Lighting equipment control method of vehicle, vehicle and electronic equipment

Abstract

The embodiment of the invention provides a lighting equipment control method of a vehicle, the vehicle and electronic equipment.The method comprises the steps that a control instruction triggered for lighting equipment is responded, an environment image set of the vehicle is obtained, and environment images in the environment image set are used for representing the brightness of the environment where the vehicle is located; based on the driving direction of the vehicle, performing partition processing on at least one environment image in the environment image set to obtain a plurality of image areas; weight information of the multiple image areas is determined respectively, brightness information of the multiple image areas is determined respectively, and the weight information is used for representing the brightness of the image areas and the degree of influence of the brightness of the image areas on the driving safety degree of the vehicle; based on the weight information and the brightness information, a control strategy of the lighting equipment is determined, and the control strategy is used for representing a rule for controlling the lighting equipment; and controlling the lighting equipment according to the control strategy. The technical problem that the control accuracy of the lighting equipment of the vehicle is low is solved.

Description

Technical Field

[0001] This application relates to the field of vehicle lighting control system technology, and more specifically, to a vehicle lighting equipment control method, a vehicle, and electronic equipment. Background Technology

[0002] Currently, in the adjustment and control of vehicle lighting equipment, related technologies often rely on cameras to sense ambient light to determine the lighting mode. However, limitations in camera hardware and insufficient light-sensing performance often lead to errors in environmental judgment, resulting in control errors in the lighting equipment and affecting driving safety and efficiency. Therefore, the technical problem of low accuracy in vehicle lighting equipment control still exists.

[0003] There is currently no good solution to the above problems. Summary of the Invention

[0004] This application provides a method for controlling vehicle lighting equipment, a vehicle, and an electronic device to at least solve the technical problem of low accuracy in controlling vehicle lighting equipment.

[0005] According to one aspect of the embodiments of this application, a method for controlling a vehicle's lighting equipment is provided. The method may include: responding to a control command triggered for the lighting equipment; acquiring a set of environmental images of the vehicle, wherein the environmental images in the set represent the brightness of the environment in which the vehicle is located; partitioning at least one environmental image in the set based on the vehicle's driving direction to obtain multiple image regions; determining weight information for each of the multiple image regions and brightness information for each of the multiple image regions, wherein the weight information represents the degree to which the brightness of the image region affects the vehicle's driving safety; determining a control strategy for the lighting equipment based on the weight information and the brightness information, wherein the control strategy represents the rules for controlling the lighting equipment; and controlling the lighting equipment according to the control strategy, wherein the controlled lighting equipment is used to increase the brightness of the environment.

[0006] Furthermore, based on the vehicle's driving direction, at least one environmental image in the environmental image set is partitioned to obtain multiple image regions, including: extracting a first image region from the environmental image, wherein the distortion degree of the first image region is less than the distortion degree of other image regions in the environmental image besides the first image region; and partitioning the first image region according to the driving direction to obtain multiple image regions.

[0007] Furthermore, the image region includes a first image sub-region, a second image sub-region, and a third image sub-region. The first image region is partitioned according to the driving direction to obtain multiple image regions, including: determining a first image sub-region from the first image region based on the driving direction and vehicle width information; determining a second image sub-region from the first image region based on the driving direction and vehicle angle information, wherein the angle information is used to represent the degree of diffusion of light emitted by the lighting device; and determining the image region from the first image region other than the first and second image sub-regions as the third image sub-region.

[0008] Further, the weight information of multiple image regions is determined separately, including: calling a weight classification model to determine the first weight information of the first image sub-region and the second weight information of the second image sub-region, wherein the weight classification model is obtained by training a recurrent neural network model, and the first weight information is greater than the second weight information; or, the brightness information of multiple image regions is determined separately, including: identifying multiple pixel blocks in the image region to obtain sub-brightness information of the pixel blocks, wherein the sub-brightness information is used to represent the brightness of the environment in the pixel block; and determining the brightness information of the image region based on the sub-brightness information corresponding to the multiple pixel blocks respectively.

[0009] Furthermore, based on weight information and brightness information, a control strategy for the lighting equipment is determined, including: determining the number of environmental images in the environmental image set; weighting the weight information and brightness information based on the number to obtain a weighted processing result; and determining the control strategy based on the weighted processing result.

[0010] Furthermore, based on the quantity, the weight information and brightness information are weighted to obtain a weighted processing result, including: in response to a quantity of one, the weight information and brightness information are weighted to obtain a weighted processing result; or, based on the quantity, the weight information and brightness information are weighted to obtain a weighted processing result, including: in response to a quantity of multiple, determining the third weight information of multiple environmental images in the environmental image set, wherein the third weight information is used to represent the degree of influence of the environmental image on driving safety; and weighting the third weight information, the weight information, and the brightness information to obtain a weighted processing result.

[0011] Furthermore, in response to the number of multiple environmental images, the third weight information of multiple environmental images in the environmental image set is determined respectively, including: in response to the number of multiple environmental images, the third weight information corresponding to each environmental image is determined based on the acquisition time of each environmental image in the environmental image set, wherein the magnitude of the third weight information is negatively correlated with the time of acquisition.

[0012] Further, based on the weighted processing results, a control strategy is determined, including: evaluating the weighted processing results of the environmental images in the environmental image set to obtain an evaluation result, wherein the evaluation result is used to represent the brightness state of the environment; in response to the evaluation result being greater than a first evaluation result threshold, a first control strategy is determined; in response to the evaluation result being greater than a second evaluation result threshold, a second control strategy is determined, wherein the brightness state corresponding to the second evaluation result threshold is greater than the brightness state corresponding to the first evaluation result threshold, and the brightness of the environment increased according to the second control strategy is higher than the brightness of the environment increased according to the first control strategy.

[0013] According to another aspect of the embodiments of this application, a vehicle lighting equipment control device is also provided. This device may include: an acquisition module, configured to acquire a set of environmental images of the vehicle in response to a control command triggered for the lighting equipment, wherein the environmental images in the set represent the brightness of the environment in which the vehicle is located; a processing module, configured to partition at least one environmental image in the set based on the vehicle's driving direction to obtain multiple image regions; a first determining module, configured to determine weight information for each of the multiple image regions and brightness information for each of the multiple image regions, wherein the weight information represents the degree to which the brightness of the image region affects the vehicle's driving safety; a second determining module, configured to determine a control strategy for the lighting equipment based on the weight information and the brightness information, wherein the control strategy represents the rules for controlling the lighting equipment; and a control module, configured to control the lighting equipment according to the control strategy, wherein the controlled lighting equipment increases the brightness of the environment.

[0014] According to another aspect of the embodiments of this application, a vehicle is also provided, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods in various embodiments of this application when it runs.

[0015] According to another aspect of the embodiments of this application, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0016] According to another aspect of the embodiments of this application, a computer program product is also provided, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0017] According to another aspect of the embodiments of this application, a computer program product is also provided, including a non-volatile computer-readable storage medium storing a computer program, which, when executed by a processor, implements the methods in various embodiments of this application.

[0018] According to another aspect of the embodiments of this application, a computer program is also provided, which, when executed by a processor, implements the methods of the various embodiments of this application.

