Method, system and device for shielding light source in front of vehicle, vehicle and program product

Abstract

The invention discloses a shielding method, system and device for a light source in front of a vehicle, the vehicle and a program product. The method comprises the following steps: acquiring image data in front of a vehicle and motion data of a light source in front of the vehicle in a current time period; determining a category of the light source based on the image data and the motion data; in response to the category being a target category, predicting predicted position information of the light source in a future time period based on the image data, the light source of the target category affecting driving safety of the vehicle, and the future time period being later than the current time period; on the basis of the predicted position information, a control strategy corresponding to a shielding unit in the vehicle in the future time period is determined; and according to a control strategy, a shielding unit is controlled to display the shielding object so as to shield the light source in front of the vehicle. The technical problem that the light glare protection effect is poor is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of automobile lighting, in particular to a method, system, device, vehicle and program product for shielding light source in front of a vehicle. BACKGROUND

[0002] Currently, for the protection of automobile high beam glare at night, the intensity of the glare reflected by the rearview mirror is detected, and the light transmittance of the lens is adjusted to reduce the impact of the glare reflected by the rearview mirror. However, the above method cannot bring glare protection effect when the oncoming vehicle high beam directly shines into the driver's eyes. Therefore, there is still the technical problem of poor light glare protection effect.

[0003] For the above problems, no effective solution has been proposed so far. SUMMARY

[0004] The embodiments of the present application provide a method, system, device, vehicle and program product for shielding light source in front of a vehicle to at least solve the technical problem of poor light glare protection effect.

[0005] According to an aspect of the embodiments of the present application, a method for shielding light source in front of a vehicle is provided, which can include: acquiring image data in front of the vehicle and motion data of a light source in front of the vehicle at a current period, wherein the image content of the image data includes the light source, and the motion data is used to represent the motion state of the light source; determining the category of the light source based on the image data and the motion data; in response to the category being a target category, predicting the predicted position information of the light source in a future period based on the image data, wherein the light source of the target category affects the driving safety of the vehicle, and the future period is later than the current period; determining the control strategy corresponding to the shielding unit in the vehicle in the future period based on the predicted position information, wherein the control strategy is used to represent the rule of the shielding unit displaying a shielding object, and the shielding object is used to shield the light source; and controlling the shielding unit to display the shielding object to shield the light source in front of the vehicle according to the control strategy.

[0006] Optionally, determining the category of the light source based on the image data and the motion data includes: performing feature extraction on the color image in the image data to obtain the brightness feature of the light source and the color temperature feature of the light source; identifying the thermal radiation image in the image data to obtain the movement characteristic of the light source, wherein the movement feature is used to represent whether the light source is a moving hot target; and in response to the brightness feature satisfying a brightness condition, the color temperature feature satisfying a color temperature condition, the movement characteristic representing that the light source is a moving hot target, and the motion data satisfying a motion condition, determining that the category is the target category.

[0007] Optionally, in response to the category being the target category, the predicting, based on the image data, the predicted position information of the light source in the future period comprises: in response to the category being the target category, identifying the image data to obtain current position information of the light source in a current period; and predicting the current position information to obtain the predicted position information.

[0008] Optionally, the identifying, in response to the category being the target category, the image data to obtain the current position information of the light source in the current period comprises: in response to the category being the target category, identifying a color image in the image data to obtain two-dimensional coordinate information of the light source in the current period; and performing fusion processing on three-dimensional coordinate information in the motion data and the two-dimensional coordinate information to obtain the current position information, wherein the fused coordinate information is used to represent a position of the vehicle in a vehicle coordinate system.

[0009] Optionally, the determining, based on the predicted position information, the control strategy of the shielding unit in the vehicle in the future period comprises: converting the predicted position information to obtain a projection position of the light source in the shielding unit; determining a shielding parameter based on the projection position, wherein the shielding parameter is used to represent a size and a light transmission characteristic of a shielding object; and converting the shielding parameter to obtain the control strategy.

[0010] Optionally, the method can further comprise: determining a projection parameter matched with the shielding parameter, wherein the projection parameter is used to represent to-be-displayed content in the shielding area.

[0011] According to another aspect of the embodiments of the present application, a shielding system for a light source in front of a vehicle is also provided, which can comprise: a perception module configured to acquire, in a current period, image data in front of the vehicle and motion data of a light source in front of the vehicle, wherein image content of the image data comprises the light source, and the motion data is used to represent a motion state of the light source; a central control module configured to determine a category of the light source based on the image data and the motion data; in response to the category being a target category, predict, based on the image data, predicted position information of the light source in a future period; and determine, based on the predicted position information, a control strategy of a shielding unit in the vehicle in the future period, wherein the control strategy is used to represent a rule of displaying a shielding object by the shielding unit, and the shielding object is used to shield the light source; and the shielding unit is configured to display shielding content according to the control strategy to shield the light source in front of the vehicle.

[0012] According to a further aspect of the embodiments of the present application, a device for shielding a light source in front of a vehicle is provided. The device can include an acquisition unit configured to acquire image data of the front of the vehicle and motion data of the light source in front of the vehicle at a current time period, wherein the image data includes the light source, and the motion data is used to represent a motion state of the light source; a first determination unit configured to determine a category of the light source based on the image data and the motion data; a prediction unit configured to predict, in response to the category being a target category, predicted position information of the light source at a future time period based on the image data, wherein the target category of the light source affects driving safety of the vehicle, and the future time period is later than the current time period; a second determination unit configured to determine, based on the predicted position information, a control strategy of a shielding unit in the vehicle at the future time period, wherein the control strategy is used to represent a rule of the shielding unit displaying a shielding object, and the shielding object is used to shield the light source; and a control unit configured to control the shielding unit to display the shielding object according to the control strategy, so as to shield the light source in front of the vehicle.

[0013] According to a further aspect of the embodiments of the present application, a computer readable storage medium is provided. The computer readable storage medium includes a stored program, wherein the program, when executed by an apparatus in which the computer readable storage medium is located, performs the method for shielding a light source in front of a vehicle according to the embodiments of the present application.

[0014] According to a further aspect of the embodiments of the present application, a processor is provided. The processor is configured to execute a program, wherein the program, when executed by the processor, performs the method for shielding a light source in front of a vehicle according to the embodiments of the present application.

[0015] According to a further aspect of the embodiments of the present application, a program product is provided. The program product includes computer instructions, wherein the computer instructions, when executed by a processor, implement the method for shielding a light source in front of a vehicle according to the embodiments of the present application.

[0016] According to a further aspect of the embodiments of the present application, a vehicle is provided. The vehicle can be configured to perform the method for shielding a light source in front of a vehicle according to the embodiments of the present application.

[0017] In the embodiment of the present application, in the current period, image data of the front of the vehicle and motion data of the light source in front of the vehicle are acquired, wherein the image content of the image data includes the light source, and the motion data is used to represent the motion state of the light source; based on the image data and the motion data, the category of the light source is determined; in response to the category being a target category, based on the image data, the predicted position information of the light source in a future period is predicted, wherein the target category of the light source affects the driving safety of the vehicle, and the future period is later than the current period; based on the predicted position information, the control strategy corresponding to the shielding unit in the vehicle in the future period is determined, wherein the control strategy is used to represent the rule of the shielding unit displaying the shielding object, and the shielding object is used to shield the light source; and the shielding unit displays the shielding object according to the control strategy to shield the light source in front of the vehicle. That is, in the embodiment of the present application, the category of the light source opposite to the vehicle is determined in real time and accurately through the image data and the motion data of the light source. Further, if the category is a target category, based on the image data, the predicted position information of the light source in the future can be predicted. Then, according to the predicted future position of the light source, a corresponding control strategy is generated to form a local light shielding area through the shielding unit in the vehicle. The harmful high beam in front of the vehicle is accurately shielded without affecting the normal visual perception of the surrounding environment, thereby realizing the technical effect of improving the light glare protection effect and solving the technical problem of poor light glare protection effect. BRIEF DESCRIPTION OF DRAWINGS

[0018] The accompanying drawings, which are included to provide a further understanding of the present application, constitute a part of the present application and illustrate the illustrative embodiments of the present application and their description serve to explain the present application, and do not constitute improper limitations on the present application. In the drawings:

[0019] Figure 1 FIG. 1 is a flowchart of a shielding method for a light source in front of a vehicle according to an embodiment of the present application;

[0020] Figure 2 FIG. 2 is a schematic diagram of a shielding system for a light source in front of a vehicle according to an embodiment of the present application;

[0021] Figure 3 FIG. 3 is a schematic diagram of a shielding device for a light source in front of a vehicle according to an embodiment of the present application. DETAILED DESCRIPTION

[0022] In order for those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor should be within the scope of protection of the present application.

[0023] It is to be understood that the terminology "first", "second" and the like used in the specification and the claims of the application as well as the preceding description of the drawings is merely used to distinguish between similar objects and does not necessarily imply a specific order or chronology. It is to be understood that the data thus used can be interchanged, where appropriate, so that the embodiments of the application described herein can be carried out in a different order than the one illustrated or described herein. Furthermore, the terms "comprising", "having", "including" and "containing" and any variations thereof used herein are intended to cover a non-exclusive inclusion, such that processes, methods, systems, articles, or apparatuses that comprise, have, include or contain a list of steps or elements, but do not necessarily comprise, have, include or contain only those steps or elements, are also covered.

[0024] According to the embodiments of the present application, an embodiment of a method for shielding a light source in front of a vehicle is provided, the steps shown in the flowchart of the drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described herein can be executed in a different order than that shown herein.

[0025] In this embodiment, a method for shielding a light source in front of a vehicle is proposed, which can determine the category of the light source in front of the vehicle in real time and accurately based on image data and motion data of the light source. Further, if the category is a target category, based on the image data, the future position information of the light source can be predicted. Subsequently, according to the predicted future position of the light source, a corresponding control strategy is generated to form a local light shielding area by a shielding unit in the vehicle. The harmful high beam in front of the vehicle is accurately shielded without affecting the normal visual perception of the surrounding environment, thereby achieving the technical effect of improving the light glare protection effect and solving the technical problem of poor light glare protection effect.

[0026] Figure 1 is a flowchart of a method for shielding a light source in front of a vehicle according to an embodiment of the present application. As shown in Figure 1 the method can include the following steps:

[0027] Step S102, in a current period, image data in front of the vehicle and motion data of a light source in front of the vehicle are acquired, wherein the image content of the image data includes the light source, and the motion data is used to represent the motion state of the light source.

