Visibility-based auxiliary driving system and method

By introducing visibility detection and assist modules into the assisted driving system, the problem of high blindness in the existing technology in foggy environments is solved, real-time monitoring and control of vehicle speeds, fog lights and obstacles is realized, and driving safety is improved .

CN119058382APending Publication Date: 2024-12-03GOERTEK INC
View PDF 0 Cites 2 Cited by

Patent Information

Application Number
CN202411035432.6
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-30
Publication Date
2024-12-03

Smart Images

  • Figure CN119058382A_ABST
    Figure CN119058382A_ABST
Patent Text Reader

Abstract

The invention provides a visibility-based auxiliary driving system and method, and the method comprises the steps: capturing a driving environment outside a vehicle through a preset camera device, so as to form a captured video, carrying out the frame division of the captured video in real time, so as to form coherent captured images, carrying out the visibility detection of the captured images, so as to obtain a visibility coefficient, and carrying out the detection of the visibility coefficient. The method comprises the following steps: acquiring a visibility coefficient, comparing the visibility coefficient with a preset visibility threshold value in real time, and if the visibility coefficient is lower than the visibility threshold value, triggering an auxiliary module, so that the auxiliary module assists the driving of the vehicle, thereby detecting the current external environment through an external camera device; the environment information detected by the glasses can be analyzed, a driver is prompted according to different conditions, then the vehicle speed is adjusted, whether a fog lamp is turned on or not is judged, the vehicle following distance is controlled, surrounding obstacles are detected, the auxiliary driving function is achieved, the coverage range of auxiliary driving is expanded, and driving safety is improved.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of assisted driving technology, and more specifically, to an assisted driving system and method based on visibility. Background Art

[0002] As the number of cars continues to increase, the number of traffic accidents is also increasing every year. Therefore, how to reduce the number of traffic accidents and how to improve the intelligence of assisted driving are becoming more and more important. Assisted driving technology can provide real-time monitoring and intervention in situations such as driver fatigue, distraction, and mistakes, thereby greatly reducing the risk of accidents, and can promptly identify potential collision risks and automatically brake when the driver fails to react in time to avoid collisions.

[0003] However, foggy weather is a test of the driver's driving skills. The existing assisted driving system can only sense the surrounding environment through sensors to trigger the auxiliary system. This technology has high blindness and low coverage.

[0004] Therefore, there is an urgent need for a consistent visibility-based assisted driving system that can detect the current external driving environment and adjust the vehicle speed, fog lights, and distance according to the external environment. Summary of the invention

[0005] In view of the above problems, the purpose of the present invention is to provide a visibility-based assisted driving method and system to solve the problem that the existing assisted driving system can sense the surrounding environment through sensors to trigger the auxiliary system, but the technology has high blindness and low coverage.

[0006] The present invention provides an auxiliary driving system based on visibility, which includes an environment monitoring module, a target detection module communicating with the environment monitoring module, and an auxiliary module connected to the target detection module; wherein,

[0007] The environment monitoring module is used to capture the driving environment outside the vehicle through a preset camera device to form a captured video;

[0008] The target detection module is used to divide the captured video into frames in real time to form a coherent captured image, perform visibility detection on the captured image to obtain a visibility coefficient, and compare the visibility coefficient with a preset visibility threshold in real time; wherein, if the visibility coefficient is lower than the visibility threshold, the auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle.

[0009] Preferably, the environment monitoring module includes a camera device and a camera auxiliary unit;

[0010] The camera device is used to capture the external environment of the vehicle;

[0011] The camera auxiliary unit includes a light detection program integrated in the vehicle's on-board system and auxiliary lights arranged around the camera device; wherein, the light detection program is used to detect the lighting coefficient and lighting color of the vehicle's driving environment and determine whether the lighting of the driving environment meets the visibility photography requirements based on the lighting coefficient and the lighting color. If it does not meet the requirements, the auxiliary lights are automatically turned on for light neutralization or fill light.