[0019] In this embodiment, if vehicle lighting equipment needs to be controlled, control commands triggered by the lighting equipment can be detected. In response to these commands, environmental images are acquired in real time, containing real-time illumination information around the vehicle. Based on the vehicle's real-time driving direction, the images are finely partitioned into multiple image regions. This method ensures that the illumination intensity from different perspectives is fully considered. Weight information is assigned to each image region, reflecting the actual impact of brightness in different regions on driving safety. Combining the brightness and weight information of each image region allows for a more accurate assessment of the overall ambient illumination state, leading to a more reasonable lighting equipment control strategy. Following this control strategy, the brightness provided by the lighting equipment can be dynamically adjusted. By introducing deep correlation analysis between the environmental image set and the vehicle's driving direction, not only are misjudgments caused by camera hardware limitations reduced, but the accuracy and safety of vehicle lighting control in complex environments are significantly enhanced. This solves the technical problem of low accuracy in vehicle lighting equipment control and achieves the technical effect of improving the accuracy of vehicle lighting equipment control. Attached Figure Description

[0020] The accompanying drawings, which are included to provide a further understanding of this application and form part of this application, illustrate exemplary embodiments of this application and are used to explain this application, but do not constitute an undue limitation of this application. In the drawings:

[0021] Figure 1 This is a flowchart of a vehicle lighting equipment control method according to an embodiment of this application;

[0022] Figure 2 This is a schematic diagram of an automatic high / low beam control architecture according to an embodiment of this application;

[0023] Figure 3 This is a schematic diagram of an image statistical region according to an embodiment of this application;

[0024] Figure 4 This is a schematic diagram of a long short-term memory network model according to an embodiment of this application;

[0025] Figure 5 This is a schematic diagram of a statistical-based automatic high beam headlight control recommendation strategy process according to an embodiment of this application;

[0026] Figure 6 This is a schematic diagram of a vehicle lighting control device according to an embodiment of this application. Detailed Implementation

[0027] To enable 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 accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present application.

[0028] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0029] According to an embodiment of this application, a method for controlling lighting equipment in a vehicle is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0030] This embodiment provides a method for controlling the lighting equipment of a vehicle. Figure 1 This is a flowchart of a vehicle lighting equipment control method according to an embodiment of this application, such as... Figure 1 As shown, the method may include the following steps.

[0031] Step S102: In response to a control command triggered by the lighting equipment, acquire a set of environmental images of the vehicle.

[0032] In the technical solution provided by step S102 of the embodiments of this application, the environmental images in the environmental image set can be used to represent the brightness of the environment in which the vehicle is located.

[0033] Optionally, the control command can be a signal indicating that the vehicle's ignition switch is on (IGN ON) and the lighting equipment is in automatic mode. This control command can instruct the vehicle to enter autonomous driving lighting mode, meaning that the lighting equipment can be autonomously adjusted based on the driving environment, eliminating the need for driver intervention. This control command not only simplifies the driver's workflow but also ensures that the lighting strategy adapts flexibly to changes in the environment, improving driving safety and comfort.

[0034] Optionally, the environmental image set can be a collection of environmental images captured by onboard cameras, designed to comprehensively reflect the lighting conditions and changes around the vehicle. Each frame of the environmental image is a snapshot of the vehicle's real-time environment, containing information such as light intensity distribution, weather conditions, road lighting, and other light sources (e.g., oncoming vehicle headlights). These environmental images not only reveal the numerical characteristics of brightness but also provide intuitive visual feedback. Through environmental image analysis, different environmental elements, such as streetlights, signs, and obstacles, can be identified and distinguished, thus providing a basis for developing precise lighting control strategies.

[0035] Optionally, the vehicle's lighting equipment can be devices on the vehicle that increase the brightness of the environment in which the vehicle is located, such as the vehicle's headlights or rear lights that activate when the vehicle is reversing. The aforementioned headlights can also be referred to as high and low beam headlights.

[0036] In this embodiment, if a control command triggered for the lighting device is detected, a set of environmental images of the vehicle can be acquired.

[0037] Optionally, if an IGN ON event is detected, the initialization of the vehicle's lighting control system can be triggered, indicating that the vehicle is ready to receive further commands. After confirming "IGN ON," the vehicle status is checked to ensure the switch is on and the headlights are in automatic control mode. In automatic control mode, the high / low beam adjustment of the lighting equipment is intelligently determined by the system based on environmental image analysis results, without manual intervention. The onboard camera can be activated to capture real-time images of the area in front of and around the vehicle. Continuous image capture by the onboard camera generates a series of continuous environmental images, forming an environmental image set.

[0038] Optionally, the acquired environmental images can be denoised, corrected, and formatted to improve the accuracy and efficiency of subsequent analysis.

[0039] Step S104: Based on the vehicle's driving direction, at least one environmental image in the environmental image set is partitioned to obtain multiple image regions.

[0040] In the technical solution provided by step S104 of this application embodiment, the vehicle's driving direction can be used to indicate the main direction of the vehicle's movement. This driving direction can include: forward direction (directly forward), reverse direction (directly backward), etc., without specific limitations. Specifically, "directly forward" can be used to indicate the vehicle's front field of vision, which is the part with the most concentrated and critical lighting needs during nighttime driving. "Directly backward" can be used to indicate its function in specific situations (such as checking the ambient brightness behind the vehicle).

[0041] Optionally, the image region can be obtained by dividing the environmental image into different parts according to the driving direction. For example, the image region can include Zone I, Zone II, and areas other than Zones I and II (non-I and non-II zones). Zone I can represent a key statistical area, located at the center of the environmental image or directly in front of the driver's line of sight. This key statistical area has a direct impact on driving safety because it covers the most direct road conditions and potential obstacles within the driver's field of vision. Zone II can be a secondary key area, including the sides or further away of the image. Although less important than Zone I, Zone II can still be monitored to help understand the overall environment in which the vehicle is located. Non-I and non-II zones can include the edges of the environmental image or other non-critical fields of view. These non-I and non-II zones have very low or even negligible weight in illumination analysis, avoiding misjudgments caused by edge distortion or interference from irrelevant light sources.

[0042] It should be noted that the image regions obtained by partitioning the environmental image based on the driving direction are merely illustrative examples and are not specifically limited here. Any process and method that can measure which image regions' brightness has a greater impact on vehicle driving safety based on the driving direction and thus perform environmental image partitioning is within the protection scope of the embodiments of this application.

[0043] In this embodiment, after acquiring the environmental image set of the vehicle, the environmental images and at least one of them can be partitioned based on the vehicle's driving direction to obtain multiple image regions.

[0044] Optionally, multi-sensor fusion technology, including the vehicle's Global Positioning System (GPS), gyroscope, and accelerometer, can be used to capture real-time sensor data such as the vehicle's current position, direction of movement, and speed. By analyzing this sensor data, it can be determined whether the vehicle is traveling straight or in reverse, thus adapting to different image partitioning rules.

[0045] Optionally, an appropriate image zoning scheme can be selected based on the acquired vehicle's driving direction and status. For example, when driving straight, the focus is on the forward view, while when reversing, the focus is more on the rear view. Environmental factors such as weather, road lighting, and time can also be considered to adjust the zoning emphasis to adapt to different driving conditions. Based on the vehicle's current driving direction, key view areas in the image that match the driving direction are determined.

[0046] Optionally, based on the aforementioned preset image partitioning rules, the image region most directly related to the driving direction and having the greatest impact on driving safety can be designated as Zone I. The brightness information analysis of Zone I has the highest priority and weight. Image regions slightly farther from the driving direction but still monitorable are marked as Zone II. The brightness analysis weight of Zone II is lower than that of Zone I, but it is still monitored to provide additional environmental information. Edges or non-critical parts of the environmental image that are unrelated to the current driving direction are designated as non-Zone I and non-Zone II. These non-Zone I and non-Zone II regions can be assigned the lowest weight or ignored in illumination analysis to reduce redundant calculations and improve control efficiency.

[0047] In this embodiment, by partitioning the environmental image set based on the vehicle's driving direction, the intelligence and adaptability of the automatic high beam control system are significantly improved. Clear image region division ensures accurate capture of key visual information, especially the light analysis of the more important driving direction, greatly enhancing responsiveness in complex road conditions.

[0048] Step S106: Determine the weight information of multiple image regions and the brightness information of multiple image regions respectively.

[0049] In the technical solution provided by step S106 of the embodiments of this application, the weight information can be used to represent the brightness of the image area and the degree of influence on the driving safety of the vehicle.

[0050] Optionally, the weighting information can be used to measure the importance of different environmental regions in the environmental image to vehicle driving safety and lighting control decisions. Weighting coefficients can reflect the contribution and priority of each image region in the lighting environment analysis. The aforementioned weighting information can include weighting coefficients for Region I and Region II. The weighting coefficient for Region I can be assigned a higher weight because Region I is directly related to the driver's field of vision and the road conditions ahead, used to promptly detect obstacles in the driving direction and judge road brightness, thus having the greatest impact on high and low beam control decisions. The weighting coefficient for Region II can be set lower than that for Region I. The weighting coefficient for Region II can cover secondary priority areas of the environmental image, such as the side front, distant roads, or potential meeting areas. While information from these secondary priority areas is not immediately necessary, it provides essential supplementation when comprehensively assessing the driving environment, helping to avoid potential risks.

