CN111824003A - A kind of control method and control system of vehicle lamp - Google Patents

Abstract

The present invention provides a control method and control system for a vehicle lamp. The control method obtains image information around a driving vehicle, and recognizes the image information to identify different targets existing in different driving environments, such as at an intersection. In a single environment, static targets including traffic lights and related traffic signs can be identified, and dynamic targets such as other vehicles and pedestrians existing in front of the current vehicle can be identified on the driving road, and traffic police can be identified on the road under the command of the traffic police. The lights are controlled according to the control method corresponding to the identified target information, which realizes different control of the lights in different driving environments, and solves the problem that the existing control methods for the lights cannot adapt to the changing driving environment. .

Description

A kind of control method and control system of vehicle lamp

technical field

The invention belongs to the technical field of intelligent control, and in particular relates to a control method and a control system for a vehicle lamp.

Background technique

During the driving process of the vehicle, it is necessary to control the turning on and off of the lights according to the road conditions and the driving state.

The control methods for vehicle lamps in the prior art include manual control methods and automatic control methods. The manual control mode is to control the opening and closing of the lights according to the driver's needs; the automatic control mode is to detect the ambient light in the driving environment of the vehicle based on the photoelectric sensor, and control the opening and closing of the lights according to the intensity of the ambient light.

In the scheme of manually controlling the lights, the driver is required to operate, which not only causes the driver to operate complicatedly, but also has low accuracy; in the scheme of automatically controlling the lights, although the control of the lights can be automatically realized, it can only control the car according to the ambient light. Turning on or off the lights has a single functional scene and cannot adapt to the complex and changeable driving environment.

SUMMARY OF THE INVENTION

In view of this, the purpose of the present invention is to provide a control method and a control system for a vehicle lamp, so as to solve the problem that the control scheme for the vehicle lamp in the prior art cannot adapt to the complex and changeable driving environment.

The technical solution is as follows:

The present invention provides a control method for a vehicle lamp, comprising:

Collect image information around the driving vehicle;

Identify the image information to obtain target information; the target information includes at least one of the following: dynamic target, static target, and traffic police;

The vehicle lights are controlled according to the control mode corresponding to the target information.

Preferably, the target information is a dynamic target, and the controlling the vehicle lights according to the control mode corresponding to the target information includes:

Track the dynamic target based on a target tracking algorithm, and calculate the relative distance information between the dynamic target tracked at the current moment and the traveling vehicle;

Calculate the difference between the relative distance information between the current moment and the previous moment, and the time difference between the current moment and the previous moment;

Calculate the relative speed information between the dynamic target and the traveling vehicle according to the time difference between the two moments and the difference between the relative distance information between the two moments;

Judging whether the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matching the relative speed information;

If it is determined that the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matched with the relative speed information, the brake light is turned on.

Preferably, if the target information is a static target, the controlling the vehicle lights according to the control mode corresponding to the target information includes:

Calculate the relative distance information between the static target and the moving vehicle;

Determine whether the relative distance information between the static target and the traveling vehicle is less than a static braking distance threshold;

If it is determined that the relative distance information between the static target and the traveling vehicle is less than the static braking distance threshold, the brake light is turned on.

Preferably, the target information is a traffic policeman, and the controlling the vehicle lights according to the control mode corresponding to the target information includes:

Estimate the attitude of the traffic policeman, and get the estimation result;

extracting the human skeleton points of the traffic policeman from the estimation result;

After processing the human skeleton points by using the pre-established spatiotemporal graph convolution network, the semantic information of the traffic police posture is obtained;

According to the semantic information, the vehicle lights matching the semantic information are controlled.

Preferably, it also includes:

According to the collected image information, determine whether the driving environment is a night environment;

If it is judged to be a night environment, turn on the high beam;

judging whether the target information includes a dynamic target;

If it is determined that the target information includes a dynamic target, then determine whether the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold;

If it is determined that the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, if the high beam is currently turned on, the high beam will be switched to the low beam, and if the low beam is currently turned on, the Low beam lights.

Preferably, before the identifying the image information and obtaining the target information, the method further includes:

Filter processing is performed on the image information, and histogram equalization processing is performed on the filtered image information to obtain image information after equalization processing.

The present invention also provides a control system for a vehicle lamp, comprising:

The data acquisition module is used to collect image information around the driving vehicle;

A machine vision algorithm module for identifying the image information and obtaining target information; the target information includes at least one of the following: dynamic target, static target, and traffic police; control.

Preferably, the machine vision algorithm module includes:

A distance calculation unit, configured to track the dynamic target based on a target tracking algorithm, and calculate the relative distance information between the dynamic target tracked at the current moment and the traveling vehicle;

The speed calculation unit is used to calculate the difference between the relative distance information between the current moment and the previous moment, and the time difference between the current moment and the previous moment; according to the time difference between the two moments and the relative distance between the two moments The difference between the distance information is calculated to obtain the relative speed information between the dynamic target and the driving vehicle;

a braking judgment unit, configured to judge whether the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than a braking distance threshold matched with the relative speed information;

The control unit is configured to turn on the brake light if it is determined that the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matched with the relative speed information.

