Vehicle fog lamp control method, device and equipment and computer readable storage medium

Abstract

The invention relates to the field of vehicle control, in particular to a vehicle fog lamp control method, device and equipment and a computer readable storage medium. The method comprises the steps of obtaining a road image in front of a vehicle; performing lane line detection on the road image to obtain position information of a lane line vertex in the road image; the lane line vertex comprises a point farthest from the vehicle in a lane line area detected in the road image; and according to the position information, fog lamps of the vehicle are controlled. According to the invention, the fog lamp control efficiency and reliability can be improved, so that the driving safety guarantee degree is improved.

Description

Technical Field

[0001] The present application relates to the field of vehicle control technology, and in particular to a vehicle fog light control method, device, equipment and computer-readable storage medium. Background Art

[0002] In the field of automobile safety, the use of fog lights is crucial to improving driving safety in adverse weather conditions. Traditional fog light systems usually require drivers to manually turn them on or off according to weather conditions, which not only increases the driver's operating burden, but may also lead to failure to use fog lights in time when needed, thereby increasing the risk of traffic accidents. With the development of technology, automated and intelligent automotive lighting systems have gradually become a research hotspot. Against this background, a variety of automatic fog light control systems have emerged, aiming to achieve automatic control of fog lights through sensors, image processing and intelligent control technologies.

[0003] In the related art, the image captured by the vehicle camera is used to determine whether it is foggy, and thus determine whether the fog lights need to be turned on. This solution can distinguish between different conditions of fog and rain by calculating the image blur value and determining the fog state. However, artificially designed visual features such as blurriness and foggy state are difficult to distinguish between fog and rain, and a single value is difficult to meet complex environmental conditions and affects accuracy.

[0004] Therefore, the existing fog lamp control solution has the problem of insufficient guarantee for driving safety. Summary of the invention

[0005] In view of this, the embodiments of the present application provide a vehicle fog light control method, device, equipment and computer-readable storage medium, which can solve the problem that the existing fog light control scheme is insufficient to ensure driving safety.

[0006] In a first aspect, an embodiment of the present application provides a vehicle fog lamp control method, the vehicle fog lamp control method comprising: Acquire a road image in front of the vehicle; Performing lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points in the lane line area detected in the road image that are farthest from the vehicle; The fog lights of the vehicle are controlled according to the position information.

[0007] By adopting the above technical solution, by directly detecting the position information of the lane line vertex in the road image, the farthest road range that the driver can see can be represented based on the point farthest from the vehicle that can be detected in the road image. Therefore, the vehicle fog lights are automatically controlled according to the position information of the detected lane line vertex. Specifically, when the distance between the lane line vertex and the vehicle is less than a preset visibility threshold (such as 200 meters), the fog lights are controlled to turn on. The embodiment of the present invention only needs to detect the position of the lane line vertex, and there is no need to manually design the image features that trigger the fog lights to turn on. While improving the efficiency of road image processing, it can also avoid errors introduced by artificially designed features. Therefore, the embodiment of the present invention can improve the efficiency and reliability of fog light control, thereby improving the degree of protection for driving safety.

[0008] In some embodiments, the method further comprises: determining road visibility based on the location information; The fog lamp is controlled according to a comparison result between the visibility and a preset visibility threshold.

[0009] In some embodiments, the road image corresponds to an image coordinate system; the road corresponds to a world coordinate system; the position information includes a first coordinate of the lane line vertex in the image coordinate system; and the method further includes: Mapping the first coordinates according to a mapping relationship between the image coordinate system and the world coordinate system to obtain second coordinates; The visibility is determined based on the second coordinate.

[0010] In some embodiments, the method further comprises: Marking a reference area of ​​the road image according to a preset visibility threshold; the reference area is used to represent an invisible area of ​​the road to the vehicle if the visibility of the road is less than the visibility threshold; The fog lamp is controlled according to a relative position relationship between the position information and the reference area.

[0011] In some embodiments, the method further comprises: According to the relative position relationship, determining whether the lane line vertex is located in the reference area; If the vertex of the lane line is outside the reference area, controlling the fog lamp to be turned off; If the lane line vertex is located in the reference area, the fog lamp is controlled to turn on.

[0012] In some embodiments, the road image includes multiple image edges; the lane line vertices include pixel points farthest from the target image edge among the multiple pixel points; and the target edge is the image edge closest to the vehicle among the multiple image edges.