[0012] Preferably, the target detection module comprises a video processing unit,

[0013] The video processing unit includes a video segmentation module, an image processing module and a gradient filter; wherein,

[0014] The video division module is used to divide the captured video into frames in real time to form a coherent captured image;

[0015] The image processing module is used to perform image layering filtering processing on the captured image through a pre-trained target detection model to remove water drop traces in the captured image and form a standard image;

[0016] The gradient filter performs a gradient convolution operation on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, matches the visibility corresponding to the overall pixel sum through a preset pixel sum-visibility matching table, and uses the matched visibility as the visibility coefficient.

[0017] Preferably, the target detection module includes an intelligent trigger unit,

[0018] The intelligent trigger unit includes a preset comparison program and a trigger command; wherein,

[0019] The comparison program is used to compare the visibility coefficient and the visibility threshold in real time to determine whether the visibility coefficient is lower than the visibility threshold; if the visibility coefficient is lower than the visibility threshold, the trigger command is called so that the trigger command triggers the auxiliary module.

[0020] Preferably, the auxiliary module comprises a vehicle speed module,

[0021] The vehicle speed module includes a vehicle speed detection and control unit for performing vehicle speed detection and vehicle speed control and a vehicle distance detection and control unit for performing vehicle distance detection and vehicle distance control;

[0022] If the vehicle speed detection and control unit determines whether the current vehicle speed meets the foggy speed requirement, if not, an overspeeding reminder pops up on the vehicle display, and the vehicle automatically slows down after a preset time, and the brake taillight in the taillights of the vehicle lights up;

[0023] If the following distance detection and control unit determines that the distance between the vehicle and the vehicle in front is less than a preset vehicle distance threshold, a reminder that the vehicle distance is too small will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake taillights in the vehicle's taillights will light up.

[0024] Preferably, the auxiliary module comprises a fog lamp module,

[0025] The fog lamp module is used to determine the fog level according to the visibility coefficient, and automatically adjust the brightness of the vehicle's fog lamp according to the fog level;

[0026] Among them, the brightness of the fog lamp is negatively correlated with the visibility coefficient.

[0027] Preferably, the auxiliary module comprises an obstacle detection module,

[0028] The obstacle detection module includes a pedestrian detection unit and an obstacle detection unit; wherein,

[0029] The pedestrian detection unit is used to recognize the captured video in real time through a pre-trained human detection network, and if a human figure is recognized, a deceleration reminder is issued on the vehicle display;

[0030] The obstacle detection unit is used to perform defogging on the standard image to form a single image, and input the single image into a pre-trained obstacle recognition model so that the obstacle recognition model can recognize obstacles in the single image. If an obstacle is recognized, an avoidance reminder is issued on the vehicle display.

[0031] The present invention also provides an assisted driving method based on visibility, wherein assisted driving control is performed based on the assisted driving system based on visibility as described above, comprising:

[0032] Capturing the driving environment outside the vehicle through a preset camera device to form a captured video;

[0033] The captured video is frame-divided in real time to form a coherent captured image, visibility detection is performed on the captured image to obtain a visibility coefficient, and the visibility coefficient is compared with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, a preset auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle.

[0034] Preferably, the process of dividing the captured video into frames in real time to form a coherent captured image, and performing visibility detection on the captured image to obtain a visibility coefficient comprises:

[0035] Frame-dividing the captured video in real time to form a coherent captured image;

[0036] Performing image layering filtering processing on the captured image through a pre-trained target detection model to remove water drop traces in the captured image to form a standard image;

[0037] A gradient convolution operation is performed on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, the visibility corresponding to the overall pixel sum is matched through a preset pixel sum-visibility matching table, and the matched visibility is used as the visibility coefficient.

[0038] Preferably, the process of enabling the auxiliary module to assist in driving the vehicle includes the process of performing vehicle speed detection and following vehicle distance detection, including:

[0039] Get the current speed of the vehicle and the distance between the vehicle and the vehicle in front of it;

[0040] Determine whether the vehicle's current speed meets the speed requirement in foggy weather. If not, an overspeeding prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake tail light in the taillight will light up. Determine whether the distance between the vehicle and the vehicle in front is less than the preset vehicle distance threshold. If so, a short distance prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake tail light in the taillight will light up.

