Vehicle light control method and device, electronic equipment and readable storage medium

Abstract

The application relates to the technical field of vehicle light, and provides a vehicle light control method and device, electronic equipment and a readable storage medium. The method comprises the following steps: when an exit feature or an entrance feature of a target scene is identified on a driving section through a front-view camera and a 4D millimeter wave radar; determining a first confidence and first distance data by using image data obtained by the front-view camera; determining a second confidence and second distance data by using three-dimensional point cloud data obtained by the 4D millimeter wave radar; fusing the first distance data and the second distance data according to the first confidence and the second confidence to determine a target distance between the vehicle and the exit feature or the entrance feature; and if the target distance meets a preset vehicle light opening condition, opening the specified light of the vehicle. The application can intelligently control the vehicle light when entering and exiting the target scene, reduce traffic accidents caused by light abrupt change at the entrance and exit of the target scene, and improve driving safety.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle light, and particularly relates to a vehicle light control method and device, electronic equipment and a readable storage medium. BACKGROUND

[0002] The automatic light control technology of a vehicle is generally realized by installing a photosensitive control system in the vehicle. The system relies on electronic components such as photosensitive resistors to perceive changes in ambient light and sends corresponding electronic signals to an electronic control unit (ECU). According to these signals, the ECU can intelligently control the switching of the headlight and the switching of the high beam and low beam. When the vehicle encounters a sudden change in light during driving, the light system will automatically start to provide sufficient illumination. Conversely, when the ambient light becomes bright enough, the ECU will issue an instruction to turn off the light to save energy.

[0003] However, in some specific scenarios such as entering or exiting a tunnel or an underground parking lot, the automatic light system needs a certain amount of time to adapt to the change in light. During this period, the driver may feel visual discomfort due to the rapid change in ambient light, and even a short visual blind area may occur, thereby increasing the risk of driving. In order to solve this problem, the prior art usually equips a camera to collect image data of the road ahead, and then analyzes whether there is an entrance and exit of a special scene. However, this method is not ideal in adverse weather conditions such as dust, snow, fog, heavy rain, night and air pollution. SUMMARY

[0004] Therefore, the embodiments of the present application provide a vehicle light control method and device, electronic equipment and a readable storage medium to solve the problem that the vehicle light in the prior art cannot be intelligently controlled in some special scenarios.

[0005] In a first aspect, the embodiments of the present application provide a vehicle light control method, comprising:

[0006] The road section characteristics in front of the vehicle are identified by the front-view camera and the 4D millimeter wave radar respectively to determine whether the driving road section contains exit characteristics or entrance characteristics of the target scene; if yes, the first confidence of the exit characteristics or the entrance characteristics is determined by using the image data obtained by the front-view camera, and the first distance data of the vehicle from the exit characteristics or the entrance characteristics is calculated by using the pinhole imaging algorithm; the second confidence of the exit characteristics or the entrance characteristics is determined by using the three-dimensional point cloud data obtained by the 4D millimeter wave radar, and the second distance data of the vehicle from the exit characteristics or the entrance characteristics is calculated by using the transmitting and receiving signals of the 4D millimeter wave radar; the first distance data and the second distance data are fused according to the first confidence and the second confidence to determine the target distance of the vehicle from the exit characteristics or the entrance characteristics; if the target distance meets the preset vehicle light turning-on condition, the specified light of the vehicle is turned on.

[0007] In a second aspect, the embodiment of the present application provides a vehicle light control device, which comprises: an identification module configured to identify road section characteristics in front of a vehicle by using a front-view camera and a 4D millimeter wave radar respectively to determine whether the driving road section contains exit characteristics or entrance characteristics of a target scene; a camera module configured to, if yes, determine the first confidence of the exit characteristics or the entrance characteristics by using image data obtained by the front-view camera, and calculate the first distance data of the vehicle from the exit characteristics or the entrance characteristics by using a pinhole imaging algorithm; a radar module configured to determine the second confidence of the exit characteristics or the entrance characteristics by using three-dimensional point cloud data obtained by the 4D millimeter wave radar, and calculate the second distance data of the vehicle from the exit characteristics or the entrance characteristics by using transmitting and receiving signals of the 4D millimeter wave radar; a fusion module configured to fuse the first distance data and the second distance data according to the first confidence and the second confidence to determine the target distance of the vehicle from the exit characteristics or the entrance characteristics; and a light control module configured to turn on the specified light of the vehicle if the target distance meets the preset vehicle light turning-on condition.

[0008] In a third aspect, the embodiment of the present application provides an electronic device, which comprises a memory, a processor, and a computer program stored in the memory and executable on the processor, and the processor implements the steps of the above method when executing the computer program.

[0009] In a fourth aspect, the embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program implements the steps of the above method when executed by a processor.