[0013] In some embodiments, the location information is obtained by detecting a target model pre-deployed on the vehicle; the method further includes: Get the trained lane detection model; Compressing model parameters of the lane detection model to obtain a compressed model; Performing model format conversion on the compressed model according to the vehicle environment corresponding to the vehicle to obtain a converted model; The converted model is quantized after training to obtain the target model.

[0014] In a second aspect, an embodiment of the present application further provides a vehicle fog lamp control device, comprising: An acquisition module, used for acquiring a road image in front of the vehicle; A detection module, configured to perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points in the lane line area detected in the road image that are farthest from the vehicle; A control module is used to control the fog lights of the vehicle according to the position information.

[0015] In a third aspect, an embodiment of the present application further provides a vehicle fog light control device, the vehicle fog light control device comprising a processor and a memory, the memory being used to store instructions, the processor being used to call the instructions in the memory, so that the vehicle fog light control device executes the vehicle fog light control method described in the first aspect.

[0016] In a fourth aspect, an embodiment of the present application further provides a computer-readable storage medium, wherein the computer-readable storage medium stores computer instructions. When the computer instructions are executed on a processor, the processor executes the vehicle fog light control method as described in the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to an embodiment of the present application.

[0018] Figure 2 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to another embodiment of the present application.

[0019] Figure 3 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to another embodiment of the present application.

[0020] Figure 4 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to another embodiment of the present application.

[0021] Figure 5 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to another embodiment of the present application.

[0022] Figure 6 The present invention is a flowchart of the steps of a vehicle fog lamp control method provided according to another embodiment of the present application.

[0023] Figure 7 The figure is a schematic diagram of the structure of a vehicle fog lamp control device provided according to an embodiment of the present application.

[0024] Figure 8 The present invention is a schematic diagram of the structure of a vehicle fog lamp control device provided according to an embodiment of the present application. DETAILED DESCRIPTION

[0025] In order to more clearly understand the above-mentioned purposes, features and advantages of the present application, the present application is described in detail below in conjunction with the accompanying drawings and specific implementation methods. It should be noted that the implementation methods of the present application and the features in the implementation methods can be combined with each other without conflict.

[0026] In the following description, many specific details are set forth to facilitate a full understanding of the present application. The described implementations are only part of the implementations of the present application, rather than all the implementations.

[0027] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which this application belongs. The terms used herein in the specification of this application are only for the purpose of describing specific embodiments and are not intended to limit this application.

[0028] It should be further noted that, in this article, the terms "comprises", "includes" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device including the element.

[0029] In this application, "at least one" means one or more, and "more" means two or more than two. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone, where A and B can be singular or plural. The terms "first", "second", "third", "fourth", etc. (if any) in the specification, claims and drawings of this application are used to distinguish similar objects, rather than to describe a specific order or sequence.

[0030] In the embodiments of the present application, words such as "exemplary" or "for example" are used to indicate examples, illustrations or descriptions. Any embodiment or design described as "exemplary" or "for example" in the embodiments of the present application should not be interpreted as being more preferred or more advantageous than other embodiments or designs. Specifically, the use of words such as "exemplary" or "for example" is intended to present related concepts in a specific way.

[0031] The traditional fog lamp system is manually turned on or off by the driver according to weather conditions, which not only increases the driver's operating burden, but also may lead to the failure to use the fog lamp in time when needed, thereby increasing the risk of traffic accidents. Based on this, in the related technology, it is proposed to automatically control the fog lamp to ensure driving safety more efficiently.

[0032] In the related art, it is possible to determine whether the vehicle is foggy through images captured by the vehicle-mounted camera, and control the fog lights to automatically turn on when the vehicle is foggy. The specific image processing method includes calculating the blur value of the image, and when the value is within a preset range and the wipers are not working, it is determined to be foggy. Optionally, the solution may also include image blur value calculation means and fog state determination means, which can distinguish different conditions of fog and rain.

[0033] However, artificially designed visual features such as blur and fog are difficult to distinguish between fog and rain, and a single value is difficult to meet complex environmental conditions, affecting the accuracy of rainy and foggy detection, which in turn affects the accuracy of fog light control and the degree of driving safety.

[0034] Optionally, a V2X (Vehicle to Everything)-based fog light automatic control method can be used to obtain weather information through real-time interconnection between the edge computing center and the meteorological observation station. When weather changes are detected, the weather changes are broadcast through the V2X system. After receiving the weather change information, the on-board unit parses and analyzes it, and finally sends the results to the vehicle control system, which controls the switch of the fog lights based on the analysis results.