[0041] From the above technical solution, it can be seen that the visibility-based assisted driving system provided by the present invention captures the driving environment outside the vehicle through a preset camera device to form a captured video, divides the captured video into frames in real time to form a coherent captured image, performs visibility detection on the captured image to obtain a visibility coefficient, and compares the visibility coefficient with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, the auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle, thereby detecting the current external environment through an external camera device. When the visibility of the external environment is detected to be low, the environmental information detected by the glasses will be analyzed, and prompts will be given to the driver according to different situations, thereby adjusting the vehicle speed, determining whether to turn on the fog lights, controlling the following distance, and detecting surrounding obstacles, thereby realizing the function of assisted driving, expanding the coverage of assisted driving, and improving driving safety. BRIEF DESCRIPTION OF THE DRAWINGS

[0042] By referring to the following description in conjunction with the accompanying drawings, and with a more comprehensive understanding of the present invention, other objects and results of the present invention will become more apparent and easy to understand. In the accompanying drawings:

[0043] Figure 1 is a schematic diagram of a visibility-based driver assistance system according to an embodiment of the present invention;

[0044] Figure 2 The figure is a flow chart of a visibility-based assisted driving method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0045] The existing assisted driving system can only sense the surrounding environment through sensors to trigger the auxiliary system. This technology has high blindness and low coverage.

[0046] In view of the above problems, the present invention provides a visibility-based assisted driving system and method. The specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0047] In order to illustrate the visibility-based auxiliary driving system provided by the present invention, Figure 1 , Figure 2 The embodiments of the present invention are exemplarily indicated.

[0048] The following description of exemplary embodiments is merely illustrative in nature and is not intended to limit the present invention and its application or use. Technologies and devices known to ordinary technicians in the relevant field may not be discussed in detail, but in appropriate cases, the technologies and devices should be considered as part of the specification.

[0049] like Figure 1 As shown, the present invention provides an assisted driving system 100 based on visibility, which includes an environment monitoring module 101, a target detection module 102 communicating with the environment monitoring module 101, and an auxiliary module 103 connected to the target detection module 102; wherein the environment monitoring module 101 is used to capture the driving environment outside the vehicle through a preset camera device to form a captured video; the target detection module 102 is used to divide the captured video into frames in real time to form a coherent captured image, perform visibility detection on the captured image to obtain a visibility coefficient, and compare the visibility coefficient with a preset visibility threshold in real time; if the visibility coefficient is lower than the visibility threshold, the auxiliary module 103 is triggered to enable the auxiliary module to perform assisted driving on the vehicle.

[0050] In a specific embodiment, the environment monitoring module 101 includes a camera device 1011 and a camera auxiliary unit 1012; the camera device 1011 is used to photograph the external environment of the vehicle; the camera auxiliary unit 1012 includes a light detection program integrated in the vehicle's on-board system and auxiliary lights arranged around the camera device; wherein the light detection program is used to detect the illumination coefficient and illumination color of the vehicle's driving environment and determine whether the illumination of the driving environment meets the visibility photography requirements based on the illumination coefficient and the illumination color. If it does not meet the requirements, the auxiliary lights are automatically turned on for light neutralization or fill light. Among them, the camera device can be set to one or more, can be set in front of the vehicle, can also be set at the front windshield inside the vehicle, and can even be integrated into the AR glasses worn by the driver through Bluetooth, and there is no specific restriction on this; the illumination coefficient refers to the brightness of the external environment of the vehicle being photographed, and the visibility photography requirements refer to the brightness requirements and color requirements, for example, the brightness needs to be greater than the preset brightness threshold, and the color cannot be pure black and pure white, etc. The brightness classification can adopt the opal brightness classification, which is divided into 1 to 9 levels, or the brightness level can be set according to a certain brightness division. There is no restriction here, as long as the brightness of the external environment of the vehicle photographed by the camera device can be distinguished and it is compatible with the visibility photography requirements. For example, the brightness level divided by itself is divided into nine levels from dark to bright, with the first level being the darkest and the ninth level being the brightest. Due to the requirement that the external environment of the vehicle being photographed must be clear, the visibility photography requirements can require that the brightness is greater than the first level and less than the ninth level. .