[0010] Compared with the prior art, the embodiment of the present application has at least the following beneficial effects:

[0011] The road section characteristics in front of the vehicle are identified by the front-view camera and the 4D millimeter wave radar respectively, and it is determined whether the driving road section contains exit characteristics or entrance characteristics of the target scene; if yes, the first confidence of the exit characteristics or the entrance characteristics is determined by using the image data obtained by the front-view camera, and the first distance data of the vehicle from the exit characteristics or the entrance characteristics is calculated by using the pinhole imaging algorithm; the second confidence of the exit characteristics or the entrance characteristics is determined by using the three-dimensional point cloud data obtained by the 4D millimeter wave radar, and the second distance data of the vehicle from the exit characteristics or the entrance characteristics is calculated by the transmitting and receiving signals of the 4D millimeter wave radar; the target distance of the vehicle from the exit characteristics or the entrance characteristics is determined by fusing the first distance data and the second distance data according to the first confidence and the second confidence; and if the target distance meets the preset vehicle light opening condition, the specified light of the vehicle is turned on. Since the front-view camera and the 4D millimeter wave radar configured by the vehicle are used to obtain data in the application, the problems of high update cost, untimely update and inability to cover all roads caused by the traditional recognition method relying on high-precision maps and only relying on cameras can be avoided. At the same time, the road section characteristics in front of the road are perceived, and the target distance of the vehicle from the exit or entrance characteristics is calculated according to the perception result, so that the vehicle light can be intelligently controlled at the exit or entrance of the target scene, the traffic accidents caused by light abrupt change at the exit or entrance of the target scene are reduced, and the driving safety is improved, and the intelligent level is improved. BRIEF DESCRIPTION OF DRAWINGS

[0012] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0013] Figure 1 is a flowchart of a vehicle light control method provided by an embodiment of the present application;

[0014] Figure 2 is a schematic diagram of a vehicle distance determination method provided by an embodiment of the present application;

[0015] Figure 3 is a schematic diagram of a 4D millimeter wave radar data processing flow provided by an embodiment of the present application;

[0016] Figure 4 is a hardware architecture diagram of a vehicle light control method provided by an embodiment of the present application;

[0017] Figure 5 is a structural schematic diagram of a vehicle light control device provided by an embodiment of the present application;

[0018] Figure 6 is a structural schematic diagram of an electronic device provided by an embodiment of the present application. DETAILED DESCRIPTION

[0019] In the following description, specific details are set forth, such as particular system configurations, techniques, etc., in order to provide a thorough understanding of embodiments of the present application. However, persons having ordinary skill in the art will readily understand that embodiments of the present application can be practiced without these specific details. In other instances, well-known structures, devices, circuits, and processes have not been described in detail in order to avoid obscuring the description of embodiments of the present application.

[0020] As used herein, the term "includes" and its variants are to be read to be analogous to "comprises," or "comprising." The term "based on" is to be read as "based, at least in part, on." The term "one embodiment" means "at least one embodiment." The term "another embodiment" means "at least one additional embodiment." The term "some embodiments" means "at least some embodiments." Related definitions are given throughout the description. It should be noted that the use of "first," "second," etc., in the present application does not connote any order, quantity, or importance, but rather are used to distinguish one element from another, and are more especially used to distinguish one entity or action from another entity or action, respectively.

[0021] It should be noted that the use of "one" or "the" in the present application does not exclude that more than one of the identified unit can be present. It should be noted that the use of "a" or "an" herein does not exclude a plurality of said elements or units. It is further noted that a specific value of a parameter can be used in a specific embodiment, but other values of the same parameter can be used in other embodiments.

[0022] It should be noted that the new energy vehicle in the embodiments of the present application refers to a vehicle that uses new energy (non-traditional oil and diesel energy) and has advanced technology. These vehicles use new power systems, which can effectively reduce vehicle emissions, reduce environmental impact, and improve energy efficiency. The new energy vehicle of the embodiments of the present application includes, but is not limited to, the following types of vehicles: electric vehicles (EV), pure electric vehicles (BEV), fuel cell electric vehicles (FCEV), plug-in hybrid electric vehicles (PHEV), and hybrid electric vehicles (HEV), etc.

[0023] A vehicle light control method, device, electronic device, and readable storage medium according to embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0024] Figure 1 is a flowchart of a vehicle light control method provided by an embodiment of the present application. Figure 1 The vehicle light control method of can be executed by a vehicle controller of a new energy vehicle.

[0025] AsFigure 1 The vehicle light control method can include:

[0026] S101, identifying road section features in front of the vehicle by the front-view camera and the 4D millimeter wave radar respectively, and determining whether the driving road section contains exit features or entrance features of the target scene;

[0027] S102, if yes, determining a first confidence of the exit features or the entrance features by using image data obtained by the front-view camera, and calculating first distance data of the vehicle from the exit features or the entrance features by using a pinhole imaging algorithm;

[0028] S103, determining a second confidence of the exit features or the entrance features by using three-dimensional point cloud data obtained by the 4D millimeter wave radar, and calculating second distance data of the vehicle from the exit features or the entrance features by using a transmitting and receiving signal of the 4D millimeter wave radar;

[0029] S104, fusing the first distance data and the second distance data according to the first confidence and the second confidence, and determining a target distance of the vehicle from the exit features or the entrance features;

[0030] S105, if the target distance meets a preset vehicle light opening condition, turning on a specified light of the vehicle.

[0031] In the embodiments of the present application, the front-view camera and the 4D millimeter wave radar configured in the vehicle can identify the scene in front of the vehicle in real time, and determine whether there are road section features in the front road under the target scene (such as entrance features or exit features of the tunnel scene, or entrance features and exit features of the underground parking lot) according to the identification result. The present application can avoid the problems of high update cost, untimely update and inability to cover all roads caused by the traditional recognition method relying on high-precision maps and only relying on cameras. At the same time, since it does not rely on high-precision maps, the method of the present application embodiment can more flexibly adapt to various road scenes, and improve the intelligence and adaptability of vehicle automatic driving.

