Anti-dazzle automobile headlight control method and control device

Abstract

The present application relates to the field of automobile lighting technology, and specifically provides a control method and device for anti-glare automobile headlamp, which comprises obtaining image data and point cloud data corresponding to the current field of view by using a camera and a laser radar, identifying target features existing in the field of view according to the image data and the point cloud data, and re-projecting the target position to the automobile headlamp coordinate system according to the target features; and reducing the light intensity of the region belonging to the target position in the automobile headlamp coordinate system. The present application can accurately match the vehicle and pedestrian profiles, and dim the light source of the corresponding region, thereby reserving most of the high beam field of view for the driver while avoiding glare interference to others.

Description

Technical Field

[0001] This invention relates to the field of automotive lighting technology, specifically providing an anti-glare automotive headlight control method and control device. Background Technology

[0002] When driving at night, traditional high beams can severely dazzle drivers of oncoming or ahead vehicles, causing them to momentarily lose clear vision and easily leading to driving accidents. This year, advancements in machine vision, complex sensors, and high-precision controllable light source technology have made intelligent anti-glare technology possible. Therefore, how to avoid glare when using high beams has become an important technical challenge.

[0003] Currently, a common method for preventing glare is to use forward-facing sensors to identify the position and distance of vehicles or pedestrians ahead, and then have the controller calculate and turn off the headlights in the corresponding area. For example, existing matrix LED headlights mainly create anti-glare dark zones by turning off some LEDs, rather than dimming the lights, which can affect the driver's visibility.

[0004] Matrix LEDs can be viewed as a "switching" or "coarse-segmentation" anti-glare technology. Their basic control unit (LED beads) typically can only be fully turned on or off, unable to achieve linear brightness adjustment. On one hand, matrix LEDs usually have only 12-10 independent control zones, forming only coarse "dark areas" to block oncoming or same-direction vehicles. The edges of these dark areas are jagged or irregular, unable to precisely match the vehicle's outline. On the other hand, completely turning off the LED beads will cause the area to lose all illumination, creating a dark zone that may obscure potential hazards (such as pedestrians or obstacles). Furthermore, the frequent switching of LED beads and the abrupt transitions are not conducive to human visual adaptation, easily leading to driver fatigue over prolonged use. In complex scenarios, such as multiple closely adjacent vehicles, overlapping pedestrians and vehicles, or rapidly changing vehicle positions on curves, the system may frequently switch large zones on and off, causing the headlights to flicker and jump, resulting in poor stability and even an inability to make optimal blocking decisions. Summary of the Invention

[0005] To address the aforementioned problems, this invention provides an anti-glare automotive headlight control method, or control device, which can precisely match the outlines of vehicles and pedestrians and dim the light sources in the corresponding areas, thus preserving most of the high beam visibility for the driver while avoiding glare that may disturb others.

[0006] The anti-glare automotive headlight control method provided by this invention includes: S1: Use a camera and lidar to acquire image data and point cloud data corresponding to the current field of view; S2: Identify target features in the field of view based on the image data and point cloud data, or based solely on the point cloud data; S3: Based on the target features, reproject the target position to the car headlight coordinate system; reduce the light intensity of the area belonging to the target position in the car headlight coordinate system.

[0007] Preferably, the target features include target type, target size, target distance, and target radial velocity.

[0008] Preferably, deep learning algorithms are used to identify target features, including detection algorithms and geometric segmentation algorithms.

[0009] Preferably, the method for reprojecting the target position to the car headlight coordinate system is as follows: first, reproject the point cloud data to the camera coordinate system, and then reproject the 3D coordinates in the camera coordinate system to the car headlight coordinate system.

[0010] Preferably, the expression for reprojecting point cloud data onto the camera coordinate system is: ; in, This represents a point in the lidar coordinate system. Represents a point in the camera coordinate system. This represents the rotation matrix from the lidar to the camera. This represents the translation vector from the lidar to the camera.

[0011] Preferably, the expression for reprojecting the 3D coordinates in the camera coordinate system to the car headlight coordinate system is: ; in, This represents the intrinsic parameter matrix of the camera. This indicates that the target contour is represented by points in a pixel coordinate system. This represents the Z-axis coordinate of a point in the camera coordinate system.

[0012] Preferably, the expression for the region belonging to the target position in the automotive headlight coordinate system is: ; in, This represents the reprojection of a 3D point in the camera coordinate system onto a point in the car headlight coordinate system. This represents the rotation matrix from the camera to the car headlights. This represents the translation vector from the camera to the car headlights.