[0051] The target detection module 102 includes a video processing unit 1021 and an intelligent triggering unit 1022, wherein the video processing unit 1021 includes a video segmentation module, an image processing module and a gradient filter; wherein the video segmentation module is used to perform frame segmentation on the captured video in real time to form a coherent captured image; the image processing module is used to perform image layering filtering on the captured image through a pre-trained target detection model to remove water drop traces in the captured image to form a standard image; wherein the target detection model is a YOLOv5 model, and the gradient filter performs a gradient convolution operation on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, matches the visibility corresponding to the overall pixel sum through a preset pixel sum-visibility matching table, and uses the matched visibility as the visibility coefficient. The intelligent trigger unit 1022 includes a preset comparison program and a trigger command; wherein the comparison program is used to compare the visibility coefficient and the visibility threshold in real time to determine whether the visibility coefficient is lower than the visibility threshold; if the visibility coefficient is lower than the visibility threshold, the trigger command is called so that the trigger command triggers the auxiliary module.

[0052] The auxiliary module 103 includes a vehicle speed module 1031, a fog light module 1032, and an obstacle detection module 1033.

[0053] Specifically, the vehicle speed module 1031 includes a vehicle speed detection and control unit for performing vehicle speed detection and speed control and a following distance detection unit for performing following distance detection and following distance control; wherein, if the vehicle speed detection and control unit determines whether the current vehicle speed meets the speed requirement in foggy weather, if not, an overspeeding prompt will pop up on the vehicle display, and the vehicle will automatically slow down after a preset time, and the brake taillights in the vehicle's taillights will light up; in this embodiment, the vehicle speed module 1031 is implemented based on the GPS module on the vehicle, that is, the vehicle's driving speed is detected by the GPS module. If the current driving speed exceeds the maximum safe speed, the vehicle will automatically slow down until the speed drops to within the specified speed limit.

[0054] If the following distance detection unit determines that the distance between the vehicle and the vehicle in front is less than a preset vehicle distance threshold, a reminder that the vehicle distance is too small will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake taillights in the taillights will light up. In a specific embodiment, the captured image pair can be defogged based on a preset deblurring model to remove interference information in the image, and then the license plate of the vehicle in front can be detected based on the pre-trained target detection model, which is a YOLOv5 model. Because the vehicle in front occupies a large proportion of the driver's field of view when the vehicle is stopped, it is difficult to accurately predict the risk of rear-end collision between the vehicle in front and the driver's vehicle based solely on the size of the target vehicle. Therefore, it is judged that the vehicle in front is rapidly approaching the driver's vehicle based on the speed of the frame size change and the frame size. When the speed at which the prediction box increases is greater than the threshold value v th And the border size is larger than s th If the vehicle ahead is judged to be at risk of colliding with the driver's vehicle, the vehicle system will automatically brake to the stop. If the distance does not exceed the preset vehicle distance threshold range, the vehicle will continue to drive.

[0055] As for how to judge whether the distance between the vehicle in front and the current vehicle is within a safe range, the target detection model mainly detects the license plate of the vehicle in front. When the license plate of the vehicle in front is displayed in a larger proportion in the image taken by the external camera device, it means that the vehicle in front is closer to the current vehicle, otherwise it is farther away. Therefore, when the proportion of the license plate target in the original image taken by the external camera device exceeds a certain threshold, the driver will be reminded to slow down until the size ratio is reduced to the specified threshold.

[0056] The fog lamp module 1032 is used to determine the fog level according to the visibility coefficient, and automatically adjust the brightness of the vehicle's fog lamp according to the fog level; wherein the fog lamp brightness is negatively correlated with the visibility coefficient, the higher the visibility coefficient, the dimmer the fog lamp, and the lower the visibility coefficient, the brighter the fog lamp. The fog level is a pre-set fog level, and a general fog level can be preset, or a fog level can be preset by itself. For example, the visibility distance is matched according to the visibility coefficient, and the horizontal visibility distance of 1-10 kilometers is regarded as a light fog level, the horizontal visibility distance of less than 1 kilometer is regarded as a medium fog level, the horizontal visibility distance of 200-500 meters is regarded as a dense fog level, and the horizontal visibility of less than 50 meters is regarded as a strong dense fog level, and then the brightness of the fog lamp is automatically adjusted according to different fog levels. Based on the captured images collected on foggy days, the corresponding visibility level is judged to decide whether to turn on the fog lights and control the light intensity of the fog lights. The light intensity will be evaluated based on the result after the previous gradient convolution. The light intensity is inversely proportional to the sum of pixels after convolution, that is, the lower the visibility, the greater the light intensity. At the same time, in this embodiment, when turning on the fog lights, the vehicle targets in the picture will be detected through the trained YOLOv5 model. For the detected vehicles, the lights will automatically turn off the lights in the target area of ​​the vehicle to avoid interfering with the sight of oncoming vehicles and vehicles in front.