[0032] In some examples, if it is determined that the front road has an exit feature or an entry feature in a target scene, the vehicle will use the front-view camera and the 4D millimeter-wave radar for more accurate data acquisition and processing. The front-view camera can acquire high-definition image data, determine a first confidence of the exit or entry feature through image recognition technology, and calculate a first distance data of the vehicle and the exit or entry feature using a pinhole imaging algorithm. At the same time, the 4D millimeter-wave radar acquires three-dimensional point cloud data by transmitting and receiving millimeter-wave signals, and then determines a second confidence of the exit or entry feature, and calculates a second distance data of the vehicle and the exit or entry feature through signal propagation time. Among them, the confidence in the embodiment can refer to the probability of determining the feature. For example, when using the front-view camera to acquire image data to determine the exit or entry feature, the first confidence can refer to the first confidence measure of the judgment of the tunnel scene entry / exit. The measure can be based on including the clarity of the image, the obviousness of the feature, the accuracy of the algorithm, etc. A higher confidence indicates a stronger confidence in the feature or event, i.e. more certain that the identified scene is a tunnel scene entry / exit, while a lower confidence indicates a lower confidence of the system.

[0033] It can be understood that the number of front-view cameras and 4D millimeter-wave radars can be one or multiple, and the specific number can be selected according to the actual needs of the vehicle and the application scenario. In addition, the vehicle can also use other sensors and algorithms for assistance. For example, the vehicle can use a laser radar for high-precision three-dimensional modeling and perception to obtain more accurate road information and obstacle information.

[0034] Further, in order to obtain more accurate target distance, the vehicle will combine the first confidence and the second confidence to fuse the first distance data and the second distance data. To improve the reliability and accuracy of the distance data. Through fusion processing, the vehicle can determine the target distance of the exit or entry feature, and if the target distance meets the preset vehicle light opening condition, the specified light of the vehicle is turned on.

[0035] It can be understood that the preset vehicle light opening condition can be a condition set by relevant personnel according to needs or a condition preset when the vehicle is shipped, and the specific condition setting rule can be set according to the vehicle light performance and parameters, which is not limited here.

[0036] By the method of the embodiments of the present application, the vehicle can rely only on the front-view camera and the 4D millimeter wave radar configured by the vehicle to perceive the section characteristics of the front road in real time, and perform corresponding light control according to the perception result, thereby improving the driving safety and intelligent level. In addition, the embodiments of the present application do not rely on satellite communication signals, high-precision maps and other information, and avoid the use of functions in satellite signal-free areas. This method is especially suitable for scenes with special section characteristics such as tunnels and underground parking lots, and provides a more convenient and safe driving experience for drivers.

[0037] In some embodiments, determining the first confidence of the exit feature or the entrance feature by using the image data acquired by the front-view camera includes: preprocessing the image data, storing the preprocessed image data as a grayscale image; performing binaryzation processing on the grayscale image to obtain a binary image, and performing edge detection on the binary image to obtain a contour image; and acquiring the pixel coordinates of the center point of the contour image by using a preset function, and determining the first confidence based on the pixel coordinates of the center point.

[0038] Specifically, the image data acquired by the front-view camera can be preprocessed first (including denoising, color correction, dynamic range adjustment, etc.), to remove noise in the image and improve image quality, and then converted into a grayscale image. Each pixel in the grayscale image has only one brightness value (usually an integer from 0 to 255), representing the depth of color. In this way, only one color channel needs to be processed, which can greatly reduce the computational load of subsequent processing.

[0039] Further, the grayscale image is binaryzation processed to convert the image into only two colors (usually black and white), so as to highlight the key information in the image, such as edges and contours, and provide a clearer image for subsequent edge detection. After obtaining the binary image, edge detection is performed to identify the contours of objects in the image (such as the entrance contour and the exit contour of the tunnel scene), and determine the position and shape of the exit or entrance.

[0040] Further, a preset function (such as a Hough function) is used to extract the pixel coordinates of the center point from the contour image. This usually means that the object in the image (such as the exit or entrance) is identified as a circle or an approximate circle. The first confidence is determined based on the pixel coordinates of the center point. It is used to represent the reliability or accuracy of the identification result. For example, if the determination of the center point coordinates is very accurate, the first confidence will be high; on the contrary, if there is a great uncertainty in the determination of the coordinates, the confidence will be lower.

[0041] According to the technical scheme provided in the embodiment of the present application, the image data is preprocessed, and the preprocessed image data is stored as a grayscale image; the grayscale image is binarized to obtain a binary image, and the binary image is edge detected to obtain a contour image; a preset function is used to obtain a center point pixel coordinate of the contour image, and a first confidence degree is determined based on the center point pixel coordinate, wherein through a series of processing of the image data, whether the target scene includes the exit feature and the entrance feature is determined, so that the vehicle can accurately capture the exit feature and the entrance feature of the target scene through the front-view camera during driving.

[0042] In addition, in some embodiments, the first distance data of the vehicle from the exit feature or the entrance feature is calculated by using a pinhole imaging algorithm, including: obtaining a ranging point of the exit feature or the entrance feature, making the ranging point incident to a pixel coordinate system constructed by an imaging element through a camera hole of the front-view camera, and determining a pixel coordinate of the ranging point in the pixel coordinate system; a first plane parallel to a horizontal plane is constructed, a vertical height, a downward angle and a focal length of the front-view camera are obtained, wherein the downward angle indicates an included angle between an optical axis of the front-view camera and the first plane, and the focal length indicates a vertical distance from the camera hole of the front-view camera to the pixel coordinate system; according to the pixel coordinate, the vertical height, the downward angle and the focal length, the first distance data of the vehicle from the exit feature or the entrance feature is determined by using a similarity relationship algorithm and a trigonometric function.