[0051] Optionally, brightness information can refer to data extracted from the environmental image that describes the illumination intensity of each image region. This brightness information can include Type I brightness calculation and Type II brightness calculation. Type I brightness calculation can be applied to region I, using a high-precision algorithm to calculate the average brightness or brightness distribution characteristics of pixels within region I to accurately assess illumination conditions in the vehicle's driving direction. Type II brightness calculation can be implemented in region II. Compared to region I, Type II brightness calculation has slightly lower accuracy and complexity, but still provides valuable information. Type II brightness calculation focuses on identifying potential illumination change patterns, contributing to the overall driving environment assessment and helping the system make more comprehensive lighting control strategies.

[0052] In this embodiment, after partitioning the environmental images in the environmental image set based on the driving direction to obtain multiple image regions, the weight information and brightness information of each image can be determined respectively.

[0053] Optionally, the weights of each image region are dynamically adjusted based on real-time changes in driving direction and ambient lighting to ensure that the analysis focus aligns with the actual situation. For regions I and II, high-precision and moderate-precision brightness calculation methods are employed to extract type I and type II brightness calculation information, respectively, to evaluate the lighting intensity within the region. Based on the selected brightness algorithm, the pixel values ​​of each image region are statistically analyzed to calculate the average brightness or brightness distribution characteristics, thereby obtaining brightness information.

[0054] In this embodiment of the application, the above method can accurately adjust the analysis weight information of different image regions according to the specific content of the vehicle driving direction and the environmental image, effectively extract and analyze brightness information, and significantly improve the safety and comfort of vehicle driving.

[0055] Step S108: Determine the control strategy for the lighting equipment based on the weight information and brightness information.

[0056] In the technical solution provided by step S108 of this application embodiment, the control strategy is used to represent the rules for controlling the lighting equipment. The control strategy can refer to a set of rules that determine the working mode of the vehicle lighting equipment based on the weight information and brightness information of the environmental image. If the lighting equipment is the headlight of a vehicle, the corresponding control strategy can be a high beam and low beam switching strategy.

[0057] In this embodiment, after determining the weight information and brightness information of the image region, the control strategy of the lighting device can be determined based on the weight information and brightness information.

[0058] Optionally, brightness information is combined with corresponding weight information to weight the brightness data of each region. The brightness information of key regions (Region I) has a more significant impact on the final decision due to the high weight of key regions. By comparing the weighted brightness data with the preset lighting condition threshold, and using the time series analysis function of Long Short-Term Memory (LSTM) network, the changing trend of lighting conditions is evaluated to determine whether the environment is approaching the dark conditions where high beams need to be turned on or the bright conditions where low beams should be switched.

[0059] Optionally, based on the above analysis results, specific lighting equipment control strategies can be generated, such as "recommended to turn on high beams" or "suggest switching to low beams".

[0060] In this embodiment of the application, the above method can intelligently generate lighting equipment control strategies, effectively and dynamically and precisely control the switching of high and low beams according to actual environmental conditions during nighttime driving. This reflects a high degree of automation, intelligence and humanization design concept, and significantly improves the safety and comfort of nighttime driving.

[0061] Step S110: Control the lighting equipment according to the control strategy.

[0062] In the technical solution provided by step S110 in the embodiments of this application, the controlled lighting device is used to increase the brightness of the environment.

[0063] In this embodiment, after determining the control strategy for the lighting equipment based on weight information and brightness information, the lighting equipment can be controlled according to the control strategy to improve the brightness of the environment.

[0064] Optionally, the above control strategy can be sent to the vehicle's lighting control system for execution. Simultaneously, through continuous learning and feedback mechanisms, threshold settings and weight allocation are gradually optimized to adapt to more diverse driving environments and personalized needs. The lighting equipment adjusts its operating status according to the received control strategy, and the system synchronously monitors the execution effect, making fine adjustments as necessary to ensure that the lighting strategy improves the driver's visibility while avoiding adverse effects on other road users, achieving good driving safety.

[0065] In steps S102 to S110 of this embodiment, if it is necessary to control the vehicle's lighting equipment, control commands triggered by the lighting equipment can be detected. In response to these control commands, environmental images are acquired in real time, containing real-time illumination information around the vehicle. Based on the vehicle's real-time driving direction, the images are finely partitioned into multiple image regions. This method ensures that the illumination intensity from different perspectives is fully considered. Weight information is assigned to each image region, reflecting the actual impact of brightness in different image regions on driving safety. Combining the brightness information and weight information of each image region allows for a more accurate assessment of the overall ambient illumination state, thereby enabling the development of a more reasonable lighting equipment control strategy. Following this control strategy, the brightness provided by the lighting equipment can be dynamically adjusted. By introducing deep correlation analysis between the environmental image set and the vehicle's driving direction, not only are misjudgments caused by camera hardware limitations reduced, but the accuracy and safety of vehicle lighting control in complex environments are significantly enhanced. This solves the technical problem of low accuracy in vehicle lighting equipment control and achieves the technical effect of improving the accuracy of vehicle lighting equipment control.

[0066] The embodiments of this application will be described in detail below with reference to the steps described above.

[0067] As an optional implementation, step S104 involves partitioning at least one environmental image in the environmental image set based on the vehicle's driving direction to obtain multiple image regions, including: extracting a first image region from the environmental image, wherein the distortion degree of the first image region is less than the distortion degree of other image regions in the environmental image besides the first image region; and partitioning the first image region according to the driving direction to obtain multiple image regions.

[0068] In this embodiment, during the process of partitioning the environmental images in the environmental image set based on the vehicle's driving direction, a first image region can be extracted from the environmental images. This first image region can be further partitioned to obtain multiple image regions. The distortion degree of the first image region is less than the distortion degree of other image regions in the environmental image besides the first image region.

[0069] Optionally, the first image region can refer to the effective detection area, such as a rectangular area covering 60% to 70% of the center of the environmental image (excluding edge distortion). The core of the aforementioned first image region lies in the accurate definition and priority analysis of the most critical parts of the environmental image, aiming to improve the efficiency and accuracy of the automatic high beam control system.

[0070] Optionally, during the distortion correction and first image region extraction process of the environmental image, software algorithms can automatically correct the distortion at the edges of the environmental image to ensure the visual accuracy of the central region. Based on the vehicle's driving direction, a rectangular area covering 60% to 70% of the center of the environmental image is defined as the first image region. This first image region, due to its smaller distortion, is considered the most reliable part for brightness information. The first image region is then accurately extracted from the environmental image to prepare for subsequent brightness analysis.

[0071] Optionally, during the process of refining the first image region into sub-regions, the vehicle's driving direction can be further analyzed, taking into account possible turns, straight-line travel, etc., to ensure that the sub-regions match the vehicle's forward path. Based on the driving direction, the first image region can be subdivided into multiple image regions. For example, the first image region can be subdivided into: Level I key regions, the area directly in front of the vehicle, which are crucial for high and low beam headlight control and have a high weight; Level II key regions, the area to the side or further ahead, which have a supplementary role in high and low beam headlight control and have a lower weight; and non-key regions, the area far from the vehicle's driving path or not expected to affect headlight control, with a low weight or even ignored.

[0072] In this embodiment of the application, the above method can finely process environmental images, reasonably allocate analysis weights for different areas, formulate lighting equipment control strategies based on high-precision brightness information, effectively improve the safety and comfort of night driving, reduce driving interference caused by light misjudgment, and realize intelligent and safe dynamic control of high and low beam headlights.

[0073] As an optional implementation, the image region includes a first image sub-region, a second image sub-region, and a third image sub-region. The first image region is partitioned according to the driving direction to obtain multiple image regions, including: determining a first image sub-region from the first image region based on the driving direction and vehicle width information; determining a second image sub-region from the first image region based on the driving direction and vehicle angle information, wherein the angle information is used to represent the diffusion degree of light emitted by the lighting device; and determining the image region in the first image region other than the first and second image sub-regions as the third image sub-region.