[0057] The obstacle detection module 1033 includes a pedestrian detection unit and an obstacle detection unit; wherein the pedestrian detection unit is used to perform real-time recognition of the captured video through a pre-trained human detection network, and if a pedestrian is recognized, a deceleration reminder is issued on the vehicle display; the obstacle detection unit is used to perform defogging processing on the standard image to form a single image, and input the single image into a pre-trained obstacle recognition model so that the obstacle recognition model can perform obstacle recognition on the single image, and if an obstacle is recognized, an avoidance reminder is issued on the vehicle display, that is, the image still needs to be defogged before obstacle detection, It detects objects on the road that may pose a danger to driving safety, that is, by detecting the scene in front of the driver, when the detected object belongs to a dangerous category, the car will automatically avoid the obstacle and will remind the driver that there is an obstacle ahead and he needs to slow down. In rainy, snowy, foggy and haze weather, the driver's vision in the car is poor and it is difficult to clearly see the pedestrians outside the car. Multiple external cameras can help the driver understand this situation. When there are many pedestrians outside the car, the driver will be reminded to slow down. When the car itself is at a high speed and there are many pedestrians around, the car will automatically slow down and keep a safe distance from the pedestrians.

[0058] That is, the system involves visibility detection, target detection, image classification and other technologies. First, the system detects the visibility of the images collected by the camera outside the vehicle based on the gradient filter. If the current visibility is high, the system will continue to monitor the current environment. If the current visibility is poor, the system will assist the driver in driving, control the vehicle speed, following distance, whether to turn on the lights, and detect obstacles on the road. In terms of vehicle speed, the system calculates the maximum speed to ensure driving safety based on different visibility. If the vehicle speed exceeds the range, the system will issue a warning to the driver. In terms of following distance, the system determines whether the distance between the current vehicle and the vehicle in front is less than the minimum following distance. If it is less than the following distance, the system determines that there is a risk of rear-end collision with the vehicle in front. In terms of lights, the system determines whether to turn on the fog lights and adjusts the brightness of the fog lights according to different visibility. In terms of obstacle detection, the system identifies objects on the road that may affect the safety of vehicle driving, thereby improving driving safety.

[0059] In short, the visibility-based assisted driving system in this embodiment is used for assisted driving. It combines target detection, vehicle speed recognition, visibility detection, image classification and other technologies to detect the current external environment. When the visibility of the external environment is detected to be low, the environmental information detected by the glasses will be analyzed, and prompts will be given to the driver according to different situations. The system will then adjust the vehicle speed, determine whether to turn on the fog lights, control the following distance, and detect surrounding obstacles to achieve the function of assisted driving.

[0060] As described above, the visibility-based assisted driving system provided by the present invention captures the driving environment outside the vehicle through a preset camera device to form a captured video, divides the captured video into frames in real time to form a coherent captured image, performs visibility detection on the captured image to obtain a visibility coefficient, and compares the visibility coefficient with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, the auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle, thereby detecting the current external environment through an external camera device. When the visibility of the external environment is detected to be low, the environmental information detected by the glasses is analyzed, and prompts are given to the driver according to different situations, thereby adjusting the vehicle speed, determining whether to turn on the fog lights, controlling the following distance, and detecting surrounding obstacles, thereby realizing the function of assisted driving, expanding the coverage of assisted driving, and improving driving safety.

[0061] like Figure 2 As shown, the present invention also provides a visibility-based assisted driving method, which performs assisted driving control based on the visibility-based assisted driving system as described above, comprising:

[0062] S1: Capturing the driving environment outside the vehicle through a preset camera device to form a captured video;

[0063] S2: Divide the captured video into frames in real time to form a coherent captured image, perform visibility detection on the captured image to obtain a visibility coefficient, and compare the visibility coefficient with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, trigger a preset auxiliary module to enable the auxiliary module to assist in driving the vehicle.