[0028] In the technical solution provided in the above step S102 of the present application, the image data can be used to represent the visual environment in front of the vehicle, can be image data of the front of the vehicle obtained by a high dynamic range (High Dynamic Range, HDR for short) forward-looking camera, and can be used to capture light source details even in extreme lighting conditions, ensuring accurate identification of light sources in night or strong light environment. The motion data can be used to represent the motion state of the light source, can include three-dimensional spatial coordinates, motion speed and other data of the front light source, and can be used to determine the distance of the light source from the vehicle, the lateral offset, the height from the ground and other data. For example, it can be data collected by a forward-looking laser radar, which can monitor the relative distance change between the light source and the vehicle, as well as the lateral position and longitudinal height change of the light source in real time.

[0029] Optionally, a HDR forward-looking camera is used to continuously shoot color images in front of the vehicle at a high frame rate to obtain image data. An infrared camera is used to obtain thermal radiation images of the light source in night low light conditions, combined with point cloud data of the forward-looking laser radar, to measure the accurate position and speed of the light source in three-dimensional space to obtain motion data. The high point cloud density and refresh rate of the laser radar enable the system to obtain the motion characteristics of the light source, which can be used to determine the acceleration and possible change in motion direction. It should be noted that this is only an example and the way of obtaining image data and the way of obtaining motion data are not limited.

[0030] For example, the HDR forward-looking camera can be installed below the rearview mirror in the vehicle, with the lens facing straight ahead, for collecting color images of the forward-looking field of view; the infrared camera is coaxially installed with the HDR camera, for capturing thermal radiation signals of the light source of the vehicle in night low light environment, excluding interference from non-vehicle light sources (such as street lamps); the forward-looking laser radar is installed in the center of the grille of the vehicle head, for measuring three-dimensional spatial coordinates (such as distance, lateral offset, height) and motion speed of the target in front. Through the HDR forward-looking camera, the infrared camera and the forward-looking laser radar, image data in front of the vehicle and motion data in front of the vehicle can be obtained.

[0031] Optionally, a perception module is used to realize accurate collection and preliminary identification of the glare light source, which can include but is not limited to: an HDR camera, an infrared camera and a laser radar. The HDR camera extracts the brightness, color temperature and shape features of the light source; the infrared camera confirms whether the light source is a "moving vehicle thermal radiation source"; the laser radar provides distance and motion data of the light source, and the three data are transmitted in real time to the central control module through the vehicle Ethernet.

[0032] Through the above step S102 of the present application, detailed information of the light source in the visual environment in front of the vehicle can be obtained. The HDR front-view camera can capture the details of the light source in the night or in an environment with extremely strong light source. In combination with the thermal radiation image captured by the infrared camera, it can be further confirmed whether the light source comes from a moving vehicle, and the interference of static light sources such as street lamps is excluded, so as to accurately lock the light source that poses a threat to driving safety. Through the high-frequency scanning of the front environment by the front laser radar, not only the position information of the light source is provided, but also the speed and acceleration of the light source can be obtained through the comparison and analysis of multiple scanning data.

[0033] In step S104, the category of the light source is determined based on the image data and the motion data.

[0034] In the technical solution provided in the above step S104 of the present application, the above-mentioned category can be used to represent the properties of the light source, including but not limited to the type of the light source (such as high beam, low beam, fog lamp), the attribution of the light source (such as whether it belongs to an oncoming vehicle or is a fixed lighting device), the intensity and color temperature of the light source, and the dynamic characteristics (such as speed, acceleration, and direction change) of the light source, etc.

[0035] Optionally, after obtaining the image data in front of the vehicle and the motion data of the light source, the category of the light source can be determined based on the image data and the motion data.

[0036] Optionally, after the image pre-processing by the central control module, the light source region is extracted by a You Only Look Once version 8 (YOLOv8) model, and the category of the light source is determined in combination with the infrared thermal radiation features and the laser radar motion features, so as to determine whether the light source is a "harmful high beam".

[0037] Optionally, determining the category of the light source based on the image data and the motion data can include: using the brightness and color temperature features in the image data, in combination with the shape and distribution pattern of the light source, to identify whether the light source belongs to a high beam, a low beam, or a fog lamp, etc. For example, a high beam usually has a high brightness and color temperature, and appears as a pair of light spots in the image with a spacing that conforms to the layout of a general vehicle lamp group. The motion data and the thermal radiation image can be used to determine whether the light source belongs to an oncoming vehicle or a fixed lighting device. For example, the motion speed and direction of the light source of an oncoming vehicle correspond to the driving state of the vehicle, while the motion state of a fixed lighting device (such as a street lamp) is almost zero. The brightness value of the light source in the image data is used to evaluate the intensity of the light source, and it is determined whether the intensity is high enough to form glare. For example, when the brightness value exceeds a threshold value, the system determines that the intensity of the light source is too high and may interfere with driving.

[0038] For example, based on the brightness of the light source (threshold > 3000 nit per square), color temperature (5000-6500 Kelvin), and paired characteristics (distance 1.2-1.8 meters, consistent with vehicle lamp group layout), it is determined whether it is a "harmful high beam"; if it is determined to be harmful, the anti-dazzling process is started.

[0039] Through the above step S104 of the present application, accurate classification of the light source can be realized, and then targeted anti-dazzling measures can be taken. Based on the comprehensive analysis of image data and motion data, the vehicle can effectively identify and distinguish light sources that pose a threat to driving safety from harmless light sources, such as static lighting devices (street lamps) or non-vehicle light sources (natural light, reflected light, etc.), avoiding misjudgment and excessive processing of harmless light sources. Thus, while ensuring the clarity of the driver's field of view, the safety and comfort of night driving are significantly improved.

[0040] Step S106, in response to the category being a target category, predicting the predicted position information of the light source in a future period based on the image data, wherein the light source of the target category affects the driving safety of the vehicle, and the future period is later than the current period.

[0041] In the technical solution provided in the above step S106 of the present application, the above-mentioned target category can be used to represent a light source category that poses a potential threat to driving safety, such as a high beam of an oncoming vehicle, a light source of a vehicle approaching at high speed, etc., and the future period can be used to represent a time range in which the vehicle predicts the motion trajectory of the light source, such as the next 300 milliseconds or longer, which can be a pre-set category of "harmful high beam". The above-mentioned predicted position information can be used to represent the future projection position of the light source on the intelligent windshield, including the horizontal and vertical coordinates, and the possible shading requirement area.

[0042] Optionally, after determining the category of the light source based on the image data and the motion data, the predicted position information of the light source in a future period can be predicted based on the image data in response to the category being a target category.

[0043] Optionally, based on the currently acquired image data and the motion data of the light source, a data prediction algorithm can be used to calculate the predicted position information of the light source in a future period. For example, through a Kalman filter algorithm, the motion trajectory and position change of the light source in a subsequent time period can be predicted based on the image data (including the size, shape, brightness and color temperature characteristics of the light source) and motion data (such as speed, acceleration) of the light source in the current period. By recording and analyzing the position change of the light source in the previous few seconds, the future position of the light source can be predicted using trend analysis. If the light source belongs to a high beam of an oncoming vehicle, the position of the light source will change constantly as the vehicle approaches, and trend analysis can assist the prediction algorithm to more accurately estimate the future position.

[0044] Through step S106 of this application, the future motion trajectory of the target category light source can be accurately predicted. Based on the image data and motion data of the light source acquired in the current time period, advanced data prediction algorithms such as Kalman filtering can be used to predict the instantaneous position change of the light source. Furthermore, the motion trend and development direction of the light source in the future time period can be estimated, including the lateral drift, longitudinal approach speed and possible turning or speed change of the light source.

[0045] Step S108: Based on the predicted location information, determine the control strategy corresponding to the occlusion unit in the vehicle in the future time period, wherein the control strategy is used to characterize the rules for the occlusion unit to display the occlusion object, and the occlusion object is used to occlude the light source.

[0046] In the technical solution provided in step S108 of this application, the aforementioned occlusion unit can be used to represent a pixel area or light-blocking area on the vehicle windshield that can be independently controlled. It can be a polymer-dispersed liquid crystal (PDLC) smart windshield, or simply a smart windshield. This smart windshield may include an augmented reality head-up display (AR-HUD) projection unit. The aforementioned area can be dynamically activated or have its transparency adjusted based on predicted light source position information to block specific light sources without affecting the overall field of vision. The aforementioned control strategy can be a control command, which may include, but is not limited to, target pixel coordinates, voltage values, brightness, etc., and can be used to represent a set of rules or algorithms to determine when, where, and how the occlusion unit displays the occluded object to adapt to dynamic changes in the light source. The control strategy not only covers the activation timing of the occlusion unit but also includes its size, shape, transparency adjustment method, and whether there is grayscale gradient edge processing, ensuring that the occlusion effect meets safety protection requirements while minimizing the impact on the driver's field of vision. An occlusion object can be used to represent the visible occlusion pattern formed by an occlusion unit on a vehicle windshield, which is used to physically isolate a specific light source.

[0047] Optionally, after predicting the predicted location information of the light source in the future time period based on the image data, the control strategy corresponding to the occlusion unit in the vehicle in the future time period can be determined based on the predicted location information.

[0048] In this embodiment, in addition to determining the predicted projection position in the future time period based on the predicted position information, the real-time position of the light source in the vehicle coordinate system can be calculated by fusing camera and radar data through Kalman filtering (e.g., x: lateral offset, y: longitudinal distance, z: height); and then the projection position of the light source on the smart windshield in the next 300ms can be predicted by the EKF algorithm.

[0049] In this embodiment, a human-machine interaction execution module is used to achieve glare shading and information compensation. This human-machine interaction execution module may include a shading unit. The intelligent windshield in the shading unit can replace the traditional windshield, covering the driver's forward core field of vision; the AR-HUD is installed behind the instrument panel, with the projection light path pointing towards the driver's field of vision area of ​​the intelligent windshield.

[0050] Optionally, the exact location of the obstruction unit on the windshield can be determined based on the predicted future location information of the light source. This location can cover the predicted area of ​​harmful light sources while minimizing obstruction of non-glare areas to maintain clear visibility for the driver.