[0035] However, this solution relies on the widespread deployment of V2X communication infrastructure such as RSU (Road Side Unit), which may not be popular in some areas. In addition, V2X communication may be affected by signal interference and delay, which may affect the response speed and accuracy of the system, thereby affecting the degree of driving safety.

[0036] Optionally, an automobile headlight system with integrated automatic fog lights may be used, the automobile headlight system including a headlight panel, a camera, an image processor, a controller and a light action execution device. The camera in the automobile headlight system collects environmental information, such as rain, fog and light intensity, and the controller automatically controls the switch of the fog lights, low beam lights and high beam lights according to the information, thereby improving driving safety and reducing the operating burden of the driver.

[0037] However, the integrated LED light group and control system in this solution may increase the production cost of the headlights, and the integrated design may make maintenance and replacement more complicated and expensive. In addition, the response of the automobile headlight system to environmental changes may need to be adjusted and optimized according to the specific use environment, and the modification cost is high. In addition, the performance limitations of the camera and image processor may affect the accuracy and reliability of the system, thereby affecting the degree of driving safety.

[0038] It can be seen from the above that the related fog lamp control has problems such as insufficient accuracy, high cost, and low reliability, which leads to insufficient driving safety guarantees.

[0039] In view of this, the embodiments of the present application provide a vehicle fog light control method, device, equipment and computer-readable storage medium, which can improve the reliability and efficiency of fog light control and enhance driving safety, and are particularly suitable for fog light control in environments with low visibility such as rainy, foggy and dusty days.

[0040] See also Figure 1 , which is a flowchart of the steps of the vehicle fog light control method provided by an embodiment of the present application. According to different requirements, the order of the steps in the flowchart can be changed, and some steps can be omitted. This method embodiment can be executed based on a preset electronic device, which has certain data storage, data processing and communication capabilities. Specifically, the electronic device may include a controller, which can communicate with a camera on the vehicle and sensors in the road environment, and the embodiment of the present invention is not limited to this.

[0041] See also Figure 1 As shown, the vehicle fog lamp control method may include the following steps: Step 101: Acquire a road image in front of the vehicle.

[0042] Specifically, a road image can be collected by a preset camera on the vehicle, and the electronic device can communicate with the camera to obtain the road image in front of the vehicle. The road image represents an image of the field of view that the driver of the vehicle can see. Specifically, the preset camera can be the front camera of the vehicle. Based on the setting position and angle of the front camera, the road surface in front of the vehicle is photographed by the front camera to obtain a road image.

[0043] Step 102: Perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image.

[0044] Among them, lane line detection is performed on the road image to obtain the lane line area included in the road image. The lane line area is the area corresponding to the lane line contour, and the pixels in the lane line area constitute the lane line contour. The point farthest from the vehicle among all the pixels in the lane line area is the lane line vertex. It can be understood that the road image intuitively represents the driver's field of vision, and when the lane line in the road image is blocked due to rain, fog, dust and other reasons, the lane line contour of the blocked area cannot be detected in the road image. Therefore, the point farthest from the vehicle that can be detected in the road image can represent the farthest position that the driver can currently see on the road ahead. Therefore, in an embodiment of the present invention, the current maximum visual range of the driver of the vehicle is determined based on the position information of the lane line vertex, so as to facilitate the subsequent control of the fog lights according to the maximum visual range, thereby ensuring driving safety under low visibility conditions.

[0045] It should be noted that, considering that there may be some outliers in the lane line area, or the fog, dust, etc. in the current driving environment may block the lane line in front of the road to varying degrees, in order to more accurately characterize the current visibility conditions for driving safety through the lane line vertices, the positions of the preset pixel points can be arranged in descending order according to the distance from the vehicle, and the average or median can be calculated to obtain the position information of the lane line vertices.