[0074] In this embodiment, during the partitioning of the first image region based on the driving direction, a first image sub-region can be determined from the first image region based on the driving direction and vehicle width information. A second image sub-region can be determined from the first image region based on the driving direction and vehicle angle information. The image region outside the first and second image sub-regions in the first image region can be determined as a third image sub-region. The aforementioned image regions may include the first, second, and third image sub-regions.

[0075] Optionally, the first image sub-region is a key area determined from the first image region based on the vehicle's driving direction and width information. For example, the first image sub-region can be a Level I key area. This first image sub-region can be located slightly below the center of the environmental image, directly corresponding to the road directly in front of the vehicle. It is the driver's focal point and a key area for detecting road conditions and obstacles ahead. The first image sub-region carries relatively important brightness information and is responsible for judging driving safety. The location of the second image sub-region is based on the driving direction and the vehicle's angle information, especially the diffusion degree of light from the lighting equipment. This second image sub-region can cover the area slightly to the side or slightly off-center from the vehicle's front, used to monitor lateral road conditions and possible changes in illumination. The third image sub-region can be the portion remaining after subtracting the first and second image sub-regions from the first image region. This third image sub-region can include the sky, ground, distant objects, etc., and has less impact on real-time high / low beam control decisions.

[0076] Optionally, the angle information can be acquired by the vehicle's angle sensor, representing the degree of diffusion and directional deviation of the light emitted by the lighting equipment relative to the vehicle's direction of travel. This angle information can influence the delineation of the position and size of the second image sub-region, determining the accuracy and range of the side-front environment monitoring. The width information can refer to the width of the vehicle itself, and the lateral extent of the first image sub-region in the image, determined based on the vehicle width. This width information is crucial for the accurate delineation of the first image sub-region, ensuring that it covers the main road width directly in front of the vehicle while eliminating the influence of edge distortion.

[0077] Optionally, during the process of determining the first image sub-region based on the driving direction and vehicle width information, basic vehicle parameters, including vehicle width information and real-time driving direction, can be collected. Combining the vehicle width and driving direction, the algorithm automatically calculates and delineates the first image sub-region, which directly faces the driving path directly in front of the vehicle. The first image sub-region is marked as a Level I priority region, assigned the highest weight, and used for priority brightness data processing.

[0078] Optionally, during the process of determining the second image sub-region based on driving direction and angle information, angle information from an angle sensor can be acquired, reflecting the diffusion degree of the lighting equipment's light and its position relative to the vehicle's driving direction. Within the first image region, a second image sub-region is delineated based on the angle information and driving direction. These regions are typically located to the side and front of the vehicle, capable of capturing brightness changes in side road conditions. The second image sub-region is considered a Level II priority area, with its brightness information having a lower weight than that of the first image sub-region, but still having a significant impact on the control strategy. The boundaries of the second image sub-region are marked and recorded for subsequent brightness analysis and weight application.

[0079] Optionally, in determining the third image sub-region, the already defined first and second image sub-regions are excluded from the first image region. The remaining region is defined as the third image sub-region, which typically includes the non-direct driving path portion of the first image region. The third image sub-region is marked as a non-I, non-II region, and its brightness information may be given lower weight in the control strategy or processed using a forgetting mechanism to reduce unnecessary data processing burden.

[0080] In this embodiment of the application, the above method can achieve fine-grained zoning processing of the vehicle's environmental image. Based on the importance of different areas and the vehicle's dynamic information, the control strategy of the lighting equipment can be determined, thereby ensuring the safety and comfort of driving the vehicle, while reducing unnecessary energy consumption and interference to other road users.

[0081] As an optional implementation, step S106 involves determining the weight information of multiple image regions, including: calling a weight classification model to determine the first weight information of a first image sub-region and the second weight information of a second image sub-region, wherein the weight classification model is obtained by training a recurrent neural network model, and the first weight information is greater than the second weight information.

[0082] In this embodiment, during the process of determining the weight information of multiple image regions, a weight classification model can be invoked to determine the first weight information of the first image sub-region and the second weight information of the second image sub-region. This weight classification model is a specific artificial intelligence model designed to automatically assign the importance (i.e., weight information) of each image sub-region in the control strategy based on the different characteristics of the first and second image sub-regions. In this embodiment, the weight classification model can be an LSTM (Laser-Based Memory Module) used to process and predict time-series data. The LSTM model has the ability to understand the brightness change trends in image sub-regions and can predict the future illumination state of image sub-regions based on historical data, providing data support for weight allocation. Through continuous learning and analysis, the weight classification model can adjust the weights of the first and second image sub-regions in real time according to environmental changes (such as the appearance of vehicles ahead, streetlights turning on, etc.) to reflect the degree of influence of the image sub-regions on driving safety under current illumination conditions.

[0083] Optionally, the first weight information can refer to the higher weight value assigned to the first image sub-region (Level I importance region) in the weight classification model. This first weight information can reflect the crucial role of the first image sub-region in the automatic high / low beam control strategy, as it directly corresponds to the main driving path directly in front of the vehicle. The second weight information can be the lower weight value assigned to the second image sub-region (Level II importance region) by the weight classification model. Although the second weight information is lower than the first weight information, it emphasizes the auxiliary role of side-front ambient light information in driving decisions.

[0084] Optionally, a pre-trained LSTM recurrent neural network model is loaded to ensure that the model has been fully learned for historical image sequences and corresponding weight assignments. The LSTM model includes an input layer, hidden layers (containing LSTM units), an output layer, and key forget, input, and output gate mechanisms to process time-series data. The latest environmental images are captured from the vehicle's camera. The images are segmented according to the vehicle's driving direction and width information, separating a first image sub-region (the key area directly in front) and a second image sub-region (the area to the side or away from the center).

[0085] Optionally, the brightness data of the first and second image sub-regions can be converted into a format suitable for model input, and normalization can be performed. Continuous image data is organized into a time series so that the LSTM model can understand and learn trends in brightness changes. The model receives current and historical image sequences and utilizes its long-term memory to predict the weight information of the first and second image sub-regions. Considering that ambient brightness changes with time and location, the LSTM model can dynamically adjust the weights, reflecting an immediate assessment of the importance of the current environment to the image sub-region.

[0086] Optionally, after analysis, the LSTM model outputs first weight information (higher weight) for the first image sub-region and second weight information (lower weight) for the second image sub-region. The model output weight information is confirmed to meet expectations, i.e., the first weight information is always greater than the second weight information, reflecting that the Level I priority area is more important for high / low beam headlight control decisions than the Level II priority area.

[0087] In this embodiment of the application, the above method can dynamically and intelligently adjust the weight of different image regions in the automatic high beam control strategy, so that the vehicle can respond more accurately and timely to environmental changes during driving, improve driving safety, and reduce interference to other road users.

[0088] As an optional implementation, step S106, determining the brightness information of multiple image regions, includes: identifying multiple pixel blocks in the image region to obtain sub-brightness information of the pixel blocks, wherein the sub-brightness information is used to represent the brightness of the environment in the pixel block; and determining the brightness information of the image region based on the sub-brightness information corresponding to the multiple pixel blocks respectively.

[0089] In this embodiment, during the process of determining the brightness information of multiple image regions, multiple pixel blocks within the image region can be identified separately to obtain sub-brightness information for each pixel block. The brightness information of the image region can be determined based on the sub-brightness information corresponding to each of the multiple pixel blocks. Here, a pixel block can be the basic unit within the image region; for example, a pixel block can refer to an 8x8 pixel matrix, also known as the smallest statistical unit, used to decompose and analyze the brightness attributes of the image region. The brightness of each pixel block can be measured individually to obtain sub-brightness information. Sub-brightness information can be a quantitative indicator of the brightness within a pixel block, reflecting the local ambient light intensity. Sub-brightness information is the measurement result of the brightness within a single pixel block, and can be the average brightness, maximum brightness, or other statistical quantities representing brightness for a single pixel block.