[0043] Specifically, in order to better illustrate the present embodiment, the following will be combined with Figure 2 to illustrate the present embodiment, Figure 2 is a schematic diagram of a distance determination method of a vehicle and a ranging point provided by the present application, as Figure 2 :

[0044] The coordinate system with X as the horizontal axis and Y as the vertical axis is a horizontal coordinate system, P is a ranging point of the exit feature or the entrance feature obtained, O is a camera hole of the front-view camera, OO1 is a vertical height of the front-view camera, at this time the ranging point can be incident to a pixel coordinate system (the coordinate system with X1 as the horizontal axis and Y1 as the vertical axis) constructed by an imaging element through the camera hole of the front-view camera, the coordinates (Px, Py) of P1 are the pixel coordinates of the ranging point in the pixel coordinate system, in order to facilitate calculation, the coordinates (Px, Py) of P1 are represented as (x, y) in the subsequent. Further, a first plane (i.e. the plane where ΔNOM is located) parallel to the horizontal plane XY is constructed, at this time the included angle between the optical axis OO2 and the horizontal line NO is the downward angle of the camera, when the optical axis is downwardly deviated, α>0, when the optical axis is upwardly deviated, α is less than 0, M is the intersection of the P1Px line segment and the first plane, and N is the intersection of the PyO2 line segment and the first plane.

[0045] It can be understood that the plane where the OMN is located constitutes a first plane perpendicular to OO1, OO2 is the focal length f of the front-view camera, since the first plane is parallel to the XY plane, ∠OPO1=∠P1OM=∠P1OPx+∠PxOM. At this time, the corresponding relationship and trigonometric function of the triangle are utilized, that is, the first distance data is calculated according to the obtained pixel coordinates, the vertical height, the downward angle and the focal length.

[0046] In some examples, in △NO2O, NO2⊥OO2 can obtain:

[0047] O2N=f tanα;

[0048] In △O2PxO, O2Px=x, OO2⊥O2Px can obtain:

[0049]

[0050] PxM=O2N, in △PxOM, OPx⊥PxM can obtain:

[0051]

[0052] In △P1OPx, OPx⊥P1Px, P1Px=O2Py=y can obtain:

[0053]

[0054] In △OPO1, OO1⊥PO1 can obtain:

[0055]

[0056] Further, according to the vertical height H of the front-view camera obtained by the sensor configured by the vehicle, the downward angle α obtained by the attitude sensor, the focal length f of the front-view camera, and the physical pixel coordinates (x, y) of the ranging point, the first distance data d of the ranging point P can be calculated according to the following formula:

[0057]

[0058]

[0059] It can be understood that after the first plane is constructed as an auxiliary plane, the first distance data can also be calculated by utilizing the corresponding relationship and trigonometric function of the remaining triangles, which will not be described here.

[0060] According to the technical scheme provided in the embodiments of the present application, the ranging point of the exit feature or the entrance feature is obtained, the ranging point is incident to the pixel coordinate system constructed by the imaging element through the imaging hole of the front-view camera, and the pixel coordinate of the ranging point in the pixel coordinate system is determined; a first plane parallel to the horizontal plane is constructed, the vertical height, the downward angle and the focal length of the front-view camera are obtained, wherein the downward angle indicates the included angle between the optical axis of the front-view camera and the first plane, and the focal length indicates the vertical distance from the imaging hole of the front-view camera to the pixel coordinate system; according to the pixel coordinate, the vertical height, the downward angle and the focal length, the first distance data between the vehicle and the exit feature or the entrance feature is determined by using the trigonometric function, and the corresponding relationship of the triangle and the trigonometric function are used to ensure the accuracy of the calculation result.

[0061] In addition, in some embodiments, the second confidence of the exit feature or the entrance feature is determined by using the three-dimensional point cloud data obtained by the 4D millimeter wave radar, including: preprocessing the three-dimensional point cloud data, storing the preprocessed three-dimensional point cloud data as a preset queue, wherein the preset queue is used to store N+1 frames of data; when the latest frame of three-dimensional point cloud data is captured, the latest frame of three-dimensional point cloud data is added to the preset queue, and the data stored in the preset queue is fused to generate a target point cloud map; the target point cloud map is subjected to coordinate conversion, and the converted target point cloud map is mined by using a frequency domain analysis method to obtain the frequency feature of the target point cloud map, and the second confidence is determined based on the frequency feature.

[0062] Specifically, the 4D millimeter wave radar can obtain the three-dimensional position and speed information of the front object, forming three-dimensional point cloud data for describing the objects and their motion states in the front environment. The obtained three-dimensional point cloud data is preprocessed, such as filtering and denoising, to ensure more accurate data, which is helpful for subsequent data processing and analysis. The preprocessed three-dimensional point cloud data is stored in a preset queue. This queue can store N+1 frames of data, and the characteristic of this preset queue is first-in first-out, that is, the earliest data will be processed first. When the latest frame of three-dimensional point cloud data is captured, it is added to the preset queue. Then, all the data stored in the queue are fused. The information of multiple data sources is combined into a unified and consistent information representation to improve the accuracy and reliability of the data.

[0063] Further, the fused data will generate a target point cloud map. The target point cloud map can be used for a more accurate representation of objects and their relationships in the environment. Then, the target point cloud map is converted in coordinates from the radar coordinate system to another more commonly used or more intuitive coordinate system, such as the world coordinate system or the vehicle coordinate system. The converted target point cloud map is mined using a frequency domain analysis method. The frequency domain analysis method can analyze data in the frequency domain, which can reveal periodicity, trends or other frequency-related characteristics in the data. Based on the results of the frequency domain analysis, a second confidence of the exit or entrance feature is determined. The second confidence is a measure of the accuracy of the identified exit or entrance feature, and if the confidence is high, the identified feature is likely to be real; if the confidence is low, further verification or processing may be required.