[0046] Specifically, lane line detection can be performed on a road image according to a preset lane line training model. The lane line detection model can adopt a LaneNet model, which is a deep learning model. The processing process of the lane line detection model for the input image can include the following: first, the RGB (Red Green Blue, red, green and blue primary colors) image captured by the vehicle's forward camera is input into the lane line detection model, and the input RGB image is extracted through the backbone network in the lane line detection model. The backbone network can be a convolutional neural network, such as MobileNet. For each anchor point in the image, local features are extracted according to its projection position on the feature map. The anchor point is modeled as a line on the image plane, and each anchor point contains starting point and angle information. Then, based on the attention mechanism of the anchor point, the global information is aggregated. The mechanism processes the local feature vector through a fully connected function, outputs a weight for each anchor point, and then generates a global feature vector by combining the weight and local features. Subsequently, for each anchor point, a classification probability (True or False, used to characterize whether it belongs to the lane line boundary), offset and lane length are given to achieve lane line prediction classification and regression. Among them, classification and regression prediction can be completed by two fully connected functions respectively. Finally, the backbone network outputs the lane boundary line, that is, the lane detection result. Among them, in the training and testing stages of the lane detection model, non-maximum suppression (NMS) based on lane distance measurement can be used to reduce the number of false positives, so as to retain only the boxes that are most likely to contain the target object and obtain the best detection results.

[0047] Correspondingly, the training process of the lane detection model is explained as follows: First, the sample preparation stage: real vehicle data collection images are used as training samples. Specifically, the camera installed on the car can be used to collect driving data in different severe weather conditions, including: heavy fog, heavy snow, heavy rain and sandstorms. These images will be used for subsequent model training and verification. In addition, the density and distance of obstructions such as fog and smoke in front of the vehicle are measured using a visibility meter. These data are helpful for evaluating and simulating the lane detection performance under different visibility conditions. In addition, the internal parameters (such as focal length) and external parameters (such as the position and direction of the camera relative to the vehicle coordinate system) of the camera are used to calibrate the image data to ensure that the lane detection distance is consistent with the actual measured visible distance. Finally, the collected images are manually calibrated and labeled to ensure that the information such as the position and angle of the lane lines in the image are accurate. The collected data is divided into training set, validation set and test set. The training set is used for model training, the validation set is used for model tuning, and the test set is used to evaluate model performance.

[0048] Through the construction and training process of the lane detection model, it is possible to ensure that the lane detection model can work accurately and reliably under various driving conditions, thereby providing reliable support for fog light control. The embodiment of the present invention only relies on existing vehicle hardware and does not require additional infrastructure. In addition, it can avoid the limitations brought about by the single feature of artificial design in related technologies, and improve the generalization and stability of the algorithm.

[0049] Taking into account that the lane line area includes multiple pixel points, in order to improve the accuracy of extracting lane line vertices from the lane line area, in some embodiments, the lane line vertex includes the pixel point farthest from the target edge among the multiple pixel points; wherein the target edge is the edge closest to the vehicle among the multiple image edges of the road image.

[0050] Based on the way the road image is acquired and the content represented, the target edge can represent the starting position of the driver's field of vision. Therefore, the pixel point farthest from the target edge is used as the lane line vertex, which can effectively represent the end position of the driver's field of vision. It can be understood that if the image edge farthest from the vehicle among multiple image edges is selected as the reference edge, the pixel point closest to the reference edge in the lane line area is the lane line vertex.

[0051] The distance between a pixel point and an image edge may be measured using existing methods, such as straight-line distance, Euclidean distance, etc., and the present invention does not limit this.

[0052] Step 103: Control the fog lights of the vehicle according to the position information.

[0053] Among them, the current farthest visible position of the road in front of the vehicle can be determined according to the position information, the visibility of the road in front of the vehicle can be determined according to the farthest visible position, and the fog lights can be controlled according to the visibility. Specifically, the visibility can be compared with the preset visibility threshold. If the comparison result indicates that the visibility is less than the visibility threshold, the fog lights are controlled to turn on automatically, and the visibility of the current driving scene is improved by the light emitted by the fog lights, thereby improving the driving safety guarantee. Among them, the visibility threshold can be determined according to relevant traffic regulations and industry standards, such as 200 meters. Compared with the related art of controlling the turning on of the fog lights through communication through V2X technology, the fog light control scheme of the embodiment of the present invention is not affected by factors such as the network, and the system responds promptly, while also avoiding data security and privacy issues caused by data exchange.

[0054] In another embodiment of the present invention, considering that the road image represents the field of vision of the driver of the vehicle, and the lane line vertex can represent the farthest visible position of the driver on the road, therefore, Figure 2 As shown, the vehicle fog lamp control method may include the following steps: Step 201: Acquire a road image in front of the vehicle.

[0055] Among them, step 201 is similar to the aforementioned step 101 and will not be repeated here.