[0090] Optionally, the first image sub-region (Level I priority region) and the second image sub-region (Level II priority region) are divided into multiple pixel blocks of 8x8 pixels. For each pixel block, its sub-brightness information is calculated. This can be done by statistically analyzing the brightness values ​​of each pixel within the pixel block, such as by calculating the average, median, or other representative indicators. Pixel blocks can be dynamically merged as needed to expand the statistical area and improve sensitivity to larger light sources (such as streetlights or oncoming vehicle headlights). The merging operation is performed based on real-time environmental and lighting conditions to ensure the accuracy of the brightness analysis. The sub-brightness information of each pixel block within the same image sub-region is summarized, and the overall brightness information of the image sub-region is obtained through weighted averaging or other statistical methods.

[0091] Optionally, an LSTM-based weighted classification model is invoked to determine first and second weight information for the first and second image sub-regions, respectively, based on historical data and the current environment. The model's learning process considers the trend of pixel brightness changes to predict which regions are more critical for high / low beam control at the current moment. The obtained brightness information of the first and second image sub-regions is then weighted according to their respective weight information to obtain a weighted brightness value, which is used as the final brightness information input into the control strategy.

[0092] In this embodiment of the application, the above method enables refined processing of brightness information in environmental images. By utilizing statistical analysis of pixel blocks and sub-brightness information, combined with the weight classification capability of the LSTM model, a more accurate automatic high beam control strategy is generated, effectively improving the safety and comfort of vehicle driving.

[0093] As an optional implementation, step S108, determining the control strategy for the lighting device based on weight information and brightness information, includes: determining the number of environmental images in the environmental image set; performing weighted processing on the weight information and brightness information based on the number to obtain a weighted processing result; and determining the control strategy based on the weighted processing result.

[0094] In this embodiment, during the process of determining the control strategy for the lighting equipment based on weight information and brightness information, the number of environmental images in the environmental image set can be determined. Based on the number, the weight information and brightness information can be weighted to obtain a weighted processing result. Based on the weighted processing result, the control strategy is determined. The weighted processing result can be a comprehensive index that integrates the brightness information of different image regions in the environmental images and the importance of these image regions relative to the overall control strategy (i.e., weight information).

[0095] Optionally, calculate the number of images in the environmental image set; confirm whether the image set contains the latest N images to ensure the timeliness and accuracy of the analysis results. Set the initial values ​​of the weight information Wi (i=1,...,N) and the brightness information Li (i=1,...,N) (possibly based on the image acquisition time and importance); prepare the data structures required by the processing algorithm, such as a buffer to store the weighted processing results.

[0096] Optionally, a weighted processing result is calculated based on the weight information and brightness information to comprehensively reflect the current brightness state of the environment. For each environmental image i, its corresponding weight information Wi is multiplied by its brightness information Li to obtain a weighted brightness value Bi = Wi × Li. The weighted brightness values ​​Bi of each image are summarized, and the weighted processing result H is calculated by summation, averaging, or other statistical methods. Based on the weighted processing result H, an appropriate control strategy for the lighting equipment is automatically selected.

[0097] Optionally, a first weighted information threshold T1 and a second weighted information threshold T2 (T2>T1) are set to represent different brightness levels and control requirements. If H>T2, it indicates that the ambient brightness is sufficient and it is suitable to use low beams or reduce the lighting intensity. The control strategy is determined to be to turn off or reduce the high beams. If H≤T2 and H>T1, the ambient brightness is moderate. Based on the current strategy and historical data, it is decided whether to maintain the low beam state or slightly adjust the brightness. If H≤T1, the ambient brightness is low. The high beams can be turned on or enhanced to improve visibility. The control strategy is determined to be to turn on or enhance the high beams.

[0098] Optionally, a control strategy is sent to the vehicle's lighting control system to adjust the operating mode of the lighting equipment. Upon receiving the control strategy instruction, the lighting controller adjusts the state of the high beams (on, off, increased, or decreased) according to the instruction's requirements. The controller also records the executed actions and the current ambient brightness state for subsequent analysis and optimization.

[0099] In this embodiment, the method described above performs refined processing on the environmental image set, combining the weight and brightness information of each image to calculate a comprehensive weighted processing result, thereby intelligently determining the control strategy for the lighting equipment. This decision mechanism based on weight and brightness information makes automatic high / low beam control more dynamic and precise, adaptable to constantly changing driving environments, and improves the safety and efficiency of nighttime driving.

[0100] As an optional implementation, the weight information and brightness information are weighted based on the quantity to obtain a weighted processing result, including: in response to a quantity of one, the weight information and brightness information are weighted to obtain a weighted processing result.

[0101] In this embodiment, during the process of weighting the weight information and brightness information based on the quantity, if the quantity is one, the weight information and brightness information can be weighted to obtain a weighted processing result.

[0102] Optionally, by comprehensively analyzing the brightness information and its weight information of each image in the environmental image set, a weighted processing result reflecting the current environmental lighting conditions is obtained. The brightness information of the first image sub-region (Level I key region) and the second image sub-region (Level II key region) is obtained from the environmental image set, and is respectively denoted as B1 and B2. The weight classification model is called to obtain the weight information of the first image sub-region and the second image sub-region, denoted as W1 and W2, where W1 > W2, reflecting the higher influence of the Level I region on the control decision. For each environmental image i (i = 1...N), the weighted value Wi of the brightness information Bi of the environmental image is calculated Bi, which is carried out separately for the first image sub-region and the second image sub-region. The weighted brightness values of all images are aggregated to obtain the total weighted brightness value sum(Wi Bi), where i traverses all images in the environmental image set. If the size N of the image set exceeds the preset maximum processing amount (for example, 30 images), then the most recent 30 images are retained for weighted processing, and the remaining images are ignored to ensure processing efficiency and real-time performance.

[0103] Optionally, according to the weighted processing result, the control strategy of the lighting device, that is, the turning on or off of the high beam, is determined. A threshold T is set to distinguish between bright environments and low-light environments. T can be a fixed value or can be dynamically adjusted according to the actual environment and conditions. Compare the total weighted brightness value with the threshold T. If sum(Wi Bi) < T, it indicates that the current environment is more suitable for using the high beam to provide better road surface visibility. If sum(Wi Bi) >= T, it means that the environment is bright enough or there are high-brightness areas (such as street lights, oncoming vehicle lights). At this time, turning on the high beam may cause glare or interference, so the high beam should be maintained or switched to the low beam mode. According to the above comparison results, a control command to turn on the high beam or maintain / switch to the low beam is output.

[0104] In the embodiment of the present application, by collecting, analyzing, and processing the brightness information and weight information in the environmental image set, the control strategy of the lighting device is intelligently determined to adapt to the changing night driving environment. Through the deep learning ability of the LSTM model, the weight information can be dynamically adjusted, effectively distinguishing the influence degree of different image regions on the control of high and low beams, ensuring that the use of the lighting device can provide necessary visual assistance while minimizing interference to other road users.

[0105] As an optional implementation, the weight information and brightness information are weighted based on the quantity to obtain a weighted processing result, including: in response to the quantity being multiple, determining the third weight information of multiple environmental images in the environmental image set, wherein the third weight information is used to represent the degree of influence of the environmental image on driving safety; and weighting the third weight information, the weight information and the brightness information to obtain a weighted processing result.

[0106] In this embodiment, during the weighted processing of weight information and brightness information based on quantity, if there are multiple images, a third weight information can be determined for each of the multiple environmental images in the environmental image set. The third weight information, weight information, and brightness information can be weighted to obtain a weighted processing result. The aforementioned third weight information refers to a specific value assigned to each environmental image during the processing of the environmental image set, considering its generation time, location within its sub-region, and potential impact on current driving safety. This specific value can be used to reflect the importance of a particular environmental image to the automatic high / low beam control strategy, i.e., its influence on determining whether to turn on the high beams.