[0013] An anti-glare automotive headlight control device for implementing an anti-glare automotive headlight control method includes: a lidar, a camera, a headlight, and a control module, wherein the lidar, camera, and headlight are respectively connected to the control module; The control module controls the lidar to collect point cloud data; the control module controls the camera to collect image data; the control module analyzes the image data and depth data collected by the camera and lidar and transmits the results to the headlight module; the control module then controls the headlights to adjust the light intensity of local areas according to the results.

[0014] Preferably, the control module includes a development board driver, which is an NVIDIA Jetson Orin NX development board.

[0015] Preferably, the development board driver triggers the camera and LiDAR to acquire image data and point cloud data through hardware triggering or software triggering.

[0016] Compared with the prior art, the present invention can achieve the following beneficial effects: From a hardware perspective, this invention provides a complete hardware system for anti-glare vehicle lights, which can more accurately adjust the projection range and brightness of the lights.

[0017] From a software perspective, this invention identifies the area corresponding to a pedestrian or vehicle within the car's headlights and then adjusts the local light intensity, rather than directly turning off the high beams. This method fully considers the needs of drivers, pedestrians, and vehicles, avoiding glare for pedestrians and vehicles while maintaining the driver's visibility. Based on real-world scenarios, it provides a more flexible and suitable adjustment strategy for car headlights, offering strong targeting and better matching accuracy. Attached Figure Description

[0018] Figure 1 This is a flowchart of an anti-glare automotive headlight control method provided in an embodiment of the present invention; Figure 2 This is a layout diagram of an anti-glare automotive headlight control device provided according to an embodiment of the present invention; Figure 3 This is a schematic diagram of an anti-glare automotive headlight control device provided according to an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only for explaining the invention and do not constitute a limitation thereof. Similar elements in different embodiments are referred to by associated similar element reference numerals. In the following embodiments, many details are described to facilitate a better understanding of the invention. However, those skilled in the art will readily recognize that some features may be omitted in different situations, or may be replaced by other elements, materials, or methods. In some cases, some operations related to the invention are not shown or described in the specification. This is to avoid obscuring the core parts of the invention with excessive description. For those skilled in the art, detailed description of these related operations is not necessary; they can fully understand the related operations based on the description in the specification and general technical knowledge in the art.

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in this invention can be combined to form various implementations. Furthermore, the order of the steps or actions in the method description can be changed or adjusted in a manner readily apparent to those skilled in the art. Therefore, the various orders in the specification and drawings are merely for the clear description of a particular embodiment and do not imply a mandatory order, unless otherwise stated that a particular order must be followed.

[0021] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," and "counterclockwise," etc., indicating orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation on this invention. Furthermore, the terms "first," "second," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, features defined with "first," "second," etc., may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0022] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art will understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0023] The invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0024] like Figure 1 As shown, the present invention provides a method for controlling anti-glare automotive headlights, as detailed below: S1: Use a camera and lidar to acquire image data and point cloud data corresponding to the current field of view.

[0025] When a car is driving at night, pedestrians or vehicles within the driver's field of vision in the direction of travel will be affected by the headlights. To minimize the impact of headlights, it is necessary to first identify the location of any objects, such as pedestrians or vehicles, within the current field of vision. Cameras are then fixedly installed on the car to collect images of the field of vision; one or more cameras can be used. The camera installation location can be adjusted as needed to ensure comprehensive acquisition of image data and point cloud data corresponding to the current field of vision. For example, cameras can be installed above the windshield, near the rearview mirrors, or around the vehicle.

[0026] Furthermore, since different targets within the vehicle's field of view are at different distances (i.e., different scales) from the vehicle, this invention also incorporates a lidar system on the vehicle to collect point cloud data. A lidar is a sensor that uses laser beams to measure distance. It emits laser pulses into the surrounding environment and measures the time it takes for these pulses to reflect back, thereby calculating the distance to surrounding objects. Lidar can quickly scan the space around the vehicle, generating a large amount of three-dimensional point cloud data. Each point cloud data point contains its coordinate information in three-dimensional space. These points combined form a three-dimensional model of the vehicle's surrounding environment, accurately describing the shape, position, and distance of targets within the field of view. The camera and lidar can be integrated or discrete; their combination can obtain image data and point cloud data of targets such as pedestrians and vehicles within the vehicle's field of view.

[0027] S2: Identify target features in the field of view based on image data and point cloud data, or only based on point cloud data.