[0056] Step 202: Perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image.

[0057] Among them, step 202 is similar to the aforementioned step 102 and will not be repeated here.

[0058] Step 203: Determine the visibility of the road according to the position information.

[0059] The distance between the vertex of the lane line and the vehicle is determined according to the position information, and the distance is determined as the visibility of the road. Specifically, considering the acquisition method of the road image and the content represented by it, the visibility can be determined according to the distance between the vertex of the lane line and the image edge of the road image.

[0060] Considering that the road image covers a certain area in front of the vehicle, the edge of the road image closest to the vehicle can be approximately simulated as the starting position of the driver's field of vision (that is, the position of the driver's eyes), and the edge of the road image farthest from the vehicle is the farthest collection range of the road image collection device (such as the aforementioned front camera).

[0061] Therefore, specifically, the distance between the lane line vertex and the target edge is determined as visibility, wherein the target edge is an image edge closest to the vehicle among multiple image edges.

[0062] Step 204: Control the fog lamp according to the comparison result between the visibility and a preset visibility threshold.

[0063] Among them, the preset visibility threshold is used to represent the distance range that the driver should at least be able to see when there is no driving hazard. For example, when the visibility threshold is 200 meters, it means that the driver should at least be able to see objects within 200 meters in front of the vehicle to avoid accidents.

[0064] The visibility is compared with the preset visibility threshold. If the visibility is less than the visibility threshold, it indicates that the driver's visual range is insufficient and there is a risk of a traffic accident. Therefore, the fog lights need to be turned on to improve driving safety.

[0065] Correspondingly, if the visibility is greater than or equal to the visibility threshold, it indicates that the current driver's visual range is far and there is no risk of a traffic accident for the time being. Therefore, there is no need to control the fog lights to turn on, thereby avoiding the use of fog lights when not necessary and saving vehicle resources.

[0066] In another embodiment of the present invention, considering that the position information obtained by detecting the road image corresponds to the image coordinate system, it cannot directly represent the position and visibility information in the real road environment. Therefore, Figure 3 As shown, the vehicle fog lamp control method may include the following steps: Step 301: Acquire a road image in front of the vehicle.

[0067] Among them, step 301 is similar to the aforementioned step 101 and will not be repeated here.

[0068] Step 302: Perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image; the position information includes the first coordinates of the lane line vertices in the image coordinate system; the road image corresponds to the image coordinate system.

[0069] The process of determining the position information is similar to the aforementioned step 102, and will not be described in detail here. The image coordinate system is used to describe the position of the pixel points in the road image.

[0070] Step 303: Mapping the first coordinate according to the mapping relationship between the image coordinate system and the world coordinate system to obtain a second coordinate; wherein the road corresponds to the world coordinate system.

[0071] Among them, the mapping relationship can be determined according to the device parameters of the road image acquisition device. Taking the acquisition device as the front camera of the vehicle as an example, the coordinate conversion matrix between the coordinate points in the image coordinate system and the coordinate points in the world coordinate system is determined according to the camera parameters of the front camera, and the coordinate conversion matrix is ​​determined as the mapping relationship. The camera parameters may include camera extrinsic parameters and camera intrinsic parameters. Specifically, the camera extrinsic parameters may include the position and orientation of the front camera, and the camera intrinsic parameters may include the focal length and pixel size of the front camera, which is not limited in the embodiment of the present invention. The second coordinate is used to characterize the position of the lane line vertex in the real road environment.

[0072] Step 304: Determine the visibility according to the second coordinate.

[0073] Specifically, according to the second coordinate and the position of the vehicle in the world coordinate system, the distance between the vertex position of the lane line and the vehicle in the real road environment is determined, and the distance is determined as the visibility. For example, the second coordinate may be (0, 150), and the coordinate of the vehicle in the world coordinate system may be (0, 0), then the visibility is 150 meters.

[0074] Step 305: Control the fog lamp according to the comparison result between the visibility and a preset visibility threshold.

[0075] Among them, step 305 is similar to the aforementioned step 204 and will not be repeated here.

[0076] In another embodiment of the present invention, considering that whether the fog lamp needs to be turned on can be determined based on a preset visibility threshold, and the visibility threshold actually corresponds to the farthest position that the driver should be able to see in a real road environment, therefore, in order to further improve the efficiency of fog lamp control, Figure 4 As shown, the vehicle fog lamp control method may include the following steps: Step 401: Acquire a road image in front of the vehicle.