[0051] Optionally, a strategy for the size and shape of the blocking unit can be developed to adapt to dynamic changes in the light source. For example, if the predicted light source is the high beam of a sedan or large vehicle, the blocking unit may be adjusted to a rectangle to cover the higher-positioned light source; for the high beam of a regular sedan, the blocking unit may remain elliptical to reduce the obstructed area. The transparency of the blocking unit can be dynamically adjusted based on the predicted intensity of the light source (e.g., brightness, color temperature) and ambient light conditions. For example, in bright light conditions, the transparency of the blocking unit can be set higher to effectively block glare; in low light conditions, the transparency needs to be reduced to avoid unnecessary darkening of the driver's field of vision. This is to reduce visual abrupt changes and driver fatigue.

[0052] Optionally, the edges of the occlusion unit are subjected to a grayscale gradient feathering process. That is, on the edges of the occlusion unit, the transparency linearly decreases from a set value at the center of the occlusion to the value of the surrounding transparent area within a few pixels, achieving a smooth transition. The activation and deactivation times of the occlusion unit are determined. For example, the occlusion unit is activated 100 milliseconds before a light source is expected to enter the harmful visual field, and gradually returns to a transparent state as the light source moves away or its intensity drops below a safe threshold, adapting to real-time changes in the light source. In conjunction with the occlusion unit, a strategy for displaying key driving information in the occluded area of ​​the HUD is determined, including information content, display brightness, and position, ensuring that the driver can still obtain necessary information when harmful light sources are obscured, thus improving driving safety.

[0053] For example, the occlusion block parameters are calculated based on the predicted position. These parameters include, but are not limited to: position P(x',y') (x'=x×windshield lateral coefficient, y'=z×windshield longitudinal coefficient, the coefficients are determined by windshield calibration), width W, height H, and transparency α. Simultaneously, the content projected onto the occlusion block by the HUD (e.g., vehicle speed, speed limit) and brightness are determined. Based on the occlusion parameters, a corresponding control strategy can be derived.

[0054] Through step S108 of this application, the optimal position of the blocking unit on the smart windshield can be accurately calculated based on the future projection position of the light source, ensuring that the blocking object can accurately cover the light source, reducing glare interference to the driver's vision, and without affecting the field of vision of the unblocked area.

[0055] In step S110, according to the control strategy, the occlusion unit is controlled to display the occlusion object to block the light source in front of the vehicle.

[0056] In the technical solution provided by step S110 of this application, after determining the control strategy corresponding to the occlusion unit in the vehicle in a future time period based on the predicted location information, the occlusion unit can be controlled to display the occlusion object in accordance with the control strategy to block the light source in front of the vehicle.

[0057] Optionally, according to the control strategy, the central control module sends pixel driving commands to the intelligent windshield driving circuit. Through the driving control commands, the corresponding pixels are energized to form a blocking block. That is, the blocking unit is controlled to display the blocking object to block the light source in front of the vehicle. Simultaneously, a projection command can be sent to the AR-HUD to project information onto the blocking block area.

[0058] Optionally, according to a predetermined strategy, the vehicle will accurately activate the masking unit at a predicted time and adjust it to the correct position to cover the projection area of ​​the light source on the windshield. This process is dynamic; the position of the masking unit updates in real time as the light source moves, ensuring the continuous effectiveness of the masking. The size and shape of the masking unit are adjusted based on the predicted nature of the light source (e.g., high beams from an SUV or a regular sedan). Simultaneously with the masking operation, the HUD's projected content, display brightness, and position are intelligently adjusted to ensure that key driving information, such as vehicle speed, speed limits, and navigation routes, is still clearly displayed in the masked unit area.

[0059] In this embodiment, the sensing module can continuously monitor the light source status: if the lidar detects that the longitudinal distance of the light source is >100m (the light source is far away), or the camera detects that the light source brightness is <1000 cd / m², then the central control module instructs the corresponding pixel of the smart windshield to power off (restoring 90% light transmittance), the AR-HUD stops projecting in that area, and returns to step 1; if the light source continues to exist, then repeat the above steps S102-S108 to dynamically update the occlusion block parameters.

[0060] For example, the system determines whether a light source is a "harmful high beam" based on its brightness (threshold > 3000 cd / m²), color temperature (5000-6500 K), and pairing characteristics (spacing 1.2-1.8 m, consistent with vehicle lighting layout). If it is determined to be harmful, the anti-glare process is initiated. Furthermore, the system can generate pixel driving instructions for the smart windshield (including target pixel coordinates and voltage values) and projection instructions for the HUD (including projection content, brightness, and position), which are sent to the execution module via the CANFD bus (instruction transmission delay < 10 ms). The execution module then controls the occlusion unit to display the occlusion object according to the control strategy, thereby blocking the light source in front of the vehicle.

[0061] Through steps S102 to S110 of this application, image data of the area in front of the vehicle and motion data of the light source in front of the vehicle are acquired. The image data includes the light source, and the motion data characterizes the motion state of the light source. Based on the image data and motion data, the category of the light source is determined. In response to the category being the target category, the predicted position information of the light source in a future time period is predicted based on the image data. The target category of the light source affects the driving safety of the vehicle, and the future time period is later than the current time period. Based on the predicted position information, a control strategy corresponding to the occlusion unit in the vehicle is determined for the future time period. The control strategy characterizes the rules for the occlusion unit to display the occlusion object, and the occlusion object is used to occlude the light source. According to the control strategy, the occlusion unit is controlled to display the occlusion object to occlude the light source in front of the vehicle. In other words, in this embodiment, the category of the light source facing the vehicle is determined in real time and accurately using image data and the motion data of the light source. Further, if the category is the target category, the predicted position information of the light source in the future can be predicted based on the image data. Subsequently, based on the predicted future position of the light source, a corresponding control strategy is generated to form a local shading area through the occlusion unit in the vehicle. It precisely blocks harmful high beams from the front of the vehicle without affecting normal visual perception of the surrounding environment, thereby improving the glare protection effect and solving the technical problem of poor glare protection.

[0062] The method described in this embodiment will be further described below.

[0063] As an optional implementation, the category of the light source is determined based on image data and motion data, including: extracting features from the color image in the image data to obtain the brightness features and color temperature features of the light source; identifying the thermal radiation image in the image data to obtain the motion characteristics of the light source, wherein the motion features are used to characterize whether the light source is a moving thermal target; and determining the category as the target category in response to the brightness features satisfying the brightness condition, the color temperature features satisfying the color temperature condition, the motion characteristics characterizing the light source as a moving thermal target, and the motion data satisfying the motion condition.

[0064] In this embodiment, the brightness feature described above can be used to represent the intensity of the light source, i.e., the brightness value of the light source area in the color image. A high brightness value may indicate a strong light source intensity, posing a potential threat to driving safety. The color temperature feature described above can be used to represent the spectral characteristics of the light source, i.e., the color temperature of the light emitted by the light source. The radiation image described above can be used to represent the image characteristics of the light source in the infrared band, captured by an infrared camera, to identify whether the light source is a moving thermal target. Moving thermal targets typically refer to vehicle high beams, which can appear as moving heat sources in infrared images and can be distinguished from fixed heat sources such as streetlights. The brightness condition described above can be used to represent a threshold or a set of thresholds to determine whether the intensity of the light source has reached a level requiring shading. The color temperature condition described above can be used to represent the effective range of the light source's color temperature, to further distinguish the type of light source. The movement characteristic described above can be used to indicate whether the light source is a moving thermal target, i.e., whether the position of the light source in the thermal radiation image changes over time, and the speed and direction of the change.

[0065] Optionally, the color image can be preprocessed using an HDR camera, including noise reduction, white balance adjustment, and distortion correction. Then, image processing algorithms, such as edge detection and thresholding, can be used to identify and extract high-brightness areas in the image to obtain the brightness characteristics of the light source. The value of the brightness characteristic reflects the intensity of the light source; high brightness values ​​indicate light sources that may cause glare. Based on the red, green, and blue (RGB) values ​​of the color image, color temperature estimation algorithms (such as the gray-world algorithm and the minimum color temperature error algorithm) can be used to calculate the color temperature characteristics of the light source. Color temperature characteristics help distinguish different types of light sources because different types of light sources (such as car high beams, low beams, and fog lights) often have different color temperatures.

[0066] Optionally, the thermal radiation images captured by the infrared camera can reflect the temperature distribution of the light source. Through image processing, such as temperature thresholding, areas with higher temperatures, i.e., heat sources, can be identified. If the position of the heat source changes in the image over time, it indicates that the light source is moving. By comparing several consecutive frames of thermal radiation images, the differences between the images can be analyzed to determine the positional changes of the heat source, thereby inferring the movement characteristics of the light source. Movement characteristics include the speed and direction of the light source's movement, used to determine whether the light source is a moving thermal target.

[0067] Optionally, the brightness characteristics, color temperature characteristics, and motion characteristics are compared with preset brightness conditions, color temperature conditions, and motion data conditions. If the characteristics of the light source simultaneously meet all conditions—that is, the brightness is higher than the threshold, the color temperature is within the effective range, it belongs to a moving thermal target, and the motion data matches an oncoming vehicle—then the system identifies the light source as a target category of "harmful high beams." Through the judgment logic, the system can accurately distinguish between harmful light sources (such as the high beams of oncoming vehicles) and harmless light sources (such as fixed streetlights and non-vehicle light sources).

[0068] In this embodiment, a "multi-feature fusion recognition" method is adopted. First, the brightness and color temperature features of the light source are extracted by an HDR camera (excluding low beam lamps with a brightness of <3000cd / m² and fog lamps with a color temperature of <5000K). Then, an infrared camera is used to confirm whether the light source is a "moving thermal target" (excluding fixed street lamps). Finally, the speed of the light source is detected by a lidar (excluding stationary targets with a speed of <10km / h) in order to accurately determine the type of light source.