[0028] Image data and point cloud data are acquired through different methods. Therefore, the target features identified by the two may not correspond; that is, target features identified in the image data may not be identified in the point cloud data, or vice versa. Therefore, this invention provides multiple target feature identification methods. After acquiring image data and point cloud data, algorithms or models are needed to analyze and process them separately to identify target features present in the field of view. This invention employs a target detection algorithm to identify target features in image data and a target detection and tracking algorithm to identify target features in point cloud data. The target detection algorithm identifies oncoming or same-direction vehicles and pedestrians from the image data acquired by the camera. The geometric segmentation algorithm segments the target contours identified by the target detection algorithm. The target detection and tracking algorithm directly obtains the coordinate information of the target's location and its radial velocity from the point cloud data. The geometric segmentation algorithm then segments the target features identified in the image data and point cloud data.

[0029] For image data, object detection algorithms are used for processing. These algorithms can be based on deep learning models, such as convolutional neural networks (CNNs), to classify and locate objects in images. For example, a model can be trained to identify common targets such as pedestrians, vehicles, and traffic signs, or existing models like the YOLO series or R-CNN series can be used. During the recognition process, the object detection algorithm analyzes each pixel in the image data, extracting object features such as edges, textures, and colors, and matching these features with a pre-trained model to determine the type and location of the target in the image data. For point cloud data, object detection and object tracking algorithms are used. Point cloud data has three-dimensional spatial information, thus allowing for a more accurate description of the shape and location of objects. Geometric segmentation algorithms can divide the identified target features in image and point cloud data into different regions, each corresponding to a potential object, forming candidate regions. Geometric features of the object, such as size, shape, and orientation, are extracted from these candidate regions to further determine the type and location of the target.

[0030] By integrating the recognition results from image data and point cloud data, target features of objects within the vehicle's field of view can be obtained, including target type, target size, target distance, and target radial velocity. During integration, if the target features identified in the image data and point cloud data belong to the same target, an overlap determination is performed. This determination confirms that the target features in the image data and point cloud data belong to the same target, preventing the identification of the same target feature as two separate targets. An overlap threshold is set for the overlap determination; if the overlapping area reaches the threshold, it is determined to be a target feature of the same target, and this feature is recognized as the target recognition result. When the target feature is obtained solely from point cloud data, it is still considered the target recognition result. However, target features obtained solely from image data are only used to assist in the overlap determination and are not considered as the target recognition result.

[0031] S3: Reproject the target location onto the car headlight coordinate system based on the target features. Reduce the light intensity of the area belonging to the target location in the car headlight coordinate system.

[0032] To achieve anti-glare functionality while preventing the headlights from affecting the driver, this invention reduces the light intensity of the target area within the vehicle's forward field of vision. This ensures that the glare does not obstruct pedestrians or other vehicles outside the vehicle, while maintaining the driver's visibility. To achieve this, the position of the target within the vehicle's field of vision within the vehicle's headlight coordinate system needs to be determined. This invention obtains the corresponding target position coordinates based on target characteristics and reprojects these coordinates onto the vehicle's headlight coordinate system. The specific reprojection process is as follows: First, the point cloud data (i.e., depth information) measured by the LiDAR is reprojected onto the camera coordinate system. The transformation expression is: , in, This represents a point in the lidar coordinate system. Represents a point in the camera coordinate system. This represents the rotation matrix from the lidar to the camera. This represents the translation vector from the lidar to the camera.

[0033] Through the above transformation, 3D points in the camera coordinate system are obtained, and then these 3D coordinates are reprojected into the vehicle headlight coordinate system. The target contour segmented by the geometric segmentation algorithm is a point in the pixel coordinate system. It requires the use of the camera's intrinsic parameter matrix. The transformation to the camera coordinate system is as follows: .

[0034] This allows us to obtain the 3D points of the target contour in the camera coordinate system. Finally, we can use the rotation matrix from the camera to the headlights... Translation vector from camera to headlights The target contour is reprojected from 3D points in the camera coordinate system to points in the headlight coordinate system. This allows you to control the dimming of the light intensity in the corresponding area of ​​the headlights. The conversion formula for reprojection is as follows: .

[0035] After reprojection, the target is projected from the camera coordinate system to the car headlight coordinate system, and the area on the car headlight corresponding to the target is darkened.