[0077] Among them, step 401 is similar to the aforementioned step 101 and will not be repeated here.

[0078] Step 402: Perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image.

[0079] Among them, step 402 is similar to the aforementioned step 102 and will not be repeated here.

[0080] Step 403: Marking a reference area of ​​the road image according to a preset visibility threshold; the reference area is used to represent an invisible area of ​​the road to the vehicle if the visibility of the road is less than the visibility threshold.

[0081] Among them, the visibility threshold is used to represent the distance range that the driver should at least be visible to when there is no driving danger. For example, when the visibility threshold is 200 meters, it represents that the driver should at least be able to see the objects within 200 meters in front of the vehicle to avoid accidents. Therefore, the boundary of the reference area is determined according to the visibility threshold, so that the reference area can represent the range of the road that is invisible to the driver of the vehicle when there is a risk of accidents due to low visibility. It should be noted that the invisible area is not completely invisible to the driver, but its visual blurriness to the driver is greater than the preset blurriness threshold, so that the driver cannot distinguish the objects in the area that affect driving. For example, in response to the reduced visibility caused by foggy weather, the driver can generally perceive the presence of fog in the invisible area through vision, but the driver cannot identify pedestrians, roadblocks, driving lanes, etc. in the invisible area that are blocked by the fog image and have an impact on driving safety or vehicle control.

[0082] By marking the reference area and visualizing the visibility threshold in the road image, the visibility requirement in the current road scene can be more intuitively represented, which facilitates subsequent comparison with the lane line vertices, thereby controlling the fog lights according to the comparison results.

[0083] Step 404: Control the fog lamp according to the relative position relationship between the position information and the reference area.

[0084] Among them, the relative position relationship is used to characterize whether the vertex of the lane line is located in the reference area. Based on the characterization content of the reference area, when the vertex of the lane line is located in the reference area, it means that the driver cannot see the area outside the reference area clearly, that is, the driver cannot see the area farther than the reference area, so the current visibility does not meet the standard, and the fog lights can be controlled to turn on. When the vertex of the lane line exceeds the reference area, it means that the driver can see the area farther than the reference area, so its visibility is greater than the visibility threshold. There is currently no risk of accidents due to low visibility, and there is no need to turn on the fog lights. By directly comparing the position information with the reference area, it is possible to more efficiently and intuitively determine whether the current visibility meets the standard, thereby improving the efficiency of fog light control.

[0085] In another embodiment of the present invention, in order to improve the efficiency of visibility judgment, thereby improving the efficiency of fog lamp control and better ensuring driving safety, as shown in FIG. Figure 5 As shown, the vehicle fog lamp control process also includes: Step 501: Acquire a road image in front of the vehicle.

[0086] Among them, step 501 is similar to the aforementioned step 101 and will not be repeated here.

[0087] Step 502: Perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image.

[0088] Among them, step 502 is similar to the aforementioned step 102 and will not be repeated here.

[0089] Step 503: Marking a reference area of ​​the road image according to a preset visibility threshold; the reference area is used to represent an invisible area of ​​the road to the vehicle if the visibility of the road is less than the visibility threshold.

[0090] Among them, step 503 is similar to the aforementioned step 403 and will not be repeated here.

[0091] Step 504: Determine whether the lane line vertex is located within the reference area based on the relative position relationship between the position information and the reference area.

[0092] The coordinates of the lane line vertex represented by the position information are compared with the coordinate range covered by the reference area to determine whether the lane line vertex is located in the reference area.

[0093] Step 505: If the lane line vertex is outside the reference area, control the fog lamp to be turned off.

[0094] Based on the lane line vertex representing the farthest position that the driver can currently see, the reference area represents an area that is visible only if the visibility is greater than the visibility threshold. Therefore, if the lane line vertex is outside the reference area, the fog lamp is controlled to turn on.

[0095] Step 506: If the vertex of the lane line is located within the reference area, control the fog lamp to turn on.

[0096] Correspondingly, if the vertex of the lane line is located within the reference area, it indicates that the driver can see the area that is only visible if the visibility is greater than the visibility threshold. Therefore, there is currently no risk of a safety accident due to insufficient visibility, and therefore there is no need to turn on the fog lights.

[0097] In another embodiment of the present invention, in order to improve the efficiency of visibility judgment, thereby improving the efficiency of fog lamp control and more efficiently ensuring driving safety, as shown in FIG. Figure 6 As shown, the vehicle fog lamp control process also includes: Step 601: Acquire a road image in front of the vehicle.