[0107] Optionally, collect all images (M in number) from the environmental image set. Ensure the image set covers images captured in the most recent time to reflect the latest environmental changes. Assign third weight information to each image in the environmental image set to reflect its potential impact on driving safety. Analyze the generation time of each image; the more recently generated the image, the higher its third weight information, because newer images more accurately reflect the current driving environment. Evaluate the impact on driving safety based on the content of the first image sub-region (Level I), the second image sub-region (Level II), and other regions in the image. If potential risk factors (such as pedestrians, animals, obstacles, etc.) are found, the third weight information of the corresponding image will increase. Consider the correlation between the image and the vehicle's driving direction, i.e., the degree to which the image shooting direction covers the vehicle's driving path; images with more direct coverage have higher third weight information. For each of the M images in the image set, calculate the corresponding third weight information Ti (i=1...M) to ensure that each image has a quantified indicator of its impact on driving safety.

[0108] Optionally, the brightness information of each image is obtained and weighted according to the third weight information, the first weight information, and the second weight information. Brightness information of the first image sub-region and the second image sub-region are extracted from each image, denoted as B1i and B2i (i=1...M), respectively. The first weight information W1 and the second weight information W2 (where W1>W2) are applied to initially weight the brightness information to obtain the weighted brightness information B1i. W1,B2i W2; Combining the third weight information Ti, the weighted brightness information is weighted again to form the final weighted brightness value WBi=(B1i W1+B2i W2) Ti, the above WBi can be used to represent the weighted brightness information of the i-th image.

[0109] Optionally, the weighted brightness values ​​of the environmental image set are aggregated to form a weighted processing result that comprehensively reflects the safety level of the current driving environment. The weighted brightness values ​​WBi of the M images are aggregated, and the weighted processing result R is obtained through summation, averaging, or other statistical methods; the formula for calculating R can be R=Σ(WBi) / M or R=Σ(WBi) / M. αi), where αi is a decay factor that takes into account the distance between images to ensure that the latest image contributes more; ensure that at least the latest 30 images are retained for weighted processing, and remove the earliest generated image in the order of generation to maintain the timeliness of the data.

[0110] As an optional implementation, in response to the number of multiple environmental images, the third weight information of multiple environmental images in the environmental image set is determined respectively, including: in response to the number of multiple environmental images, the third weight information corresponding to each environmental image is determined based on the acquisition time of each environmental image in the environmental image set, wherein the magnitude of the third weight information is negatively correlated with the time of acquisition.

[0111] In this embodiment, if there are multiple environmental images, the third weight information corresponding to each environmental image can be determined based on the acquisition time of each environmental image in the environmental image set.

[0112] Optionally, the environmental images to be processed and their acquisition times are organized to prepare for subsequent weight calculation. A set of environmental images is collected, assuming there are M images in the set. The acquisition time T1, T2, ..., TM of each image is recorded, forming a time series. The calculation of the third weight information based on the time series is clarified. The basic principle for calculating the third weight information is defined as follows: as the image acquisition time progresses, the third weight information decreases negatively, meaning the newest image has the highest weight, and the weights of earlier images decrease sequentially.

[0113] Optionally, a third weight information is assigned to each environmental image to reflect its influence on the current decision. A weight decay function is defined, which describes the relationship between the weight value and the time over time. Common function forms can be exponential decay, linear decay, etc., depending on the system design requirements. Based on the acquisition time series, the weight decay function is applied to each environmental image to calculate the corresponding third weight information W3i (i=1...M). For example, the latest image (acquisition time T1) has the largest W3i, while the earliest image (acquisition time TM) has the smallest W3i. The brightness information of the environmental images is initially weighted according to the first and second weight information. The brightness information of the first image sub-region (Level I) and the second image sub-region (Level II) in each environmental image are extracted respectively, denoted as B1i and B2i (i=1...M); according to the first weight information W1 and the second weight information W2, B1i and B2i are weighted respectively to form the preliminary weighted brightness information B1i. W1 and B2i W2.

[0114] Optionally, the preliminary weighted brightness information is further weighted by incorporating third weighting information to reflect the dynamic impact of time on decision-making. For each environmental image, its preliminary weighted brightness information B1i is... W1 and B2i W2 is multiplied by the third weight information W3i to generate the final weighted brightness value WB1i=B1i. W1 W3i and WB2i=B2i W2 W3i. In this way, the newest image (with the highest W3i) contributes the most to the weighted brightness value, while the contribution of earlier images decreases as W3i decreases.

[0115] Optionally, the weighted brightness values ​​of each environmental image are aggregated to provide a data basis for the formulation of control strategies. The weighted brightness values ​​of each image are aggregated, including the weighted brightness values ​​of the first image sub-region and the second image sub-region, to form a total weighted brightness value. To avoid data redundancy, only the weighted brightness values ​​of the most recent 30 images are retained for aggregation, and earlier images exceeding the number limit are discarded.

[0116] In this embodiment, the method described above illustrates how to calculate the third weight information based on the acquisition time of the environmental image, and then combine it with brightness information, first weight information, and second weight information for secondary weighting processing to finally generate a lighting equipment control strategy. By dynamically adjusting the weights, the impact of recent environmental changes on driving safety can be assessed more accurately, thereby making more intelligent high and low beam control decisions.

[0117] As an optional implementation, determining a control strategy based on the weighted processing result includes: evaluating the weighted processing result of environmental images in the environmental image set to obtain an evaluation result, wherein the evaluation result is used to represent the brightness state of the environment; determining a first control strategy in response to the evaluation result being greater than a first evaluation result threshold; and determining a second control strategy in response to the evaluation result being greater than a second evaluation result threshold, wherein the brightness state corresponding to the second evaluation result threshold is greater than the brightness state corresponding to the first evaluation result threshold, and the brightness of the environment increased according to the second control strategy is higher than the brightness of the environment increased according to the first control strategy.

[0118] In this embodiment, during the process of determining the control strategy based on the weighted processing results, the weighted processing results of the environmental images in the environmental image set can be evaluated to obtain an evaluation result. If the evaluation result is greater than a first evaluation result threshold, the control strategy can be determined as the first control strategy. Conversely, if the evaluation result is greater than a second evaluation result threshold, the control strategy can be determined as the second control strategy. The first evaluation result threshold can be a brightness index boundary used to distinguish between "low light" or "dark" environments and "normal light" environments. Under the aforementioned first evaluation result threshold, the current ambient brightness can be considered low, insufficient for safe driving, and more lighting assistance can be provided. For example, the aforementioned first evaluation result threshold can be greater than 50% in a dark environment, meaning that if the weighted processing results of the environmental image set show that more than half of the image area is under low light conditions, it will be considered a condition for triggering the first control strategy. The second evaluation result threshold is another brightness index boundary used to distinguish between "normal light" environments and "high light" or "bright" environments. The aforementioned second evaluation result threshold is set higher than the first evaluation result threshold so that different measures can be taken when the ambient brightness is high. For example, the threshold for the second evaluation result mentioned above could be that the arbitration result is greater than 50% in a bright environment. This could mean that the weighted processing result is evaluated by the arbitration mechanism, and if more than half of the image area shows that the ambient brightness is sufficient or even too high, it will be considered as a condition to trigger the second control strategy.

[0119] Optionally, the first control strategy can be a series of lighting equipment control actions taken when the weighted processing result is lower than the first evaluation result threshold, indicating that the environment is in a "dark" state. This ensures sufficient illumination under low-light conditions, improving driving safety at night or in adverse weather conditions. For example, the first control strategy could be to activate the high beams to increase the illumination range and brightness to compensate for the low-light environment. The second control strategy can be lighting equipment control actions taken when the weighted processing result is higher than the second evaluation result threshold, i.e., the environment is considered too bright (such as city streets, oncoming traffic, etc.). This avoids excessive illumination, reduces glare interference to other road users (such as drivers of oncoming vehicles), and saves energy. For example, the second control strategy could be to turn off the high beams and switch to low beam mode to reduce direct light and avoid glare.