[0064] According to the technical scheme provided by the embodiments of the present application, the 4D millimeter wave radar is used to obtain three-dimensional point cloud data, and a series of processing and analysis steps are used to determine the confidence of the exit or entrance feature, which can ensure the accuracy of the identified entrance feature and exit feature of the target scene.

[0065] In some embodiments, the first distance data and the second distance data are fused using the first confidence and the second confidence to determine the target distance of the vehicle from the exit feature or the entrance feature, including: determining a proportionality coefficient of the first distance data and the second distance data according to the numerical value of the first confidence and the second confidence; and calculating the target distance of the vehicle from the exit feature or the entrance feature according to the first distance data, the second distance, and the proportionality coefficient.

[0066] Specifically, the embodiments determine the proportionality coefficient of the first distance data and the second distance data according to the numerical value of the first confidence and the second confidence, and when the data is fused, the proportion of the first distance data and the second distance data is allocated according to the proportionality coefficient, so that the target distance result obtained is closer to the real distance.

[0067] In order to better illustrate the embodiments, the embodiments will be exemplarily illustrated in combination with the following table, as shown in the following table:

[0068]

[0069] Then the target distance is calculated by the following calculation method:

[0070] Target distance = first distance data * first confidence coefficient + second distance data * second confidence coefficient

[0071] In an example, according to the numerical values of the first confidence and the second confidence, a proportion coefficient of the first distance data and the second distance data is determined, for example, if the first confidence is 0.9 and the second confidence is 0.8, the proportion coefficient is 4:6 according to the above table, at this time the target distance can be calculated by the calculation formula of the target distance: target distance = first distance data * 4 + second distance data * 6.

[0072] It can be understood that the proportion coefficient of the first distance data and the second distance data can be adjusted according to different confidence numerical values to adapt to the data fusion requirements in different scenes.

[0073] In summary, the embodiment fuses the first distance data and the second distance data by using the first confidence and the second confidence to determine the target distance of the vehicle and the exit feature or the entrance feature, which improves the accuracy and safety of vehicle navigation and automatic driving. At the same time, the algorithm also has good scalability and flexibility, which can adapt to the application requirements in different scenes and provides strong support for the development of future intelligent transportation and automatic driving technology.

[0074] According to the technical scheme provided by the embodiment of the application, the first distance data and the second distance data are fused by using the first confidence and the second confidence to determine the target distance of the vehicle and the exit feature or the entrance feature, including: according to the numerical values of the first confidence and the second confidence, a proportion coefficient of the first distance data and the second distance data is determined; according to the first distance data, the second distance, and the proportion coefficient, the target distance of the vehicle and the exit feature or the entrance feature is calculated, in the case that the confidence of the data collected by a device is not high, the data collected by another device can be used to make up for it, so as to ensure that the result obtained tends to be the true value.

[0075] In addition, in some embodiments, if the target distance meets the preset vehicle light opening condition, the specified light of the vehicle is turned on, including: obtaining an external light intensity value, determining a time scene according to the external light intensity value; if the target distance of the vehicle and the entrance feature is less than a first preset distance threshold, and the time scene is in a first time scene, the low beam of the vehicle is controlled to be turned on; if the target distance of the vehicle and the exit feature is less than a second preset distance threshold, and the time scene is in a second time scene, the high beam of the vehicle is controlled to be turned on.

[0076] Specifically, after determining the target distance, the light intensity value of the external environment light of the vehicle can also be obtained by the photosensitive sensor configured by the vehicle, and the current time scene is determined according to the external light intensity value, so as to control the light of the vehicle in combination with the target distance.

[0077] As an example, when the vehicle is driving in the daytime, the light intensity value detected by the photosensitive sensor is high, and it can be determined that the current time scenario is daytime. If the target distance between the vehicle and the entrance feature (tunnel entrance and parking lot entrance) is less than the first preset distance threshold (for example, 100 meters) at this time, it means that the vehicle is about to enter a relatively dark area, and therefore the low beam of the vehicle can be automatically controlled to be turned on to provide more sufficient illumination and ensure driving safety.

[0078] When the vehicle is driving at night or in a relatively dark environment, the light intensity value detected by the photosensitive sensor is low, and it can be determined that the current time scenario is night. If the target distance between the vehicle and the exit feature (tunnel exit) is less than the second preset distance threshold (for example, 50 meters) at this time, it means that the vehicle is about to leave the current illumination area and enter a more dim environment. In order to ensure the driver's sight distance and driving safety, the high beam of the vehicle can be automatically controlled to be turned on to provide a longer illumination distance and better sight effect.

[0079] According to the technical scheme provided in the embodiments of the present application, by combining the judgment of the external light intensity value and the time scenario, the light control strategy of the embodiments can more accurately adapt to different driving scenarios, such as when entering or leaving a tunnel or an underground parking lot, to avoid possible temporary visual blind area, thereby increasing the risk of driving and improving driving safety. At the same time, the automatic light control also reduces the burden of the driver, so that he can pay more attention to the driving process and reduce the safety hazards that may be caused by manual operation of the light.

[0080] In addition, in some embodiments, the method further comprises: analyzing the image data obtained by the front-view camera by using an RGB model and an optical flow analysis method to obtain real-time weather data; and turning on a specified light of the vehicle if the real-time weather data meets a preset condition.