[0069] In this embodiment, the above steps enable accurate identification and classification of light source types. An HDR camera is used to preprocess the color image, including image noise reduction and white balance adjustment, to reduce the impact of ambient light and image noise on brightness feature extraction. High-brightness areas in the image are identified, and the brightness features of the light source are extracted. Brightness features can be used to determine whether the light source intensity reaches a threshold that may cause glare, thus serving as an important basis for occlusion decisions. Color temperature feature extraction is based on the RGB values ​​of the color image. The system calculates the color temperature of the light source using color temperature estimation algorithms, such as the gray-world algorithm. Different types of light sources, such as high beams, low beams, and fog lights, have different color temperatures; color temperature features help the system further distinguish the type of light source. Through thermal radiation images captured by an infrared camera, the system can identify whether the light source is moving. This identification process can be accomplished by comparing the changes in the position of the heat source in several consecutive frames of thermal radiation images. If the position of the heat source changes significantly over time, it indicates that the light source has moving characteristics, and it is likely the high beam of an oncoming vehicle. By combining motion data provided by lidar, the movement characteristics of the light source and the likelihood that the light source belongs to an oncoming vehicle can be further verified. For example, the relative speed and direction of the light source can be calculated and compared with the system's preset oncoming vehicle motion characteristics. The system compares the brightness characteristics, color temperature characteristics, and movement characteristics with preset brightness conditions, color temperature conditions, and motion conditions. If the light source's brightness characteristics are higher than the preset brightness threshold, the color temperature characteristics are within the preset effective range, and the movement characteristics indicate that the light source is a moving thermal target, and the light source's motion data matches the motion characteristics of the oncoming vehicle, then the light source can be identified as a "harmful high beam," meaning it is a target requiring anti-glare treatment.

[0070] As an optional implementation, in response to the category being the target category, predicting the predicted location information of the light source in a future time period based on image data includes: in response to the category being the target category, identifying the image data to obtain the current location information of the light source in the current time period; and predicting the current location information to obtain the predicted location information.

[0071] In this embodiment, the aforementioned current position information can be used to represent the position of the light source relative to the vehicle coordinate system at the current time point, and the aforementioned predicted position information can be used to represent the expected position information of the light source in the future period of time.

[0072] Optionally, the color image captured by the HDR camera is used to identify the light source area, and the position of the light source is located by image processing algorithms (such as background subtraction and contour detection); combined with the thermal radiation image captured by the infrared camera, the light source is further confirmed to be a moving thermal target, eliminating the interference of static light sources (such as streetlights); and the precise coordinates of the light source, including the distance from the vehicle, lateral offset and height information, are obtained by using data measured by the lidar to improve the accuracy of the position information.

[0073] Optionally, a Kalman filter algorithm is used to fuse image data and motion data to construct a dynamic model of the light source, which can reflect the motion trend of the light source. Based on the constructed dynamic model, the motion trajectory of the light source in future time periods is predicted, including the prediction of the position of the light source on the windshield, as well as the prediction of lateral offset and height. Considering the motion parameters of the light source (e.g., oncoming vehicles), such as acceleration and steering, the predicted position information is corrected by extending the prediction algorithm such as Kalman filter or particle filter to adapt to the nonlinear motion of the light source.

[0074] For example, when an oncoming vehicle approaches the vehicle at 50 kilometers per hour, it can be predicted that the light source will shift laterally by 0.05 meters (m) and longitudinally by 0.03 meters after 30 milliseconds. Based on this predicted position information, the position of the blocking block can be adjusted in advance to ensure that the blocking block always covers the light source and avoid "lagging blocking".

[0075] Optionally, a Kalman filter algorithm is used to fuse the light source's coordinates in the "two-dimensional image coordinates of the forward-looking camera" and the "three-dimensional spatial coordinates of the lidar" to calculate the precise position of the light source in the vehicle's coordinate system.

[0076] For example, the central control module calculates the projection position of the light source on the smart windshield (e.g., coordinates P (x=0.3m, y=1.5m), with the lower left corner of the windshield as the origin); based on the projection position, it generates the parameters of the occlusion block (width W=2m, height H=0.3m, transparency α=68%), sends a command to the smart windshield drive circuit, controls the corresponding pixel unit to be powered on (voltage 3.5V), forming an occlusion block with a light-blocking rate of 68%, and the 5 pixels (5 mm) at the edge of the occlusion block adopt "grayscale gradient feathering" - linearly reducing the light-blocking rate from 68% at the center to 10% at the edge, avoiding harsh boundaries that cause visual fatigue to the driver.

[0077] In this embodiment, the above steps enable dynamic prediction of the target light source. In response to a signal that the light source category is identified as the target category, image data processing is initiated. The high dynamic range color image captured by the HDR camera first undergoes image processing algorithms (such as background subtraction and contour detection) to pinpoint the precise location of the light source at the current moment. The thermal radiation image captured by the infrared camera distinguishes between moving thermal targets and static light sources. Temperature distribution characteristics further confirm that the light source is a moving thermal target, eliminating interference from non-vehicle light sources and ensuring the accuracy of light source identification.

[0078] As an optional implementation, in response to the category being the target category, the image data is identified to obtain the current position information of the light source in the current time period, including: in response to the category being the target category, the color image in the image data is identified to obtain the two-dimensional coordinate information of the light source in the current time period; the three-dimensional coordinate information and the two-dimensional coordinate information in the motion data are fused to obtain the current position information, wherein the fused coordinate information is used to characterize the position of the vehicle in its own coordinate system.

[0079] In this embodiment, the aforementioned two-dimensional coordinate information can be used to represent the position of the light source in the image, typically the two-dimensional coordinates of the light source on the color image captured by an HDR camera. The aforementioned three-dimensional coordinate information can be used to represent the precise position of the light source in real space, obtained through sensors such as LiDAR, including the straight-line distance between the light source and the vehicle, as well as the lateral and vertical offset of the light source relative to the vehicle's centerline. The aforementioned current position information can be used to represent the relative position of the light source with respect to the vehicle coordinate system at the current time point, i.e., the precise position of the light source in the vehicle coordinate system, including two-dimensional projected coordinates and three-dimensional distance information. The aforementioned fused coordinate information is used to characterize the vehicle's position in the vehicle coordinate system.

[0080] Optionally, the outline of the light source can be identified using the color image captured by the HDR camera through image processing algorithms (such as edge detection and threshold segmentation); and the specific position of the light source in the image can be located using an object detection algorithm to obtain the two-dimensional coordinate information of the light source.

[0081] Optionally, fusing the three-dimensional and two-dimensional coordinate information in the motion data to obtain the current position information may include using a Kalman filter algorithm to fuse the two-dimensional image coordinate information provided by the HDR camera with the three-dimensional coordinate information provided by the LiDAR, taking into account the mapping relationship between image coordinates and real-world coordinates, as well as the real-time performance and accuracy of the LiDAR data; during the fusion process, the algorithm will correct the real-time position of the light source to eliminate any possible positioning errors and obtain the fused coordinate information, that is, the current position information of the light source in the vehicle coordinate system, including two-dimensional projected coordinates and three-dimensional distance information.

[0082] Optionally, the "vehicle coordinate system - intelligent windshield coordinate system" can be calibrated in advance by setting a standard calibration board (including marker points with known coordinates) 10 meters in front of the vehicle, capturing images of the calibration board with an HDR camera, measuring the three-dimensional coordinates of the marker points with a lidar, establishing a mapping relationship between the two (lateral coefficient, longitudinal coefficient), and storing it in the storage unit of the central control module to ensure that the light source position conversion error is <0.05 meters.

[0083] In this embodiment, the above steps enable precise real-time positioning of the target light source. Image processing algorithms (e.g., edge detection, threshold segmentation) are used to identify the light source's outline, and then object detection algorithms are used to pinpoint the light source's specific location in the image, obtaining its two-dimensional coordinate information. Three-dimensional coordinate information is acquired using measurement data from sensors such as LiDAR. By fusing the two-dimensional and three-dimensional coordinate information, and using a Kalman filter algorithm to correct the light source's real-time position, positioning errors are eliminated, resulting in fused coordinate information of the light source in the vehicle's coordinate system—that is, precise position information, including two-dimensional projected coordinates and three-dimensional distance information.

[0084] As an optional implementation, based on predicted location information, the control strategy corresponding to the occlusion unit in the vehicle is determined in a future time period, including: converting the predicted location information to obtain the projection position of the light source in the occlusion unit; determining occlusion parameters based on the projection position, wherein the occlusion parameters are used to characterize the size and light transmission characteristics of the occluded object; and converting the occlusion parameters to obtain the control strategy.

[0085] In this embodiment, the aforementioned projection position can be used to represent the precise projection position of the light source on the windshield in a future time period, including the coordinates of the light source on the windshield glass and the position adjustment information after the light source moves. The aforementioned occlusion parameters can be used to characterize the size and light transmission characteristics of the occluded object, such as the specific width, height, transparency, and shape of the occlusion block.

[0086] Optionally, the predicted three-dimensional coordinate information (distance, lateral offset, height) is converted into two-dimensional projected coordinates. Based on the predicted projection position of the light source, the width W and height H of the occlusion block are calculated. Based on the brightness characteristics of the light source, the transparency α of the occlusion block is adjusted to ensure that the occlusion effect matches the light source intensity and avoid excessive occlusion affecting other visual information. Edge feathering can be performed to soften the boundary of the occlusion block and reduce the impact of visual abrupt changes on the driver.

[0087] Optionally, the occlusion parameters (W, H, α) are converted into specific control instructions, such as controlling which pixels are powered to a specified voltage value; the display content and brightness parameters of the HUD are simultaneously converted into specific projection instructions, including content layout, brightness adjustment and projection position; the generation and disappearance logic of the occlusion block is determined, such as starting occlusion when the light source approaches and stopping occlusion when the light source moves away or the brightness decreases.

[0088] Optionally, based on the predicted location information (x, y, z), the occlusion position P(x', y') (x' = x × windshield lateral coefficient, y' = z × windshield longitudinal coefficient, the coefficients are determined by windshield calibration), width W, height H, and transparency α can be calculated; simultaneously, the content projected onto the occlusion block by the HUD (vehicle speed, speed limit) and brightness can be determined. By transforming the occlusion parameters, the corresponding control strategy can be obtained.

[0089] For example, calculate the position of the light source in the vehicle's coordinate system (x=0.3m, y=100m, z=1.3m); predict the projection position of the light source on the smart windshield after 300 milliseconds (ms) (T0+358ms) using the EKF algorithm: x'=0.3m×1.2 (windshield lateral calibration coefficient)=0.36m, y'=1.3m×0.6 (windshield longitudinal calibration coefficient)=0.78m (the lower left corner of the windshield is the origin). Based on the projection position, the occlusion block parameters (i.e., occlusion parameters) can be calculated. These parameters may include, but are not limited to: width W = (0.2m × 100m) / 10 = 2m (actual width of the light source is 0.2m); height H = 0.4m (SUV high beams are higher, so this is adjusted to 0.4m); transparency α = (230 / 255) × 80% ≈ 72%; edge feathering: 5 pixels (5mm), linearly decreasing from 72% to 10%. Furthermore, a 3.8V voltage command (corresponding to a 72% occlusion rate) can be sent to the pixel unit of the smart windshield centered at (0.36m, 0.78m) with a length of 2m × 0.4m via a driver chip. This voltage command can serve as a control strategy. According to this control strategy, the pixels are energized to form an occlusion block (response time 45m / s, occlusion completed in T0 + 123m / s).