[0098] Among them, step 601 is similar to the aforementioned step 101 and will not be repeated here.

[0099] Step 602: Obtain the trained lane detection model.

[0100] The training process of the lane line detection model may refer to the aforementioned step 102 and will not be repeated here.

[0101] Step 603: compress model parameters of the lane detection model to obtain a compressed model.

[0102] Among them, model parameter compression can use QAT (Quantization-Aware Training) to accurately capture the quantization parameters required to convert model parameters from floating-point representation to low-precision integers, laying the foundation for the quantization conversion of the model.

[0103] Step 604: convert the model format of the compressed model according to the vehicle environment corresponding to the vehicle to obtain a converted model.

[0104] The QNN (Qualcomm Neural Network) model conversion tool can be used to convert the QAT-trained model into a specific file format suitable for the vehicle environment. By adopting advanced neural network structure design and model compression deployment methods, fog lamp control can be performed more efficiently.

[0105] Step 605: Perform post-training quantization on the converted model to obtain a target model.

[0106] Among them, post-training compression is used to embed the quantization parameters into the json configuration file of the converted neural network model to ensure the integrity and accuracy of the quantization information of the model during the conversion process. The json configuration file is compiled into a binary format executable by QNN to facilitate efficient reasoning on the NPU (Neural network Processing Unit).

[0107] Step 606: Deploy the target model onto the vehicle, and perform lane line detection on the road image through the target model to obtain position information of lane line vertices in the road image; the lane line vertices include the points farthest from the vehicle in the lane line area detected in the road image.

[0108] Among them, the process of lane line detection is roughly the same as the aforementioned step 102, and will not be repeated here. Through the aforementioned deployment process, the inference performance of the target model on the NPU can achieve more than 10 times the performance improvement compared to the traditional CPU (Central Processing Unit), significantly improving the inference speed and energy efficiency, and can meet the high requirements of the vehicle environment for real-time performance and energy efficiency. It can maximize the computing power of the vehicle chip and improve the reliability of providing driving safety protection through fog light control.

[0109] Step 607: Control the fog lights of the vehicle according to the position information.

[0110] Among them, step 607 is similar to the aforementioned step 103 and will not be repeated here.

[0111] Please refer to Figure 7 , is a schematic diagram of the hardware structure of the vehicle fog lamp control device 70 provided in the embodiment of the present application. Figure 7 As shown, the vehicle fog lamp control device 70 may include: An acquisition module 701 is used to acquire a road image in front of the vehicle; A detection module 702 is used to perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points farthest from the vehicle in the lane line area detected in the road image; The control module 703 is used to control the fog lights of the vehicle according to the position information.

[0112] Please refer to Figure 8 , is a schematic diagram of the hardware structure of the vehicle fog lamp control device 80 provided in the embodiment of the present application. Figure 8 As shown, the vehicle fog lamp control device 80 may include a processor 801 and a memory 802. The memory 802 is used to store one or more computer programs 803. The one or more computer programs 803 are configured to be executed by the processor 801. The one or more computer programs 803 include instructions, and the above instructions can be used to implement the above vehicle fog lamp control method in the vehicle fog lamp control device 80.

[0113] It is understood that the structure shown in this embodiment does not constitute a specific limitation on the vehicle fog lamp control device 80. In other embodiments, the vehicle fog lamp control device 80 may include more or fewer components than shown, or combine or split some components, or arrange the components differently.

[0114] The processor 801 may include one or more processing units, for example, the processor 801 may include an application processor (AP), a modem, a graphics processing unit (GPU), an image signal processor (ISP), a controller, a video codec, a digital signal processor (DSP), a baseband processor, and / or a neural-network processing unit (NPU), etc. Different processing units may be independent devices or integrated into one or more processors.

[0115] The processor 801 may also be provided with a memory for storing instructions and data. In some embodiments, the memory in the processor 801 is a cache memory. The memory may store instructions or data that the processor 801 has just used or circulated. If the processor 801 needs to use the instruction or data again, it may be directly called from the memory. This avoids repeated access, reduces the waiting time of the processor 801, and thus improves the efficiency of the system.