[0081] Specifically, the RGB model is the basis of color representation, and through the analysis of the red, green and blue color channels, the basic information of the color in the image can be obtained. The optical flow analysis method can estimate the motion of objects in the image sequence by analyzing the motion patterns of the pixel points or feature points in the image.

[0082] In some examples, in the image data obtained by the front-view camera, the RGB model can understand the lighting conditions, color distribution and other information of the current environment. It is used to judge the weather conditions, such as sunny, cloudy, rainy and the like. For example, if the colors in the image are generally dark, it may mean that it is cloudy or night; and if there are a large number of blue or white pixels in the image, it may mean that it is raining or snowing.

[0083] In some examples, the optical flow analysis rule can be used to analyze the motion of objects in the image, so as to further infer the weather condition. For example, if the optical flow analysis result shows that the pixels in the image generally present a downward motion trend, it may mean that it is raining, and the raindrops falling from the sky form such a motion pattern.

[0084] It can be understood that, in combination with the results of the RGB model and the optical flow analysis method, a more accurate judgment can be made on the current weather condition. If the judgment result meets the preset condition, for example, a rainy day or a night, etc. low visibility condition, then the high beam light of the vehicle can be turned on to improve the driver's line of sight distance. In the case of heavy fog, the fog lamp of the vehicle is turned on to ensure the safety of driving.

[0085] According to the technical scheme of the embodiment of the present application, the RGB model and the optical flow analysis method are used to analyze the image data obtained by the front-view camera to obtain real-time weather data. If the real-time weather data meets the preset condition, the high beam light of the vehicle is controlled to be turned on. In general, the application of the RGB model and the optical flow analysis method in weather recognition can improve the accuracy of weather recognition, and in turn serve as a control condition for the light of the vehicle, ensuring the safety of the driver during driving and improving the safety of driving.

[0086] Figure 3 is a schematic diagram of a 4D millimeter wave radar data processing flow provided by an embodiment of the present application, as shown in Figure 3

[0087] Millimeter wave point cloud refers to a set of points obtained by scanning the surrounding environment by a 4D millimeter wave radar. Each point contains information such as distance, speed, angle (sometimes also including height), which can be regarded as a three-dimensional representation of the target in the radar coordinate system. Data fusion refers to integrating data from multiple different sources or different time points together to improve the accuracy and reliability of the data. In the scenario in the present embodiment, the millimeter wave point clouds of the previous n frames, the previous n-1 frames, the previous 1 frame and the current frame are fused to obtain a higher density of point cloud data. Errors caused by radar scanning intervals or target motion can be eliminated, and the accuracy of target position and speed can be improved. In addition to the traditional 3-dimensional coordinates (x, y, z), the fused data also contains information such as reflectivity / RCS (Radar Cross Section, radar cross section, used to measure the reflection ability of the target to radar waves), Doppler speed / absolute speed (representing the speed of the target relative to the radar), and target category (such as vehicle, pedestrian, bicycle, etc.). The target detection network is a deep learning network for identifying and locating targets from the fused high-density millimeter wave point cloud data. By processing the fused high-density millimeter wave point cloud, the center point position, size, orientation and category of each detected target can be output.​

[0088] Figure 4 is a hardware architecture diagram of another vehicle light control method provided by the embodiment of the present application, as shown in Figure 4

[0089] In the embodiment, the vehicle automatic headlight function is in an open state, and the front-view camera and the 4D millimeter wave radar configured by the vehicle can identify the scene in front of the vehicle in real time, determine whether the front road has a road section feature in a target scene (for example, an entrance feature or an exit feature of a tunnel scene, or an entrance feature and an exit feature of an underground parking lot) according to the identification result, if it is determined that the front road has an exit feature or an entrance feature in a target scene, the vehicle will use the front-view camera and the 4D millimeter wave radar to obtain more accurate data and process, and transmit the target distance obtained after processing to the intelligent driving domain controller, when the target distance meets the preset light-on condition, the intelligent driving domain controller will transmit the corresponding light control instruction to the headlight controller, and the headlight controller will execute the instruction of turning on the low beam or the high beam combined with the time scene information transmitted by the photosensitive sensor.

[0090] All the optional technical solutions described above can be combined to form optional embodiments of the present application, which will not be described one by one.

[0091] It should be understood that the size of the serial number of each step in the above embodiment does not mean the order of execution, the execution order of each process should be determined according to its function and inherent logic, and should not constitute any limitation on the process of the embodiment of the present application.

[0092] Figure 5 is a schematic diagram of a vehicle light control device provided by an embodiment of the present application. As shown in Figure 5 the device comprises:

[0093] The identification module 501 is configured to identify the road section feature in front of the vehicle by the front-view camera and the 4D millimeter wave radar respectively, and determine whether the driving road section contains the exit feature or the entrance feature of the target scene;

[0094] The camera module 502 is configured to determine the first confidence of the exit feature or the entrance feature by using the image data obtained by the front-view camera, and calculate the first distance data of the vehicle from the exit feature or the entrance feature by using the pinhole imaging algorithm if so;

[0095] The radar module 503 is configured to determine the second confidence of the exit feature or the entrance feature by using the three-dimensional point cloud data obtained by the 4D millimeter wave radar, and calculate the second distance data of the vehicle from the exit feature or the entrance feature by using the transceiving signal of the 4D millimeter wave radar;

[0096] ​The fusion module 504 is configured to fuse the first distance data and the second distance data according to the first confidence and the second confidence, and determine a target distance of the vehicle to the exit feature or the entrance feature;

[0097] The light control module 505 is configured to turn on a specified light of the vehicle if the target distance meets a preset vehicle light turning-on condition.