[0090] Optionally, the width W is calculated as follows: W = (actual width of the light source × longitudinal distance between the light source and the vehicle) / 10 (the actual width of the light source is 0.2 meters by default, based on the size of common sedan high beam headlights); for example, when the light source is 100 meters away from the vehicle, W = 2 meters; when the distance is 50 meters, W = 1 meter – ensuring that the size of the obstruction block matches the image size of the light source in the field of vision, avoiding excessive obstruction; the height H is fixed at 0.3 meters, covering the height range of common vehicle high beam headlights (1.1-1.4 meters, corresponding to the projection height on the smart windshield). It should be noted that the above figures are only examples and are not specifically limited here.

[0091] In this embodiment, the transparency of the blocking block can be adaptive, with transparency α (blocking rate): α = (light source grayscale value / 255) × 80% (the light source grayscale value is collected by an HDR camera, ranging from 0-255); for example, when the light source grayscale value is 220 (strong light), α = 68%; when the grayscale value is 150 (weak light), α = 47%—ensuring that the blocking rate matches the glare intensity and avoiding excessive or insufficient blocking. The shape of the blocking block can also be adaptive, defaulting to an "ellipse" (horizontal major axis, vertical minor axis) to reduce the area obstructed from the field of vision; when a light source such as an "SUV or other tall vehicle" is detected, it automatically adjusts to a "rectangular" shape (height increased to 0.4m) to ensure coverage of higher-positioned high beams.

[0092] In this embodiment, through the above steps, the predicted three-dimensional coordinate information (light source distance, lateral offset, and height) can be used to calculate the two-dimensional projection coordinates of the light source on the windshield in the future time period using a conversion algorithm between the vehicle coordinate system and the windshield coordinate system. These projection coordinates are used to guide the generation position of the occlusion block. Based on the predicted projection position of the light source, the system calculates the width W and height H of the occlusion block to ensure that the occlusion area can fully cover the light source while minimizing the impact on the field of vision of non-light source areas. The transparency α of the occlusion block is adjusted according to the brightness of the light source to match the occlusion effect with the light source intensity, avoiding loss of visual information due to excessive occlusion. The occlusion parameters (W, H, α) are converted into control commands for the intelligent windshield, and the driving circuit controls specific pixel areas to be energized to a specified voltage value to achieve the occlusion effect. Simultaneously, the display content, brightness, and position of the HUD are converted into specific projection commands to ensure that key driving information can be displayed promptly and clearly on the occluded area.

[0093] As an optional implementation, the method further includes: determining projection parameters that match the occlusion parameters, wherein the projection parameters are used to characterize the content to be displayed in the occlusion area.

[0094] In this embodiment, the projection parameters can be used to represent the content to be displayed in the occluded area, and may include data such as the vehicle's speed and the vehicle's movement in the occluded area.

[0095] Optionally, in this embodiment, determining the projection parameters is a crucial step in ensuring that driving information is clearly visible in the obscured area. The projection parameters can be used to represent the content information to be displayed in the obscured area, including but not limited to the type of content to be projected (such as vehicle speed, speed limit ahead, navigation route, etc.), the layout and position of the content, and brightness adjustment.

[0096] Optionally, the type of information to be displayed on the HUD is determined based on the size, position, and shape of the obstruction block. For example, vehicle speed and the speed limit ahead can be displayed in a larger obstruction block to ensure that the information is large enough for the driver to quickly identify; in a smaller obstruction block, only the vehicle speed can be displayed to reduce information interference. The projection layout of the HUD is adjusted according to the center position and size of the obstruction block to ensure that the information content completely covers the obstructed area.

[0097] Optionally, the HUD's projection brightness is automatically adjusted to ensure that information is clearly visible in the obscured area, without creating a strong contrast with the ambient light outside the obscured area due to excessive brightness, thus avoiding new visual interference. As the obscuration parameters (W, H, α) are dynamically adjusted, the projection parameters, including content, layout, and brightness, are updated synchronously to ensure that the information displayed in the obscured area always matches the current driving scenario.

[0098] For example, by sending commands to the AR-HUD simultaneously, in addition to displaying the obstruction block in the obstructed area, the text and icons of "vehicle speed 60km / h + speed limit ahead 70km / h" can also be projected in the obstruction block area. The projection brightness is adjusted to 150cd / m² and matched with the ambient light intensity in real time by the light sensor to ensure that the information is clear and does not interfere with the field of vision.

[0099] In the embodiments of this application ， By combining image data and light source motion data, the type of oncoming light source from the vehicle is determined in real time and with precision. Furthermore, if the type is the target type, the future predicted position of the light source can be generated based on the image data. Subsequently, based on the predicted future position of the light source, a corresponding control strategy is generated to create a local glare-blocking area using shading units within the vehicle. This precisely blocks harmful high beams from the front of the vehicle without affecting normal visual perception of the surrounding environment, thereby improving the glare protection effect and solving the technical problem of poor glare protection.

[0100] The technical solutions of the embodiments of this application will be illustrated below with reference to preferred embodiments.

[0101] Currently, automotive high beam glare protection technology mainly revolves around "passive protection" or "single module adjustment," with core solutions including three categories: automatic anti-glare interior / exterior rearview mirrors, driver-wearing anti-glare glasses, and ordinary head-up display (HUD) brightness adjustment.

[0102] In related technologies, existing automatic anti-glare rearview mirrors can only reduce glare interference by detecting the intensity of glare reflected from the rearview mirror and adjusting the light transmittance of the electrochromic layer of the lens. However, they cannot handle the direct glare from the high beams of oncoming vehicles in the driver's forward field of vision. When the high beams of oncoming vehicles shine directly into the driver's eyes, it can cause the driver to experience "night blindness" for 0.5-2 seconds, increasing the risk of rear-end collisions and lane departure. The fundamental reason is that this type of solution only protects the "rearview mirror reflective surface" and does not cover the driver's core forward field of vision.

[0103] Wearing anti-glare glasses is a common passive solution for drivers, but it has two key drawbacks: First, the lenses are mostly fixed in their glare reduction (e.g., 30%-50%), unable to dynamically adapt to different glare intensities (e.g., strong oncoming light at close range versus weak light at a distance), easily leading to excessively dark vision in low-light environments; second, some anti-glare glasses have color shifts, which may affect the recognition accuracy of traffic lights (e.g., red and green lights), increasing the risk of misjudgment. Ordinary in-vehicle HUDs can only project basic information such as vehicle speed and navigation, and their nighttime brightness adjustment relies on manual operation or a fixed light threshold. When encountering oncoming high beam glare, the light projected by the HUD overlaps with the glare, actually increasing visual interference. The core problem is that ordinary HUDs lack synergy with glare protection, lacking a "partial glare reduction + information compensation" linkage mechanism. While existing solutions attempt to adjust the forward field of vision transmittance using a single piece of electrochromic glass, they cannot achieve "precise local shading"—if the entire piece of glass blocks the light, the field of vision in non-glare areas will darken, affecting the driver's observation of road markings and pedestrians; at the same time, existing solutions do not combine multi-sensor fusion for dynamic tracking and prediction of glare sources, so the shading area cannot follow in real time when oncoming vehicles move, resulting in poor protection effectiveness.

[0104] To address the aforementioned issues, this embodiment utilizes "precise local glare blocking + clear visibility in unobstructed areas" to reduce the "night blindness time" caused by glare in the driver's forward vision from 0.5-2 seconds to less than 0.1 seconds, thereby reducing lane departure and rear-end collision risks caused by glare by more than 80% (based on simulation test data: in oncoming high beam scenarios at night, this system can improve the driver's reaction time from 1.2 seconds to 0.3 seconds).

[0105] In this embodiment, the driving field of vision and information acquisition efficiency are optimized. The partial shading design of the intelligent windshield avoids the "darkening of the field of vision caused by the whole block of light." At the same time, the AR-HUD projects key information such as vehicle speed, road speed limit, and navigation route in the shading area—blocking glare without losing necessary driving information, improving information acquisition efficiency by 50% compared to ordinary HUDs. It achieves full automation and low operational burden. The system automatically completes the "glare recognition-localization-prediction-shading-feedback" closed loop through multi-sensor fusion, eliminating the need for manual adjustments by the driver (such as switching rearview mirror modes or adjusting HUD brightness), reducing the operational burden of nighttime driving, and is especially suitable for long-distance driving scenarios. It is highly adaptable and covers a wide range of scenarios, supporting protection against different glare intensities (such as strong light at close range and weak light at long distance), different light source types (such as high beams of cars and high beams of SUVs), and different road conditions (such as urban roads and rural roads). Furthermore, it can be upgraded through algorithms to adapt to the light source characteristics of new vehicle models, exhibiting better compatibility than existing fixed-parameter solutions.

[0106] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them; the numbers in these embodiments are only illustrative and are not intended to be specific limitations. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application and are not specifically limited here.

[0107] The "Vehicle High Beam Directional Anti-Glare System Based on Intelligent Windshield and Collaborative Perception" of this application consists of three core modules: a perception module, a central control module, and a human-machine interaction execution module. The modules interact with each other through vehicle Ethernet and bus. The composition, layout, and function of each module are as follows.

[0108] In this embodiment, the perception module may include: accurately acquiring and initially identifying glare sources, comprising: a high dynamic range (HDR) forward-facing camera, an infrared camera, and a forward-facing lidar; deployment locations: the HDR forward-facing camera (dynamic range 140dB, frame rate 30fps): installed below the rearview mirror inside the vehicle, with the lens facing directly forward, for acquiring color images of the forward field of view; the infrared camera (resolution 160×120, spectral range 8-14μm): coaxially mounted with the HDR camera, for capturing the thermal radiation signal of vehicle light sources in low-light environments at night, eliminating interference from non-vehicle light sources (such as streetlights); and the forward-facing lidar (point cloud density 1536 points / °, refresh rate 10Hz): installed in the center of the front grille, for measuring the three-dimensional spatial coordinates (distance, lateral offset, height) and speed of the target in front.