[0116] In some embodiments, the processor 801 may include one or more interfaces. The interface may include an inter-integrated circuit (I2C) interface, an inter-integrated circuit sound (I2S) interface, a pulse code modulation (PCM) interface, a universal asynchronous receiver / transmitter (UART) interface, a mobile industry processor interface (MIPI), a general-purpose input / output (GPIO) interface, a SIM interface, and / or a USB interface, etc.

[0117] In some embodiments, the processor 801 is used to execute acceleration schemes such as single instruction multiple data (SIMD) and very long instruction word (VLIW).

[0118] In some embodiments, the memory 802 may include a high-speed random access memory and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0119] This embodiment also provides a computer-readable storage medium, which stores computer instructions. When the instructions are executed on a processor, a vehicle fog light control device executes the above-mentioned related method steps to implement the vehicle fog light control method in the above-mentioned embodiment.

[0120] Among them, the vehicle fog light control device and computer-readable storage medium provided in this embodiment are used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0121] In practical applications, the above functions can be distributed to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0122] In several embodiments provided in the present application, the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are schematic. For example, the division of the modules or units is a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0123] The unit described as a separate component may or may not be physically separated, and the component shown as a unit may be one physical unit or multiple physical units, that is, it may be located in one place or distributed in multiple different places. Some or all of the units may be selected according to actual needs to achieve the purpose of the embodiment.

[0124] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0125] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application is essentially 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, which is stored in a storage medium and includes several instructions to enable a device (which can be a single-chip microcomputer, chip, etc.) or a processor (processor) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory (Random Access Memory, RAM), disk or optical disk and other media that can store program code.

[0126] The above description is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions within the technical scope disclosed in the present application should be included in the protection scope of the present application.

Claims

1. A vehicle fog lamp control method, characterized in that: The method comprises: Acquire a road image in front of the vehicle; Performing lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points in the lane line area detected in the road image that are farthest from the vehicle; The fog lights of the vehicle are controlled according to the position information.

2. The method according to claim 1, characterized in that The control process of the fog lamp also includes: determining road visibility based on the location information; The fog lamp is controlled according to a comparison result between the visibility and a preset visibility threshold.

3. The method according to claim 2, characterized in that The road image corresponds to an image coordinate system; the road corresponds to a world coordinate system; the position information includes a first coordinate of the lane line vertex in the image coordinate system; the visibility determination process further includes: Mapping the first coordinates according to a mapping relationship between the image coordinate system and the world coordinate system to obtain second coordinates; The visibility is determined based on the second coordinate.

4. The method according to claim 1, characterized in that: The control process of the fog lamp also includes: Marking a reference area of ​​the road image according to a preset visibility threshold; the reference area is used to represent an invisible area of ​​the road to the vehicle if the visibility of the road is less than the visibility threshold; The fog lamp is controlled according to a relative position relationship between the position information and the reference area.

5. The method according to claim 4, characterized in that The control process of the fog lamp also includes: According to the relative position relationship, determining whether the lane line vertex is located in the reference area; If the vertex of the lane line is outside the reference area, controlling the fog lamp to be turned off; If the lane line vertex is located in the reference area, the fog lamp is controlled to turn on.

6. The method according to any one of claims 1 to 5, characterized in that The road image includes multiple image edges; the lane line area includes multiple pixel points; the lane line vertex includes the pixel point farthest from the target edge among the multiple pixel points; the target edge is the image edge closest to the vehicle among the multiple image edges.

7. The method according to any one of claims 1 to 5, characterized in that The position information is obtained by detecting a target model pre-deployed on the vehicle; the target model determination process includes: Get the trained lane detection model; Compressing model parameters of the lane detection model to obtain a compressed model; According to the vehicle environment corresponding to the vehicle, converting the model format of the compressed model to obtain a converted model; The converted model is quantized after training to obtain the target model.

8. A vehicle fog lamp control device, characterized in that: The vehicle fog lamp control device comprises: An acquisition module, used for acquiring a road image in front of the vehicle; A detection module, configured to perform lane line detection on the road image to obtain position information of lane line vertices in the road image; the lane line vertices include points in the lane line area detected in the road image that are farthest from the vehicle; A control module is used to control the fog lights of the vehicle according to the position information.

9. A vehicle fog lamp control device, comprising a processor and a memory, characterized in that: The memory is used to store instructions, and the processor is used to call the instructions in the memory, so that the vehicle fog light control device executes the vehicle fog light control method according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions, and when the computer instructions are executed on a processor, the processor is caused to execute the vehicle fog lamp control method according to any one of claims 1 to 7.

Images