[0064] The process of dividing the captured video into frames in real time to form a coherent captured image, and performing visibility detection on the captured image to obtain a visibility coefficient includes:

[0065] SA21: Frame-dividing the captured video in real time to form a coherent captured image;

[0066] SA22: performing image layering filtering processing on the captured image through a pre-trained YOLOv5 model to remove water drop traces in the captured image to form a standard image;

[0067] SA23: Perform a gradient convolution operation on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, match the visibility corresponding to the overall pixel sum through a preset pixel sum-visibility matching table, and use the matched visibility as the visibility coefficient.

[0068] The process of enabling the auxiliary module to assist the vehicle in driving includes the process of performing vehicle speed detection and following vehicle distance detection, including:

[0069] SB21: Get the current speed of the vehicle and the distance between the vehicle and the vehicle in front of it;

[0070] SB22: Determine whether the vehicle's current speed meets the speed requirements for foggy weather. If not, a speeding prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake taillight in the taillight will light up. Determine whether the distance between the vehicle and the vehicle in front is less than the preset vehicle distance threshold. If so, a distance too small prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake taillight in the taillight will light up.

[0071] For a more specific implementation, please refer to the above-mentioned embodiment of the visibility-based driver assistance system, which will not be described in detail here.

[0072] As described above, the visibility-based assisted driving method provided by the present invention captures the driving environment outside the vehicle through a preset camera device to form a captured video, divides the captured video into frames in real time to form a coherent captured image, performs visibility detection on the captured image to obtain a visibility coefficient, and compares the visibility coefficient with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, the auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle, thereby detecting the current external environment through an external camera device. When the visibility of the external environment is detected to be low, the environmental information detected by the glasses is analyzed, and prompts are given to the driver according to different situations, thereby adjusting the vehicle speed, determining whether to turn on the fog lights, controlling the following distance, and detecting surrounding obstacles, thereby realizing the function of assisted driving, expanding the coverage of assisted driving, and improving driving safety.

[0073] As described above with reference to the accompanying drawings, the assisted driving system and method based on visibility proposed in accordance with the present invention are described by way of example. However, those skilled in the art should understand that various improvements can be made to the assisted driving system and method based on visibility proposed in the present invention without departing from the content of the present invention. Therefore, the protection scope of the present invention should be determined by the content of the attached claims.

Claims

1. A visibility-based assisted driving system, characterized in that: It includes an environment monitoring module, a target detection module communicating with the environment monitoring module, and an auxiliary module connected to the target detection module; wherein, The environment monitoring module is used to capture the driving environment outside the vehicle through a preset camera device to form a captured video; The target detection module is used to divide the captured video into frames in real time to form a coherent captured image, perform visibility detection on the captured image to obtain a visibility coefficient, and compare the visibility coefficient with a preset visibility threshold in real time; wherein, if the visibility coefficient is lower than the visibility threshold, the auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle.

2. The visibility-based driver assistance system according to claim 1, characterized in that: The environment monitoring module includes a camera device and a camera auxiliary unit; The camera device is used to capture the external environment of the vehicle; The camera auxiliary unit includes a light detection program integrated in the vehicle's on-board system and auxiliary lights arranged around the camera device; wherein, the light detection program is used to detect the lighting coefficient and lighting color of the vehicle's driving environment and determine whether the lighting of the driving environment meets the visibility photography requirements based on the lighting coefficient and the lighting color. If it does not meet the requirements, the auxiliary lights are automatically turned on for light neutralization or fill light.

3. The visibility-based driver assistance system according to claim 1, characterized in that: The target detection module includes a video processing unit, and the video processing unit includes a video segmentation module, an image processing module and a gradient filter; wherein, The video division module is used to divide the captured video into frames in real time to form a coherent captured image; The image processing module is used to perform image layering filtering processing on the captured image through a pre-trained target detection model to remove water drop traces in the captured image and form a standard image; The gradient filter performs a gradient convolution operation on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, matches the visibility corresponding to the overall pixel sum through a preset pixel sum-visibility matching table, and uses the matched visibility as the visibility coefficient.