[0109] In this embodiment, the core functions of the perception module may include: HDR camera extracting the brightness, color temperature, and shape features of the light source (referring to the light source in the forward field of view); infrared camera confirming whether the light source is a "moving vehicle heat radiation source"; and lidar providing the distance (distance between the light source and the vehicle) and motion data of the light source. All three data are transmitted to the central control module in real time (transmission delay < 50ms).

[0110] In this embodiment, the central control module may include: data fusion, decision-making, and drive control. Hardware components include: an in-vehicle high-performance artificial intelligence (AI) chip, peripheral drive circuits (including a smart windshield drive chip and a HUD control chip), and a storage unit (EMMC128GB, used to store algorithm models and logs); Installation location: installed in the electronic compartment below the vehicle's center console, connected to the vehicle's cooling system via heat sinks to ensure a stable operating temperature between -40℃ and 85℃.

[0111] In this embodiment, the core functions of the central control module may include: image processing: preprocessing HDR camera images (including noise reduction, white balance, and distortion correction), and extracting the light source area using a TensorRT-accelerated YOLOv8 model (detection accuracy > 98%, detection speed < 20ms); data fusion: using a Kalman filter algorithm to fuse the light source in the "two-dimensional image coordinates of the forward-looking camera" and the "three-dimensional spatial coordinates of the lidar", and calculating the precise position of the light source in the vehicle coordinate system (positioning error < 0.1m); logical decision-making: determining whether the light source is a "harmful high beam" based on its brightness (threshold > 3000cd / m²), color temperature (5000-6500K), and pairing characteristics (spacing 1.2-1.8m, conforming to the vehicle's lamp layout); if determined to be harmful, initiating the anti-glare process; and drive control: generating pixel drive instructions for the intelligent windshield (including target pixel coordinates and voltage values) and projection instructions for the HUD (including projection content, brightness, and position), and sending them to the execution module via the CANFD bus (instruction transmission delay < 10ms).

[0112] In this embodiment, the human-machine interaction execution module may include glare shading and information compensation, and its components are: PDLC smart windshield (hereinafter referred to as "smart windshield") and AR-HUD projection unit; the deployment location is as follows: smart windshield: replaces the traditional windshield and covers the driver's core forward field of vision (1.2m horizontal coverage and 0.8m vertical coverage); AR-HUD: installed behind the instrument panel, with the projection light path facing the driver's field of vision area of ​​the smart windshield (projection distance 12m, field of view 13°).

[0113] In this embodiment, the core functions and technical details of the human-computer interaction execution module may include: an intelligent windshield, using polymer dispersed liquid crystal (PDLC) material, divided into 1mm×1mm independent pixel units (total number of pixels: 1.2m / 1mm×0.8m / 1mm=960,000); the working principle is as follows: when not powered, the liquid crystal molecules are randomly arranged, with a light transmittance of 90% (consistent with traditional windshields); when powered, the liquid crystal molecules are arranged in an orderly manner, and the light transmittance can be changed by adjusting the voltage (0-5V). Shading rate (0%-90%) – The higher the voltage, the higher the shading rate, achieving dynamic shading; AR-HUD: Utilizes DLP projection technology, projecting visible light with a wavelength of 550nm (the wavelength most sensitive to the human eye), supporting separate projection of content in the "shaded area" and "unshaded area" of the smart windshield: The shaded area projects key information such as vehicle speed, road speed limits, and collision warnings (brightness 150cd / m², avoiding superposition with glare); the unshaded area projects navigation paths (brightness 200cd / m², ensuring clear visibility). AR-HUD: Installed behind the dashboard, the projection light path faces the driver's field of vision area of ​​the smart windshield (projection distance 12m, field of view 13°).

[0114] In this embodiment, the core innovations and technical details are as follows: Innovation 1: The collaborative execution mechanism between the smart windshield and AR-HUD. Collaborative logic: When a specific pixel area of ​​the smart windshield is powered on and darkens (forming an "occlusion block"), the AR-HUD projects driving information onto the occlusion block—both physically isolating glare using the occlusion block and compensating for the field of vision information in the occlusion block area through the HUD, thus avoiding "information loss caused by shading".

[0115] Optionally, the central control module calculates the projection position of the light source on the smart windshield (e.g., coordinates P(x=0.3m, y=1.5m), with the lower left corner of the windshield as the origin).

[0116] Optionally, the parameters of the occlusion block (width W=2m, height H=0.3m, transparency α=68%) are generated, and a command is sent to the intelligent windshield drive circuit to control the corresponding pixel unit to be powered on (voltage 3.5V) to form an occlusion block with a light-blocking rate of 68%. The edge 5 pixels (5mm) of the occlusion block adopt "grayscale gradient feathering" - linearly reducing the light-blocking rate from 68% at the center to 10% at the edge to avoid the harsh boundary causing visual fatigue to the driver.

[0117] Optionally, commands can be sent to the AR-HUD simultaneously to project the text and icons "Vehicle speed 60km / h + speed limit ahead 70km / h" onto the obstructed area, with the projection brightness adjusted to 150cd / m² (matching ambient light intensity in real time via a light sensor) to ensure clear information without interfering with the field of vision.

[0118] Key information (brightness 150 cd / m², to avoid superimposed glare); projection navigation path in non-shaded areas (brightness 200 cd / m², to ensure clear visibility). AR-HUD: installed behind the dashboard, with the projection light path facing the driver's field of vision area of ​​the smart windshield (projection distance 12m, field of view 13°).

[0119] In this embodiment, the core innovations and technical details may include: Innovation 1: The collaborative execution mechanism between the smart windshield and AR-HUD. Collaborative logic: When a specific pixel area of ​​the smart windshield is powered on and darkened (forming an "occlusion block"), the AR-HUD projects driving information onto the occlusion block—both physically isolating glare using the occlusion block and compensating for glare through the HUD. Innovation 2: The perception, recognition, and predictive tracking algorithm for glare sources.

[0120] Optionally, the light source recognition algorithm adopts "multi-feature fusion recognition": First, the brightness and color temperature features of the light source are extracted through an HDR camera (excluding low beams with brightness <3000cd / m² and fog lights with color temperature <5000K); then, the light source is confirmed to be a "moving thermal target" through an infrared camera (excluding fixed streetlights); finally, the speed of the light source is detected by a lidar (excluding stationary targets with speed <10km / h), with a comprehensive recognition accuracy >99%; Dynamic tracking and prediction algorithm: Based on the extended Kalman filter (EKF) algorithm, the historical motion data collected by the lidar (such as the speed and acceleration of the past 5 frames) is used to predict the projection position of the light source on the smart windshield within the next 300ms (prediction error <0.05m); for example, when an oncoming vehicle approaches the vehicle at 50km / h, the EKF algorithm predicts that the light source will shift laterally by 0.05m and longitudinally by 0.03m after 300ms, and adjusts the position of the blocking block in advance to ensure that the blocking block always covers the light source and avoids "lagging blocking".

[0121] Optionally, an adaptive occlusion block generation strategy. Occlusion block size is adaptive: Width W: W = (Actual width of the light source × Longitudinal distance between the light source and the vehicle) / 10 (The actual width of the light source is 0.2m by default, based on the size of common sedan high beam headlights); for example: when the light source is 100m away from the vehicle, W = 2m; when the distance is 50m, W = 1m—ensuring the occlusion block size matches the image size of the light source in the field of view, avoiding excessive occlusion; Height H: Fixed at 0.3m, covering the height range of common vehicle high beam headlights (1.1-1.4m, corresponding to the projection height on the smart windshield); Occlusion block transparency is adaptive: Transparency α (occlusion rate): α = (Light source grayscale value / 255) × 80% (Light source grayscale value is collected by HDR camera, range 0-255); For example: when the light source grayscale value is 220 (strong light), α=68%; when the grayscale value is 150 (weak light), α=47% — ensuring that the shading rate matches the glare intensity, avoiding excessive or insufficient shading; the shape of the shading block is adaptive: the default is "ellipse" (horizontal major axis, vertical minor axis) to reduce the shading area on the field of vision; when the light source is detected as "tall vehicles such as SUVs", it is automatically adjusted to "rectangle" (height increased to 0.4m) to ensure coverage of high beams at higher positions.

[0122] Optionally, perception data acquisition (continuous execution, 50ms cycle): the HDR forward-looking camera acquires forward-facing color images, the infrared camera acquires thermal radiation images, and the lidar acquires the three-dimensional coordinates and velocity data of the target in front. The three types of data are transmitted to the central control module via the vehicle Ethernet.

[0123] Optionally, glare source identification (executed per frame, time <20ms): After image preprocessing, the central control module extracts the light source area using the YOLOv8 model, and combines infrared thermal radiation characteristics (thermal radiation signal of vehicle light source) and lidar motion characteristics (three-dimensional spatial coordinates (distance, lateral offset, height) and movement speed of the target in front) to determine whether it is a "harmful high beam"; if not, return to step 1; if yes, proceed.

[0124] Optionally, light source localization and prediction (executed per frame, time <10ms): by fusing camera and radar data through Kalman filtering, the real-time position of the light source in the vehicle coordinate system (x: lateral offset, y: longitudinal distance, z: height) is calculated; then, the projection position of the light source on the smart windshield within the next 300ms is predicted using the EKF algorithm.

[0125] Optionally, the decision-making process for the occlusion block and HUD instructions (time < 10ms): Calculate the occlusion block parameters based on the predicted position: position P(x',y') (x' = x × windshield lateral coefficient, y' = z × windshield longitudinal coefficient, coefficients determined by windshield calibration), width W, height H, and transparency α; simultaneously determine the content (vehicle speed, speed limit) and brightness of the HUD projected onto the occlusion block; Step 5: Perform anti-glare operation (time < 10ms): The central control module sends pixel driving instructions to the intelligent windshield drive circuit to control the corresponding pixels to be powered on to form the occlusion block; simultaneously send projection instructions to the AR-HUD to project information in the occlusion block area.

[0126] Optionally, closed-loop feedback and exit (continuous execution): The perception module continuously monitors the light source status: If the lidar detects that the longitudinal distance of the light source is >100m (the light source is far away), or the camera detects that the light source brightness is <1000cd / m² (oncoming vehicles switch to low beam), the central control module instructs the corresponding pixel of the smart windshield to power off (restoring 90% transmittance), the AR-HUD stops projecting in this area, and returns to step 1; if the light source continues to exist, steps 3-5 are repeated to dynamically update the occlusion block parameters.