[0036] To achieve the above method, this invention also provides an anti-glare automotive headlight control device, including a LiDAR module, a camera module, a control module, and a headlight module. The LiDAR module, camera module, and headlight module are all connected to the control module. The LiDAR module includes a LiDAR sensor, the camera module includes a camera, the control module includes a development board driver, and the headlight module includes headlights, i.e., automotive headlights. The control module mainly operates in three aspects: first, driving the camera and LiDAR; second, analyzing and processing the image and depth data acquired from the camera and LiDAR and transmitting the results to the headlight module; and third, controlling the headlight module to darken the corresponding area based on the data processing results. Specifically, the control module controls the camera module to take pictures and acquire image data, and controls the LiDAR module to scan and acquire point cloud data (depth data) of the current field of view. Target features are identified based on the image data and point cloud data. Target features include target type, target size, target distance, and target radial velocity. The identification algorithm includes, but is not limited to, target detection algorithms, geometric segmentation algorithms, etc. The headlights darken the corresponding area based on the algorithm results.

[0037] It should be noted that the development board driver used in this embodiment of the invention is the NVIDIA Jetson Orin NX development board, which triggers the camera and LiDAR to acquire image data and point cloud data through hardware or software triggering. Addressing the shortcomings of matrix LED resolution and the "on / off" strategy of matrix LEDs, this embodiment of the invention replaces them with high-resolution pixelated LED technology and Digital Light Processing (DLP) technology as the headlight source for automobiles.

[0038] Although embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those skilled in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

[0039] The specific embodiments of the present invention described above do not constitute a limitation on the scope of protection of the present invention. Any other corresponding changes and modifications made in accordance with the technical concept of the present invention should be included within the scope of protection of the claims of the present invention.

Claims

1. A control method for anti-dazzle vehicle headlamps, characterized in that, include: S1: Use a camera and lidar to acquire image data and point cloud data corresponding to the current field of view; S2: Identify target features in the field of view based on the image data and point cloud data, or based solely on the point cloud data; S3: Based on the target features, reproject the target position to the car headlight coordinate system; reduce the light intensity of the area belonging to the target position in the car headlight coordinate system.

2. The anti-dazzle vehicle headlamp control method according to claim 1, characterized in that, The target features include target type, target size, target distance, and target radial velocity.

3. The anti-glare automotive headlight control method according to claim 1, characterized in that, The target features are identified using deep learning algorithms, which include target detection algorithms and geometric segmentation algorithms.

4. The anti-glare automotive headlight control method according to claim 1, characterized in that, The method for reprojecting the target position to the car headlight coordinate system is as follows: first, reproject the point cloud data to the camera coordinate system, and then reproject the 3D coordinates in the camera coordinate system to the car headlight coordinate system.

5. The anti-glare automotive headlight control method according to claim 4, characterized in that, The expression for reprojecting the point cloud data onto the camera coordinate system is: ； in, This represents a point in the lidar coordinate system. Represents a point in the camera coordinate system. This represents the rotation matrix from the lidar to the camera. This represents the translation vector from the lidar to the camera.

6. The anti-glare automotive headlight control method according to claim 4, characterized in that, The expression for reprojecting 3D coordinates from the camera coordinate system to the car headlight coordinate system is: ； in, This represents the intrinsic parameter matrix of the camera. This indicates that the target contour is represented by points in a pixel coordinate system. This represents the Z-axis coordinate of a point in the camera coordinate system.

7. The anti-glare automotive headlight control method according to claim 4, characterized in that, The expression for the region belonging to the target position in the car headlight coordinate system is: ； in, This represents the reprojection of a 3D point in the camera coordinate system onto a point in the car headlight coordinate system. This represents the rotation matrix from the camera to the car headlights. This represents the translation vector from the camera to the car headlights.

8. An anti-glare automotive headlight control device, characterized in that, The method for controlling anti-glare automotive headlights as described in any one of claims 1 to 7 includes: a lidar, a camera, a headlight, and a control module, wherein the lidar, camera, and headlight are respectively connected to the control module; The control module controls the lidar to acquire point cloud data; the control module controls the camera to acquire image data; the control module analyzes the image data and depth data acquired by the camera and lidar and transmits the results to the headlight module; the control module controls the headlights to adjust the light intensity of local areas.

9. The anti-glare automotive headlight control device according to claim 8, characterized in that, The control module includes a development board driver, which is an NVIDIA Jetson Orin NX development board.

10. The anti-glare automotive headlight control device according to claim 9, characterized in that, The development board driver triggers the camera and lidar to acquire image data and point cloud data through hardware or software triggering.

Images