computer-readable storage medium stores a light curtain control program, and when the light curtain control program is executed by the processor, the steps of the light curtain control method according to any one of claims 1 to 8 are implemented.

Images