4. The visibility-based driver assistance system according to claim 1, characterized in that: The target detection module includes an intelligent trigger unit, The intelligent trigger unit includes a preset comparison program and a trigger command; wherein, The comparison program is used to compare the visibility coefficient and the visibility threshold in real time to determine whether the visibility coefficient is lower than the visibility threshold; if the visibility coefficient is lower than the visibility threshold, the trigger command is called so that the trigger command triggers the auxiliary module.

5. The visibility-based driver assistance system according to claim 1, characterized in that: The auxiliary module includes a vehicle speed module, The vehicle speed module includes a vehicle speed detection and control unit for performing vehicle speed detection and vehicle speed control and a vehicle distance detection and control unit for performing vehicle distance detection and vehicle distance control; If the vehicle speed detection and control unit determines whether the current vehicle speed meets the foggy speed requirement, if not, an overspeeding reminder pops up on the vehicle display, and the vehicle automatically slows down after a preset time, and the brake taillight in the taillights of the vehicle lights up; If the following distance detection and control unit determines that the distance between the vehicle and the vehicle in front is less than a preset vehicle distance threshold, a reminder that the vehicle distance is too small will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake taillights in the vehicle's taillights will light up.

6. The visibility-based driver assistance system according to claim 1, characterized in that: The auxiliary module includes a fog lamp module, The fog lamp module is used to determine the fog level according to the visibility coefficient, and automatically adjust the brightness of the vehicle's fog lamp according to the fog level; Among them, the brightness of the fog lamp is negatively correlated with the visibility coefficient.

7. The visibility-based driver assistance system according to claim 1, characterized in that: The auxiliary module includes an obstacle detection module, The obstacle detection module includes a pedestrian detection unit and an obstacle detection unit; wherein, The pedestrian detection unit is used to recognize the captured video in real time through a pre-trained human detection network, and if a human figure is recognized, a deceleration reminder is issued on the vehicle display; The obstacle detection unit is used to perform defogging on the standard image to form a single image, and input the single image into a pre-trained obstacle recognition model so that the obstacle recognition model can recognize obstacles in the single image. If an obstacle is recognized, an avoidance reminder is issued on the vehicle display.

8. A visibility-based assisted driving method, characterized in that: The assisted driving control is performed based on the visibility-based assisted driving system as described in claims 1 to 7, comprising: Capturing the driving environment outside the vehicle through a preset camera device to form a captured video; The captured video is frame-divided in real time to form a coherent captured image, visibility detection is performed on the captured image to obtain a visibility coefficient, and the visibility coefficient is compared with a preset visibility threshold in real time. If the visibility coefficient is lower than the visibility threshold, a preset auxiliary module is triggered to enable the auxiliary module to assist in driving the vehicle.

9. The visibility-based assisted driving method according to claim 8, characterized in that: The process of dividing the captured video into frames in real time to form a coherent captured image, and performing visibility detection on the captured image to obtain a visibility coefficient includes: Frame-dividing the captured video in real time to form a coherent captured image; Performing image layering filtering processing on the captured image through a pre-trained target detection model to remove water drop traces in the captured image to form a standard image; A gradient convolution operation is performed on the standard image through a preset gradient filtering algorithm to obtain the overall pixel sum of the current standard image, the visibility corresponding to the overall pixel sum is matched through a preset pixel sum-visibility matching table, and the matched visibility is used as the visibility coefficient.

10. The visibility-based assisted driving method according to claim 8, characterized in that: The process of enabling the auxiliary module to assist the vehicle in driving includes the process of performing vehicle speed detection and following vehicle distance detection, including: Get the current speed of the vehicle and the distance between the vehicle and the vehicle in front of it; Determine whether the vehicle's current speed meets the speed requirement in foggy weather. If not, an overspeeding prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake tail light in the taillight will light up. Determine whether the distance between the vehicle and the vehicle in front is less than the preset vehicle distance threshold. If so, a short distance prompt will pop up on the vehicle display, and the vehicle will automatically slow down after the preset time, and the brake tail light in the taillight will light up.

Citation Information

Cited By

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

    CN119893799A

  • Control method, device and equipment after vehicle triggers danger warning lamp and medium

    CN120534270A