[0127] Optionally, the mechanism for generating beneficial effects through the synergistic effect of "partial shading + information compensation" to improve safety and information efficiency may include: the 1mm-level pixel unit of the smart windshield achieves "shading only the glare area," while the non-glare area remains transparent, avoiding overall darkening of the field of vision; AR-HUD projects information in the shading area, compensating for the information gap caused by shading, and the two work together to ensure that the driver is not disturbed by glare and can obtain key driving information in real time, directly reducing the risk of accidents; multi-sensor fusion + prediction algorithm to improve the timeliness and accuracy of protection: HDR camera ensures the accuracy of light source feature extraction, infrared camera eliminates interference from non-vehicle light sources, and LiDAR provides accurate distance and speed data, and the fusion of the three makes the light source recognition accuracy >99%; EKF prediction algorithm adjusts the position of the shading block in advance to avoid "shading lag caused by light source movement", and the protection response time is <100ms; adaptive parameter adjustment to improve scene adaptability: the size, transparency, and shape of the shading block are dynamically adjusted according to the distance, intensity, and vehicle type of the light source, without manual intervention, and can adapt to different road conditions such as urban and rural areas, as well as glare scenarios of different vehicle types such as sedans and SUVs, with better adaptability than existing fixed parameter solutions.

[0128] Table 1 is a component selection and parameter table according to the embodiments of this application, as shown in Table 1:

[0129] Table 1 Component Selection and Parameter Table According to Embodiments of this Application

[0130]

[0131] Example of working scenario and process: Scenario setting: Time can be 22:00 at night, rural road without streetlights (ambient light intensity <10 lux); Vehicle status: sedan (wheelbase 2800mm), traveling at a constant speed of 60km / h, with low beam headlights on; Oncoming target: SUV (wheelbase 2900mm), traveling at a constant speed of 50km / h, with high beam headlights on (brightness 5000cd / m², color temperature 6000K, light group spacing 1.5m), initial distance from the vehicle is 100m, lateral offset from the vehicle's centerline is 0.3m, and high beam headlight height is 1.3m.

[0132] In this embodiment, the perception data acquisition (time T0) includes: the camera acquiring two bright light sources in opposite directions, with an image grayscale value of 230; the infrared camera detecting the light source as a "moving thermal target" (temperature 30℃, consistent with the vehicle's headlight temperature); the lidar measuring the longitudinal distance of the light source as 100m, the lateral offset as 0.3m, the height as 1.3m, and the relative speed as 110km / h (60km / h for this vehicle + 50km / h for the opposite vehicle); the data is transmitted to the chip via the vehicle's Ethernet (delay 40ms). Glare source recognition (time T0 + 40ms): after the chip denoises the camera image, the YOLOv8 model extracts the light source area (detection time 18ms), and combined with "brightness 5000cd / m² > 3000cd / m², color temperature 6000K, headlight spacing 1.5m", it is determined to be "harmful high beam", and the anti-glare process is initiated. Light source positioning and prediction (time T0+58ms): Kalman filter fusion data: calculate the position of the light source in the vehicle coordinate system (x=0.3m, y=100m, z=1.3m); predict the projection position of the light source on the smart windshield 300ms later (T0+358ms) using the EKF algorithm: x'=0.3m×1.2 (windshield lateral calibration coefficient)=0.36m, y'=1.3m×0.6 (windshield longitudinal calibration coefficient)=0.78m (the lower left corner of the windshield is the origin).

[0133] In this embodiment, the decision between the occlusion block and the HUD command (time T0+68ms) is made. The parameters of the occlusion block are calculated as follows: width W = (0.2m×100m) / 10 = 2m (actual width of the light source is 0.2m); height H = 0.4m (the height of the SUV high beam is relatively high, so it is adjusted to 0.4m); transparency α = (230 / 255)×80%≈72%; edge feathering: 5 pixels (5mm), linearly reduced from 72% to 10%; command: project "vehicle speed 60km / h + speed limit ahead 70km / h" (speed limit on rural roads) in the occlusion block area, brightness 150cd / m² (matching ambient light). Perform anti-glare operation (time T0+78ms): Orin-X sends a 3.8V voltage command (corresponding to 72% shading rate) to the pixel unit of the smart windshield with (0.36m, 0.78m) as the center and 2m×0.4m as the center through the driver chip. The pixel is powered on to form a shading block (response time 45ms, shading completed in T0+123ms); at the same time, the chip sends a projection command to the AR-HUD, and the HUD projects information in the shading block area (response time 8ms, projection completed in T0+131ms). Closed-loop feedback and exit (time T0+131ms to T0+2.5s): LiDAR continuously monitors: At T0+1.5s, the oncoming SUV is 55m away from the vehicle, and the LiDAR predicts that it will switch to low beam in 1s; At T0+2.5s, the camera detects that the light source brightness has dropped to 800cd / m² (SUV switches to low beam), the Orin-X command intelligent windshield pixels are powered off (transmittance recovers to 90%, response time 40ms), HUD projection stops, and the process ends.

[0134] In this embodiment, the effectiveness of the implementation is verified, and the safety features include: only the area of ​​the oncoming high beams in the driver's forward field of vision is blocked (2m×0.4m), and the light transmittance of the remaining areas is 90%, with no "night blindness" phenomenon, and the reaction time is shortened from 1.2 seconds in the traditional unprotected mode to 0.3 seconds; information acquisition: the vehicle speed and speed limit information projected by the HUD are clearly visible, and the driver does not need to look down at the instrument panel, improving the information acquisition efficiency by 60%; adaptability: the size of the blocking block automatically adjusts as the SUV approaches (at T0+1.5s, W=(0.2m×55m) / 10=1.1m), always matching the size of the light source imaging, without excessive obstruction.

[0135] Other information that may aid understanding includes the following regarding the pixel driving principle of the intelligent windshield: The intelligent windshield employs a "row-column matrix driving" system, where each row of pixels is controlled by a channel of the chip, and each column of pixels is controlled by a ground terminal; when a pixel at the intersection of a row and column needs to be blocked, the driving chip outputs the target voltage to that row, and the column is grounded, forming an electric field that causes the PDLC liquid crystal molecules to align in an orderly manner, thus achieving light blocking; the driving circuit supports "simultaneous driving of some pixels," ensuring that the blocking block generation speed is <50ms.

[0136] In this embodiment, the coordinate calibration method requires the system to undergo "vehicle coordinate system - intelligent windshield coordinate system" calibration before leaving the factory: a standard calibration board (containing marker points with known coordinates) is set up 10m in front of the vehicle, an HDR camera captures images of the calibration board, and a lidar measures the three-dimensional coordinates of the marker points to establish a mapping relationship between the two (lateral coefficient, longitudinal coefficient), which is stored in the storage unit of the central control module to ensure that the light source position conversion error is <0.05m. Algorithm model training data: The training dataset of the YOLOv8 light source detection model contains 100,000 frames of nighttime driving images, covering different weather conditions (rainy days, foggy days), different light sources (car high beams, SUV high beams, truck high beams), and different ambient light (0-100 lux) scenarios to ensure the model's recognition accuracy in complex scenarios.

[0137] In this embodiment, the key point of the collaborative execution mechanism between the intelligent windshield and AR-HUD is: the PDLC intelligent windshield uses 1mm×1mm independent pixel units. When harmful high beams are detected, the corresponding pixel area is energized to form a local shading block; at the same time, the AR-HUD projects key driving information (vehicle speed, speed limit, navigation) in the shading block area, realizing the synergy of "glare shading and information compensation" - this mechanism is different from the existing "only shading without compensation" or "only HUD adjustment" solutions, and is the core of this application to improve safety and information efficiency.

[0138] In this embodiment, the glare light source perception and prediction algorithm based on multi-sensor fusion has the following key points: combining an HDR forward-looking camera (extracting brightness and color temperature features), an infrared camera (excluding non-vehicle light sources), and a forward-facing LiDAR (acquiring three-dimensional position and velocity), data fusion is achieved through Kalman filtering to accurately locate the light source; then, the extended Kalman filter (EKF) is used to predict the motion trajectory of the light source within the next 300ms, and the position of the occlusion block is adjusted in advance. This algorithm ensures high accuracy (>99%) of glare recognition and high timeliness of occlusion (response time <100ms), which is different from existing single-sensor or no-prediction solutions.

[0139] In this embodiment, the adaptive occlusion block generation strategy dynamically calculates the width of the occlusion block based on "actual width of the light source × distance from the light source", adjusts the transparency linearly based on "light source grayscale value", and adaptively switches the shape based on "vehicle type" (ellipse / rectangle). The edges are also treated with grayscale gradient feathering. This strategy ensures that the occlusion block can effectively block glare while minimizing the impact on the driver's forward vision, which is different from the existing fixed size and fixed occlusion rate schemes.

[0140] In this embodiment, the intelligent windshield material alternative is: Suspended Particle Device (SPD) glass. Technical details: SPD glass achieves light blocking by controlling the arrangement of suspended particles through an electric field. Its response speed (<10ms) is superior to PDLC glass (<50ms), and its light blocking uniformity is higher. However, its cost is 30% higher than PDLC glass, making it suitable for high-end vehicles with higher response speed requirements. Compatibility: SPD glass can directly replace PDLC glass; only the driving voltage range of the central control module needs adjustment (SPD glass driving voltage 0-12V). Other modules (sensing, HUD) do not require modification and can still achieve the invention's purpose.

[0141] In this embodiment, the sensing module is replaced by a millimeter-wave radar instead of a lidar. Technical details: Millimeter-wave radar (detection range 160m, ranging accuracy ±0.5m) is used, costing 50% less than lidar, making it suitable for cost-sensitive economy vehicles. Although the ranging accuracy is slightly lower (±0.5m vs ±0.1m), by optimizing the Kalman filter algorithm (increasing the weight of historical data), the positioning error can be controlled to <0.3m, still meeting the requirements for occlusion block positioning. Compatibility: The signal format of the millimeter-wave radar needs to be adapted to the CANFD protocol of the central control module; other algorithms (recognition, prediction) do not need modification, thus achieving the invention's objective.