[0098] In some embodiments, the camera module 502 is further configured to pre-process the image data, store the pre-processed image data as a grayscale image, perform binaryzation processing on the grayscale image to obtain a binary image, and perform edge detection on the binary image to obtain a contour image; and obtain a center point pixel coordinate of the contour image using a preset function, and determine the first confidence based on the center point pixel coordinate.

[0099] In some embodiments, the camera module 502 is further configured to obtain a ranging point of the exit feature or the entrance feature, and cause the ranging point to be incident on a pixel coordinate system constructed by an imaging element through a camera hole of the front-view camera, and determine a pixel coordinate of the ranging point in the pixel coordinate system; construct a first plane parallel to a horizontal plane, obtain a vertical height, a downward angle and a focal length of the front-view camera, wherein the downward angle indicates an included angle between an optical axis of the front-view camera and the first plane, and the focal length indicates a vertical distance from the camera hole of the front-view camera to the pixel coordinate system; and determine the first distance data of the vehicle to the exit feature or the entrance feature using a similarity relationship algorithm and a trigonometric function based on the pixel coordinate, the vertical height, the downward angle and the focal length.

[0100] In some embodiments, the radar module 503 is further configured to pre-process the three-dimensional point cloud data, and store the pre-processed three-dimensional point cloud data in a preset queue, wherein the preset queue is used to store N+1 frames of data; when a latest frame of three-dimensional point cloud data is captured, add the latest frame of three-dimensional point cloud data to the preset queue, and fuse the data stored in the preset queue to generate a target point cloud image; perform coordinate conversion on the target point cloud image, and mine the converted target point cloud image using a frequency domain analysis method to obtain a frequency feature of the target point cloud image, and determine the second confidence based on the frequency feature.

[0101] In some embodiments, the fusion module 504 is further configured to determine a proportionality coefficient of the first distance data and the second distance data according to a numerical size of the first confidence and the second confidence; and calculate the target distance of the vehicle to the exit feature or the entrance feature according to the first distance data, the second distance and the proportionality coefficient.

[0102] ​In some embodiments, the light control module 505 is further configured to acquire an external light intensity value, determine a time scene according to the external light intensity value, control a low beam of the vehicle to be turned on if a target distance between the vehicle and the entrance feature is less than a first preset distance threshold and the time scene is in a first time scene, and control a high beam of the vehicle to be turned on if a target distance between the vehicle and the exit feature is less than a second preset distance threshold and the time scene is in a second time scene.

[0103] In some embodiments, the light control module 505 is further configured to analyze image data acquired by the front-view camera by using an RGB model and an optical flow analysis method to obtain real-time weather data, and turn on a specified light of the vehicle if the real-time weather data meets a preset condition.

[0104] The apparatus provided by the embodiments of the present application can implement all the method steps of the above-mentioned method embodiments, and achieve the same technical effects. Therefore, repeated details will not be described herein.

[0105] Figure 6 is a schematic diagram of an electronic device 6 provided by an embodiment of the present application. As shown in the figure, the electronic device 6 of this embodiment includes a processor 601, a memory 602, and a computer program 603 stored in the memory 602 and executable on the processor 601. The processor 601 implements the steps in the above-mentioned various method embodiments when executing the computer program 603. Alternatively, the processor 601 implements the functions of the modules / units in the above-mentioned various apparatus embodiments when executing the computer program 603. Figure 6 The electronic device 6 can be a desktop computer, a notebook computer, a palm computer, a cloud server, or the like. The electronic device 6 can include but is not limited to the processor 601 and the memory 602. Those skilled in the art can understand that the electronic device 6 can include more or fewer components or different components than those shown in the figure.

[0106] The electronic device 6 shown in the figure is merely an example of the electronic device 6 and does not constitute a limitation on the electronic device 6, which can include more or fewer components or different components than those shown in the figure. Figure 6 The processor 601 can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gates or transistor logic components, discrete hardware components, or the like.

[0107]

[0108] ​The memory 602 can be an internal storage unit of the electronic device 6, for example, a hard disk or a memory of the electronic device 6. The memory 602 can also be an external storage device of the electronic device 6, for example, a plug-in hard disk, a smart media card (SMC), a secure digital (SD) card, a flash card, etc. equipped on the electronic device 6. The memory 602 can also include both the internal storage unit and the external storage device of the electronic device 6. The memory 602 is used to store computer programs and other programs and data required by the electronic device.

[0109] Those skilled in the art can clearly understand that, for the convenience and brevity of description, only the division of the above functional units and modules is exemplified, and in actual application, the above functions can be completed by different functional units and modules according to needs, that is, the internal structure of the device is divided into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit, and the integrated unit can be realized in the form of hardware or software.

[0110] The integrated module / unit, if realized in the form of a software function unit and sold or used as an independent product, can be stored in a readable storage medium. Based on this understanding, all or part of the processes in the above-mentioned embodiment methods can also be completed by a computer program instructing related hardware, and the computer program can be stored in a readable storage medium. When the processor executes the computer program, the steps of each method embodiment described above can be implemented. The computer program can include computer program code, which can be in the form of source code, object code, executable files or some intermediate forms, etc. The readable storage medium can include any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc.