[0120] Optionally, based on the weighted processing of the environmental images in the previous stage, an overall weighted processing result reflecting the current environmental brightness state is obtained. The weighted processing results of each environmental image in the environmental image set are summarized to form an overall weighted processing result H. The calculation of H considers the different weighted brightness information and third weight information of each image, reflecting the true state of ambient lighting and the degree of potential impact on driving safety. By comparing the weighted processing result H with the preset evaluation result threshold, an evaluation result representing the environmental brightness state is generated. A first evaluation result threshold T1 and a second evaluation result threshold T2 are set, where T2>T1 represents different brightness states; if H>T1 but does not reach T2, an evaluation result R1 is generated indicating that the current environment is in a "low light" or "dark" state; if H>T2, an evaluation result R2 is generated indicating that the current environment is in a "high light" or "bright" state; if H≤T1, an evaluation result R0 is generated indicating that the ambient brightness is moderate or above and no special lighting measures are required.

[0121] Optionally, an appropriate control strategy can be automatically selected based on the evaluation results to adjust the operating mode of the lighting equipment. The first control strategy is triggered when the ambient brightness is deemed insufficient, requiring enhanced lighting, such as turning on high beams or increasing the brightness output of the lighting equipment to improve the driver's visibility. The second control strategy is triggered when the ambient brightness is determined to be high, potentially posing a risk of glare or energy waste; in this case, the lighting should be switched to low beams or the brightness output reduced to avoid interfering with other road users. Alternatively, the current lighting status can be maintained, or fine-tuned according to the system's default settings to ensure that the lighting is neither excessive nor insufficient.

[0122] In this embodiment, the method distinguishes the brightness state of the environment by setting different evaluation result thresholds, and automatically selects an appropriate lighting equipment control strategy based on the state. This provides necessary lighting assistance in low-light environments while avoiding the adverse effects of excessive lighting in high-light environments. This mechanism ensures the efficient and intelligent operation of the lighting equipment while the vehicle is in motion, improving driving safety and comfort.

[0123] The following uses high and low beam lamps as an example to illustrate the technical solution of the embodiments of this application.

[0124] Currently, the automatic high beam function can help drivers use high and low beam headlights appropriately. By recognizing the surrounding road environment through a camera and combining it with the vehicle's signals, it can monitor ambient brightness, road lighting conditions, traffic participants, and vehicle speed in real time. In the case of no streetlights at night, it can automatically request the headlight control to turn on the high beam headlights, and in the case of meeting oncoming traffic or overtaking, it can automatically request the headlight control to turn off the high beam headlights.

[0125] The current sensing solution controls the output of signals by statistically analyzing the light intensity within the camera area. When the calculated brightness value of the area exceeds a certain value for a certain period, the system recommends using low beam lighting; when the calculated brightness value exceeds a certain value but falls below a certain value for a certain period, the system recommends using high beam lighting. However, due to camera hardware and light sensitivity issues, there is a certain degree of false recommendations.

[0126] The methods of the embodiments of this application will be further illustrated below.

[0127] Figure 2 This is a schematic diagram of an automatic high / low beam control architecture according to an embodiment of this application, as shown below. Figure 2As shown, the architecture may include a gateway interface 21, an angle sensor 22, a high / low beam headlight controller 23, an instrument panel display 24, a braking system 25, and a lighting controller 26. The gateway interface 21 coordinates and manages data communication from various subsystems within the vehicle, including but not limited to receiving and sending data to the angle sensor 22, the high / low beam headlight controller 23, the instrument panel display 24, the braking system 25, and the lighting controller 26. The angle sensor 22 detects changes in the vehicle's steering wheel angle, wheel angle, or vehicle posture. In automatic high / low beam control, the angle sensor helps the system understand the vehicle's direction of travel and possible steering maneuvers, which is used to predict potential road users and road lighting conditions ahead. For example, when the vehicle is turning, the angle sensor data can be used to adjust the high beam's illumination range to avoid glare to vehicles in the side lanes. The high / low beam headlight controller 23 calculates and outputs high beam control commands based on various received signals (such as ambient brightness, vehicle status, and road conditions ahead). The instrument panel display 24 provides a visual interface between the driver and the vehicle control system. The large instrument cluster displays key driving information, including but not limited to vehicle speed, engine status, fuel level, and warning signals. The automatic high / low beam control system can also display the current lighting mode (high beam / low beam), ambient light information, and other driver assistance prompts to help the driver understand the system's operating status.

[0128] like Figure 2 As shown, the braking system 25 is used for vehicle deceleration and stopping operations, and may include mechanical and electronic braking systems. Although the braking system's direct involvement in lighting control is limited, its integration with the automatic high / low beam control architecture provides an additional layer of safety. For example, when the vehicle brakes suddenly, the high / low beam controller can automatically turn off the high beams to avoid interfering with other vehicles in an emergency. The lighting controller 26 directly controls the switching, brightness adjustment, and beam angle adjustment of the vehicle's front and rear lighting equipment. It receives instructions from the high / low beam controller and actually executes the turning on, turning off, or brightness adjustment of the high beams. The lighting controller is closely connected to the vehicle's electrical system, ensuring rapid and stable response of the lighting equipment, and is the execution end of the automatic high / low beam control architecture.

[0129] In this embodiment, after IGN ON, the system switch is turned on, the headlights are in automatic mode, and after the lighting control logic passes the judgment, the camera detects the vehicle status, surrounding environment, and road conditions ahead. If the conditions for automatic high beam activation are met, the system requests to turn on the high beams; if the system detects that following vehicles, oncoming vehicles, or the vehicle-related environment (including the presence of multiple streetlights, whether the ambient brightness exceeds a threshold, etc.) do not meet the conditions for automatic high beam activation, the system requests to turn off the high beams. Once the system activation conditions are restored, the system will follow a certain delay mechanism to issue a high beam request without interfering with other road users. The request to switch between high and low beams is transmitted from the controller to the lighting control system via a control signal, and the driver can change the lighting status at any time via the light lever.

[0130] Figure 3 This is a schematic diagram of an image statistical region according to an embodiment of this application, such as... Figure 3 As shown, due to camera distortion and the influence of lighting, different weights are used to calculate brightness for different areas of the image. For Level I key statistical areas, a larger weight is used for statistical calculation, while for Level II secondary key areas, a smaller weight is used for statistical calculation. Areas not included in the statistical area are forgotten and ignored.

[0131] Optionally, the effective detection area is a rectangular region covering 60% to 70% of the center of the image (excluding edge distortion). The smallest statistical unit is an 8x8 pixel block (supporting dynamic merging). The area used by the camera to calculate streetlight brightness is generally located directly in front of vehicles, divided into Level I key areas, Level II key areas, and areas not calculated, referencing image brightness calculation methods. And utilizing... Figure 2 The weighting coefficients for brightness calculation based on the regional relationships in the image statistical region map are configured as follows: for key areas of Class I brightness calculation, a larger weighting coefficient is applied; for key areas of Class II brightness calculation, a smaller weighting coefficient is applied; and for areas that are not Class I or Class II, they are forgotten and are not included in the brightness calculation. This is used to determine whether a scene is dark and to calculate an output result.

[0132] Figure 4 This is a schematic diagram of a Long Short-Term Memory (LSTM) network model according to an embodiment of this application, as shown below. Figure 4As shown, recurrent neural networks (RNNs) all have a chain of repeating modules. In a standard RNN, this repeating module has a very simple structure, such as a single tanh layer. LSTM also has the above chain structure, but its repeating units differ from those in standard RNNs, which have only one network layer; it has four network layers internally. Long Short-Term Memory (LSTM) is a type of temporal recurrent neural network specifically designed to address the long-term dependency problem inherent in general RNNs. Each RNN has a chain of repeating neural network modules. LSTM consists of four core components: forget gate, input gate, candidate memory, and output gate. The σ function determines which information should be forgotten and which should be retained in the cell state. An output value close to 1 indicates that information is retained, and close to 0 indicates that information is forgotten. Similarly, the σ function determines which new information should be added to the cell state. Simultaneously, another different activation function (e.g., tanh) is used to generate candidate cell state values, which are then multiplied by the output of the σ gate to update the cell state. The σ function determines which information in the cell state should be output to the hidden state of the next time step.