[0142] An alternative shape for the occlusion block is a cross-shaped occlusion block. Technical details: The "horizontal bars" of the cross-shaped occlusion block cover the width of the light source, and the "vertical bars" cover the height of the light source, reducing the occlusion area by 20% compared to an ellipse, further reducing field-of-view occlusion. However, the light source contour extraction accuracy of the YOLOv8 model needs to be optimized to ensure that the cross shape accurately covers the core area of ​​the light source. Compatibility: Only the occlusion block shape decision logic of the central control module needs to be modified; the execution module does not require modification, thus achieving the invention's objective.

[0143] According to an embodiment of this application, a vehicle front light source blocking system is also provided. It should be noted that the vehicle front light source blocking system of this embodiment can be used to execute the vehicle front light source blocking method in the embodiments of this application.

[0144] Figure 2 This is a schematic diagram of a vehicle front light source blocking system according to an embodiment of this application. Figure 2 As shown, the vehicle's front light source blocking device 200 may include: a sensing module 202, a central control module 204, and a blocking module 206.

[0145] The perception module 202 is used to acquire image data of the front of the vehicle and motion data of the light source in front of the vehicle during the current time period. The image data includes the light source, and the motion data is used to characterize the motion state of the light source.

[0146] The central control module 204 is used to determine the category of the light source based on image data and motion data; in response to the category being the target category, it predicts the predicted position information of the light source in the future time period based on image data; and based on the predicted position information, it determines the control strategy corresponding to the occlusion unit in the vehicle in the future time period, wherein the control strategy is used to characterize the rules for the occlusion unit to display the occlusion object, and the occlusion object is used to occlude the light source.

[0147] The occlusion module 206 is used to display occlusion content according to the control strategy in order to block the light source in front of the vehicle.

[0148] In this embodiment, the vehicle front light source blocking system can also be called a vehicle high beam directional anti-glare system based on intelligent windshield and collaborative perception. It consists of three core modules: a perception module, a central control module, and a human-machine interaction execution module (which may include the blocking module). The modules interact with each other through in-vehicle Ethernet and a Controller Area Network with Flexible Data-rate (CANFD) bus.

[0149] In this embodiment, the perception module acquires image data of the area in front of the vehicle and motion data of the light source in front of the vehicle during the current time period. The image data includes the light source, and the motion data represents the motion state of the light source. The central control module determines the category of the light source based on the image data and motion data. In response to the category being the target category, the predicted position information of the light source in the future time period is predicted based on the image data. Based on the predicted position information, the control strategy corresponding to the occlusion unit in the vehicle in the future time period is determined. The control strategy represents the rules for the occlusion unit to display the occlusion object, and the occlusion object is used to occlude the light source. The occlusion unit displays the occlusion content according to the control strategy to occlude the light source in front of the vehicle, thereby achieving the technical effect of improving the headlight glare protection effect and solving the technical problem of poor headlight glare protection effect.

[0150] According to an embodiment of this application, a device for blocking a vehicle's front light source is also provided. It should be noted that the device for blocking a vehicle's front light source in this embodiment can be used to execute the method for blocking a vehicle's front light source in the embodiments of this application.

[0151] Figure 3 This is a schematic diagram of a vehicle front light source blocking device according to an embodiment of this application. Figure 3As shown, the vehicle's front light source blocking device 300 may include: an acquisition unit 302, a first determination unit 304, a prediction unit 306, a second determination unit 308, and a control unit 310.

[0152] The acquisition unit 302 is used to acquire image data of the front of the vehicle and motion data of the light source in front of the vehicle during the current time period. The image data includes the light source, and the motion data is used to characterize the motion state of the light source.

[0153] The first determining unit 304 is used to determine the type of light source based on image data and motion data.

[0154] The prediction unit 306 is used to predict the predicted location information of the light source in a future time period based on image data, in response to the category being the target category, wherein the light source of the target category affects the driving safety of the vehicle, and the future time period is later than the current time period.

[0155] The second determining unit 308 is used to determine the control strategy corresponding to the occlusion unit in the vehicle in a future time period based on the predicted location information. The control strategy is used to characterize the rules for the occlusion unit to display the occlusion object, and the occlusion object is used to occlude the light source.

[0156] The control unit 310 is used to control the occlusion unit to display the occlusion object according to the control strategy, so as to block the light source in front of the vehicle.

[0157] The vehicle front light source blocking device of this embodiment acquires image data and light source motion data in real time through an acquisition unit. A first determination unit determines the category of the light source based on this data. A prediction unit predicts the future position of the light source belonging to the target category. A second determination unit determines the control strategy of the blocking unit based on the predicted position information. Finally, the control unit executes the blocking operation according to the control strategy, thereby achieving the technical effect of improving the light glare protection effect and solving the technical problem of poor light glare protection effect.

[0158] According to an embodiment of this application, a computer-readable storage medium is also provided, the storage medium including a stored program, wherein the program executes the methods described in the embodiments of this application.

[0159] According to an embodiment of this application, a processor is also provided for running a program, wherein the program executes the methods described in the embodiments of this application during runtime.

[0160] According to another aspect of the embodiments of this application, an electronic device is also provided. The electronic device includes a memory and a processor, wherein the memory stores a computer program, and the processor is configured to run the computer program to perform the methods described in the embodiments of this application.

[0161] According to another aspect of the embodiments of this application, a computer program product is also provided. This computer program product includes a computer program that, when executed by a processor, implements the methods described in the embodiments of this application.

[0162] According to another aspect of the embodiments of this application, a vehicle is also provided. The vehicle includes a memory and a processor. The memory stores an executable program; the processor is used to run the program, which, when running, implements the methods described in the embodiments of this application.

[0163] In the above embodiments of this application, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0164] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. The device embodiments described above are merely illustrative; for example, the division of units can be a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the displayed or discussed mutual couplings, direct couplings, or communication connections may be through some interfaces; indirect couplings or communication connections between units or modules may be electrical or other forms.

[0165] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0166] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0167] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as a USB flash drive, read-only memory (ROM), random access memory (RAM), portable hard drive, magnetic disk, or optical disk.

[0168] The above are merely preferred embodiments of this application. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method of obscuring a light source in front of a vehicle, characterized by, The method comprises: acquiring image data of a front of a vehicle and motion data of a light source in front of the vehicle in a current period, wherein image content of the image data comprises the light source, and the motion data is used to represent a motion state of the light source; determining a category of the light source based on the image data and the motion data; in response to the category being a target category, predicting, based on the image data, prediction position information of the light source in a future period, wherein the target category of the light source affects driving safety of the vehicle, and the future period is later than the current period; determining, based on the prediction position information, a control strategy corresponding to a shielding unit in the vehicle in the future period, wherein the control strategy is used to represent a rule of the shielding unit displaying a shielding object, and the shielding object is used to shield the light source; controlling the shielding unit to display the shielding object according to the control strategy, so as to shield the light source in front of the vehicle.

2. The method of claim 1, wherein, The determination of the category of the light source based on the image data and the motion data comprises: extracting features of a color image in the image data to obtain a brightness feature of the light source and a color temperature feature of the light source; identifying a thermal radiation image in the image data to obtain a movement characteristic of the light source, wherein the movement characteristic is used to represent whether the light source is a moving thermal target; in response to the brightness feature satisfying a brightness condition, the color temperature feature satisfying a color temperature condition, the movement characteristic being used to represent that the light source is the moving thermal target, and the motion data satisfying a motion condition, determining that the category is the target category.

3. The method of claim 1, wherein, The prediction of the prediction position information of the light source in the future period based on the image data in response to the category being the target category comprises: in response to the category being the target category, identifying the image data to obtain current position information of the light source in the current period; predicting the current position information to obtain the prediction position information.

4. The method of claim 3, wherein, The identification of the image data to obtain the current position information of the light source in the current period in response to the category being the target category comprises: in response to the category being the target category, identifying a color image in the image data to obtain two-dimensional coordinate information of the light source in the current period; performing fusion processing on three-dimensional coordinate information in the motion data and the two-dimensional coordinate information to obtain the current position information, wherein the fusion coordinate information is used to represent a position of the vehicle in a vehicle coordinate system.

5. The method of claim 1, wherein, The determination of the control strategy corresponding to the shielding unit in the vehicle in the future period based on the prediction position information comprises: converting the prediction position information to obtain a projection position of the light source in the shielding unit; determining a shielding parameter based on the projection position, wherein the shielding parameter is used to represent a size and light transmission characteristic of the shielding object; converting the shielding parameter to obtain the control strategy.

6. The method of claim 5, wherein, The method further comprises: determine a projection parameter matching the occlusion parameter, wherein the projection parameter is used to represent the content to be displayed in the occlusion region.

7. A system for obscuring a light source in front of a vehicle, characterized by Comprising: an awareness module, configured to acquire image data in front of a vehicle and motion data of a light source in front of the vehicle in a current period, wherein image content of the image data comprises the light source, and the motion data is used to represent a motion state of the light source; a central control module, configured to determine a category of the light source based on the image data and the motion data, predict, in response to the category being a target category, predicted position information of the light source in a future period based on the image data, and determine a control strategy corresponding to an occlusion unit in the vehicle in the future period based on the predicted position information, wherein the control strategy is used to represent a rule of the occlusion unit displaying an occlusion object for occluding the light source; the occlusion module, configured to display the occlusion content according to the control strategy to occlude the light source in front of the vehicle.

8. A device for shielding a light source in front of a vehicle, characterized in that Comprising: an acquisition unit, configured to acquire image data in front of a vehicle and motion data of a light source in front of the vehicle in a current period, wherein image content of the image data comprises the light source, and the motion data is used to represent a motion state of the light source; a first determination unit, configured to determine a category of the light source based on the image data and the motion data; a prediction unit, configured to predict, in response to the category being a target category, predicted position information of the light source in a future period based on the image data, wherein the target category of the light source affects driving safety of the vehicle, and the future period is later than the current period; a second determination unit, configured to determine a control strategy corresponding to an occlusion unit in the vehicle in the future period based on the predicted position information, wherein the control strategy is used to represent a rule of the occlusion unit displaying an occlusion object for occluding the light source; a control unit, configured to control the occlusion unit to display the occlusion object according to the control strategy to occlude the light source in front of the vehicle.

9. A vehicle characterized by comprising: A computer program product for performing the method of any one of claims 1 to 6.

10. A computer program product, characterised in that, A computer program product comprising computer instructions which, when executed by a processor, implement the method of any one of claims 1 to 6.

Images