[0111] The above examples are only used to illustrate the technical solutions of the present application, but not limit the same; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that the technical solutions recorded in the foregoing examples can be modified, or some technical features can be replaced by equivalent ones; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application, and should be included in the protection scope of the present application.

Claims

1. A method for controlling vehicle lights, characterized in that, include: The road features ahead of the vehicle are identified by at least one forward-looking camera and at least one 4D millimeter-wave radar, respectively, to determine whether the road features contain exit or entrance features of the target scene. If so, the first confidence level of the exit feature or entrance feature is determined using the image data acquired by the forward-looking camera, and the first distance data between the vehicle and the exit feature or entrance feature is calculated using the pinhole imaging algorithm; The second confidence level of the exit feature or entrance feature is determined using the three-dimensional point cloud data acquired by the 4D millimeter-wave radar, and the second distance data between the vehicle and the exit feature or entrance feature is calculated using the transmit and receive signals of the 4D millimeter-wave radar. The first distance data and the second distance data are fused based on the first confidence level and the second confidence level to determine the target distance between the vehicle and the exit feature or entrance feature. A ratio coefficient between the first distance data and the second distance data is determined based on the magnitude of the first confidence level and the second confidence level. The target distance between the vehicle and the exit feature or entrance feature is calculated based on the first distance data, the second distance, and the ratio coefficient. If the target distance meets the preset headlight activation conditions, then the designated headlights of the vehicle are activated.

2. The method according to claim 1, characterized in that, Determining the first confidence level of the exit feature or entrance feature using image data acquired by the forward-looking camera includes: The image data is preprocessed, and the preprocessed image data is stored as a grayscale image; The grayscale image is binarized to obtain a binarized image, and edge detection is performed on the binarized image to obtain a contour image; The pixel coordinates of the center point of the contour image are obtained using a preset function, and the first confidence level is determined based on the pixel coordinates of the center point.

3. The method according to claim 1, characterized in that, The calculation of the first distance data between the vehicle and the exit or entrance feature using the pinhole imaging algorithm includes: The ranging point of the exit feature or entrance feature is obtained, and the ranging point is incident on the pixel coordinate system constructed by the imaging element through the camera hole of the forward-looking camera, and the pixel coordinates of the ranging point in the pixel coordinate system are determined. Construct a first plane parallel to the horizontal plane, and obtain the vertical height, downward tilt angle, and focal length of the front-view camera, wherein the downward tilt angle indicates the angle between the optical axis of the front-view camera and the first plane, and the focal length indicates the vertical distance from the camera aperture of the front-view camera to the pixel coordinate system; Based on the pixel coordinates, the vertical height, the downward angle, and the focal length, a first distance data between the vehicle and the exit feature or entrance feature is determined using a similarity algorithm and trigonometric functions.

4. The method according to claim 1, characterized in that, The second confidence level for determining the exit feature or entrance feature using the three-dimensional point cloud data acquired by the 4D millimeter-wave radar includes: The 3D point cloud data is preprocessed, and the preprocessed 3D point cloud data is stored in a preset queue, wherein the preset queue is used to store N+1 frames of data; when the latest frame of the 3D point cloud data is captured, the latest frame of the 3D point cloud data is added to the preset queue, and the data stored in the preset queue is fused to generate a target point cloud map; the target point cloud map is subjected to coordinate transformation, and the transformed target point cloud map is mined using frequency domain analysis to obtain the frequency characteristics of the target point cloud map, and the second confidence level is determined based on the frequency characteristics.

5. The method according to claim 1, characterized in that, The step of turning on the designated lights of the vehicle if the target distance meets the preset headlight turning-on conditions includes: Obtain the external light intensity value, and determine the time scene based on the external light intensity value; If the target distance between the vehicle and the entrance feature is less than a first preset distance threshold, and the time scene is a first time scene, then control the low beam headlights of the vehicle to be turned on. If the distance between the vehicle and the target of the exit feature is less than a second preset distance threshold, and the time scenario is a second time scenario, then the high beams of the vehicle are turned on.

6. The method according to claim 1, characterized in that, The method further includes: Real-time weather data is obtained by analyzing the image data acquired by the forward-looking camera using the RGB model and optical flow analysis method. If the real-time weather data meets the preset conditions, then the designated lights of the vehicle are turned on.

7. A vehicle lighting control device, characterized in that, include: The identification module is configured to identify road features ahead of the vehicle using at least one forward-facing camera and at least one 4D millimeter-wave radar, and determine whether the road features contain exit or entrance features of the target scene. The camera module is configured to, if so, determine a first confidence level of the exit feature or entrance feature using image data acquired by the forward-looking camera, and calculate a first distance data between the vehicle and the exit feature or entrance feature using a pinhole imaging algorithm; The radar module is configured to determine a second confidence level of the exit feature or entrance feature using three-dimensional point cloud data acquired by the 4D millimeter-wave radar, and to calculate a second distance data between the vehicle and the exit feature or entrance feature using the transceiver signals of the 4D millimeter-wave radar. The fusion module is configured to fuse the first distance data and the second distance data according to the first confidence level and the second confidence level to determine the target distance between the vehicle and the exit feature or entrance feature; wherein a ratio coefficient between the first distance data and the second distance data is determined according to the magnitude of the first confidence level and the second confidence level; and the target distance between the vehicle and the exit feature or entrance feature is calculated according to the first distance data, the second distance, and the ratio coefficient. The lighting control module is configured to turn on the designated lights of the vehicle if the target distance meets the preset headlight turning-on conditions.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the steps of the method as described in any one of claims 1 to 6.

9. A readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method as described in any one of claims 1 to 6.

Images