[0133] Figure 5 This is a schematic diagram of a statistical-based automatic high / low beam headlight control recommendation strategy process according to an embodiment of this application, such as... Figure 5 As shown, the long and short gates and forgetting mechanism of LSTM are used to calculate different weights for data generated at different times. Following a time-series analysis, similarly, when calculating the illumination of images generated by the camera, different calculation strategies are selected for different regions. For some edge areas, a forgetting gate-like approach similar to LSTM is used to discard the calculation. For newly generated images (e.g., data b30 is a newly generated image compared to data b2), the closer it is to the latest scene, the larger the influence factor for calculating the high and low beam control should be, thus a larger weight is used. Conversely, the earlier the image is generated, the greater the difference between the calculated camera statistical region and the latest scene, thus the smaller the calculation factor should be. By setting weights for camera statistical regions generated at different times (e.g., coefficients a1, a2, ..., a30, etc.), a statistical method is used to calculate the recommended values ​​over a period of time. Through arbitration, a comprehensive calculation of the high and low beam recommendations is performed.

[0134] like Figure 5As shown, for the calculation of photosensitivity of newly input images, the brightness intensity of newly input images is calculated with a larger weight. For previous input images, the weight coefficient gradually decreases over time to statistically determine brightness. Each output process processes information from no more than 30 images and outputs the corresponding results. For more than 30 images, the earliest generated image is removed according to the generation order, and the newly generated image is added to the front of the calculation queue and given a larger weight coefficient. The output results of 31 times are statistically analyzed and arbitration is performed. When the arbitration result is greater than 50% in a dark environment, the recommended output result is high beam headlights; when the arbitration result is greater than 50% in a bright environment, the recommended output result is low beam headlights.

[0135] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Furthermore, the collection, use and processing of the relevant data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0136] According to another aspect of the embodiments of this application, corresponding to the embodiments of the above-described vehicle lighting equipment control method, this specification also provides a vehicle lighting equipment control device.

[0137] Figure 6 This is a schematic diagram of a vehicle lighting equipment control device according to an embodiment of this application, as shown below. Figure 6 As shown, the vehicle lighting control device 60 may include: an acquisition module 602, a processing module 604, a first determination module 606, a second determination module 608, and a control module 610. The acquisition module 602 is used to acquire a set of environmental images of the vehicle in response to a control command triggered by the lighting device; the processing module 604 is used to partition at least one environmental image in the environmental image set based on the vehicle's driving direction to obtain multiple image regions; the first determination module 606 is used to determine the weight information and brightness information of each of the multiple image regions; the second determination module 608 is used to determine the control strategy of the lighting device based on the weight information and brightness information; and the control module 610 is used to control the lighting device according to the control strategy.

[0138] Embodiments of this application also provide a vehicle, including: a memory storing an executable program; and a processor for running the program, wherein the program executes the methods described in various embodiments of this application when it runs.

[0139] Embodiments of this application also provide a computer-readable storage medium including a stored executable program, wherein, when the executable program is running, it controls the device where the computer-readable storage medium is located to perform the methods of various embodiments of this application.

[0140] Embodiments of this application also provide a computer program product, including a computer program that, when executed by a processor, implements the methods of various embodiments of this application.

[0141] Embodiments of this application also provide a computer program product, including a non-volatile computer-readable storage medium for storing a computer program that, when executed by a processor, implements the methods in various embodiments of this application.

[0142] Embodiments of this application also provide a computer program that, when executed by a processor, implements the methods described in the various embodiments of this application.

[0143] 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.

[0144] 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 instance, 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 coupling, direct coupling, or communication connection may be through some interfaces; the indirect coupling or communication connection between units or modules may be electrical or other forms.

[0145] 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.

[0146] 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.

[0147] 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 described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, read-only memory (ROM), random access memory (RAM), portable hard drives, magnetic disks, or optical disks.

[0148] The above description is only a preferred embodiment of this application. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the principle of this application, and these improvements and modifications should also be considered within the scope of protection of this application.

Claims

1. A method for controlling lighting equipment in a vehicle, characterized in that, include: In response to a control command triggered by the lighting device, an environmental image set of the vehicle is acquired, wherein the environmental images in the environmental image set are used to represent the brightness of the environment in which the vehicle is located; Based on the vehicle's driving direction, at least one environmental image in the environmental image set is partitioned to obtain multiple image regions; Weight information for multiple image regions is determined, and brightness information for multiple image regions is determined, wherein the weight information is used to represent the degree to which the brightness of the image region affects the driving safety of the vehicle; Based on the weight information and the brightness information, a control strategy for the lighting device is determined, wherein the control strategy is used to represent the rules for controlling the lighting device; The lighting equipment is controlled according to the control strategy, wherein the controlled lighting equipment is used to increase the brightness of the environment.

2. The method according to claim 1, characterized in that, Based on the vehicle's driving direction, at least one environmental image in the environmental image set is partitioned to obtain multiple image regions, including: A first image region is extracted from the environmental image, wherein the distortion degree of the first image region is less than the distortion degree of other image regions in the environmental image besides the first image region; According to the driving direction, the first image region is divided into multiple image regions.

3. The method according to claim 2, characterized in that, The image region includes a first image sub-region, a second image sub-region, and a third image sub-region. The first image region is divided into multiple image regions according to the driving direction, including: Based on the driving direction and the width information of the vehicle, the first image sub-region is determined from the first image region; Based on the driving direction and the vehicle's angle information, a second image sub-region is determined from the first image region, wherein the angle information is used to represent the degree of diffusion of the light emitted by the lighting device; The image region other than the first image sub-region and the second image sub-region in the first image region is determined as the third image sub-region.

4. The method according to claim 3, characterized in that, Determine the weight information of each of the image regions, including: The weight classification model is invoked to determine the first weight information of the first image sub-region and the second weight information of the second image sub-region, wherein the weight classification model is obtained by training a recurrent neural network model, and the first weight information is greater than the second weight information. or, Determine the brightness information of multiple image regions respectively, including: Multiple pixel blocks in the image region are identified respectively to obtain sub-brightness information of the pixel blocks, wherein the sub-brightness information is used to represent the brightness of the environment in the pixel block; The brightness information of the image region is determined based on the sub-brightness information corresponding to multiple pixel blocks.

5. The method according to any one of claims 1 to 4, characterized in that, Based on the weight information and the brightness information, a control strategy for the lighting device is determined, including: Determine the number of environmental images in the environmental image set; Based on the quantity, the weight information and the brightness information are weighted to obtain the weighted processing result; Based on the weighted processing results, the control strategy is determined.

6. The method according to claim 5, characterized in that, Based on the stated quantity, the weight information and the brightness information are weighted to obtain a weighted processing result, including: In response to the quantity being one, the weight information and the brightness information are weighted to obtain the weighted processing result; or, Based on the stated quantity, the weight information and the brightness information are weighted to obtain a weighted processing result, including: In response to the number being multiple, third weight information of multiple environmental images in the environmental image set is determined respectively, wherein the third weight information is used to represent the degree of influence of the environmental images on the driving safety. The third weight information, the weight information, and the brightness information are weighted to obtain the weighted processing result.

7. The method according to claim 6, characterized in that, In response to the number being multiple, third weight information for multiple environmental images in the environmental image set is determined, including: In response to the number being multiple, based on the acquisition time corresponding to each of the multiple environmental images in the environmental image set, the third weight information corresponding to each environmental image is determined, wherein the magnitude of the third weight information is negatively correlated with the time of acquisition.

8. The method according to claim 5, characterized in that, Based on the weighted processing result, the control strategy is determined, including: The weighted processing results of the environmental images in the environmental image set are evaluated to obtain an evaluation result, wherein the evaluation result is used to represent the brightness state of the environment; In response to the evaluation result being greater than the first evaluation result threshold, the control strategy is determined to be the first control strategy; In response to the evaluation result being greater than a second evaluation result threshold, the control strategy is determined to be a second control strategy, wherein the brightness state corresponding to the second evaluation result threshold is greater than the brightness state corresponding to the first evaluation result threshold, and the brightness of the environment increased according to the second control strategy is higher than the brightness of the environment increased according to the first control strategy.

9. A vehicle, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

10. An electronic device, characterized in that, include: Memory, which stores executable programs; A processor for running the program, wherein the program, when running, performs the method according to any one of claims 1 to 8.

Images