Preferably, the machine vision algorithm module further includes:

an environment judging unit, used for judging whether the driving environment is a night environment according to the collected image information;

a target judging unit, configured to judge whether the target information includes a dynamic target if it is judged to be a night environment;

a switching judgment unit, configured to judge whether the relative distance information between the dynamic target and the driving vehicle is less than a switching distance threshold when judging that the target information includes a dynamic target;

The control unit is configured to turn on the high beams when the environment judgment unit judges that it is a night environment; and when judging that the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, if the currently turned on is far If the high beam is switched to the low beam, if the low beam is currently turned on, keep the low beam.

Preferably, it also includes:

The preprocessing module is used for filtering the image information, and performing histogram equalization processing on the filtered image information to obtain the equalized image information.

Compared with the prior art, the above-mentioned technical solution provided by the present invention has the following advantages:

It can be seen from the above technical solutions that in the present application, the image information around the driving vehicle can be obtained, and the image information can be identified to identify different targets existing in different driving environments. , The static target of the relevant traffic signs, other dynamic targets such as other vehicles and pedestrians existing in front of the current driving vehicle can be identified on the driving road, and the traffic police can be identified on the road under the command of the traffic police, and control according to the determined target information. The vehicle lights are controlled by the method, different control of the vehicle lights in different driving environments is realized, and the problem that the existing control methods for the vehicle lights cannot be adapted to the changing driving environment is solved.

Description of drawings

In order to illustrate the embodiments of the present invention or the technical solutions in the prior art more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are For some embodiments of the present invention, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without creative efforts.

1 is a flowchart of a method for controlling a vehicle lamp provided by an embodiment of the present invention;

2 is a flowchart of a method for controlling vehicle lights in a dynamic target scenario provided by an embodiment of the present invention;

3 is a flowchart of a method for controlling vehicle lights in a static target scenario provided by an embodiment of the present invention;

FIG. 4 is a flowchart of a method for controlling a vehicle lamp provided by an embodiment of the present invention in a scenario where a traffic police exists;

5 is a flowchart of a method for controlling vehicle lights in a night scene provided by an embodiment of the present invention;

FIG. 6 is a schematic structural diagram of a control system for a vehicle lamp provided by an embodiment of the present invention.

Detailed ways

In order to make the purposes, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments These are some embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

This embodiment discloses a method for controlling a vehicle lamp, which is applied to an advanced driver assistance system (Advanced Driver Assistance Systems, ADAS). Referring to FIG. 1 , this embodiment includes the following steps:

S101. Collect image information around the traveling vehicle.

During the driving process of the vehicle, the on-board visual sensor is used to collect image information around the vehicle.

In this embodiment, the vehicle-mounted visual sensor is an ADAS forward-looking camera on the ADAS system. The ADAS forward-looking camera has a long detection distance and can obtain image information farther away from the currently traveling vehicle.

In other embodiments, in addition to using the ADAS forward-looking camera to obtain image information around the vehicle, the wide-angle fisheye camera in the 360 surround view system of the vehicle is also used to obtain image information around the vehicle. The wide-angle fisheye camera has a large imaging angle and clear imaging, and can obtain image information in a wide range around the driving vehicle.

S102. Identify the image information to obtain target information; the target information includes at least one of the following: a dynamic target, a static target, and a traffic policeman.

After the image information is collected, the pre-established visual target detection and classification network is used to identify the image information to obtain target information. The target information includes dynamic targets, static targets, and traffic police.

Vehicles and pedestrians in the actual scene are classified as dynamic targets, traffic light devices and traffic sign devices appearing on the driving road are classified as static targets, and traffic police appearing on the driving road are classified as traffic police.

Exemplarily, the visual target detection and classification network is established based on a deep learning target detection algorithm, such as the YOLO V3 target detection algorithm, to achieve classification and identification of dynamic targets, static targets and traffic police in image information.

S103: Control the vehicle lights according to the control mode corresponding to the target information.

The target information is classified into dynamic target, static target and traffic police. The control methods corresponding to different types of targets are different. If the obtained target information is a dynamic target, the lights are controlled according to the control method corresponding to the dynamic target. Among them, the lights include brake lights, high beams, low beams, turn signals, etc. The control of the lights includes turning on, off and switching.

It can be seen from the above technical solutions that in this embodiment, the image information around the driving vehicle is acquired, and the image information can be identified to identify different targets existing in different driving environments. The static targets of traffic lights and related traffic signs can identify dynamic targets such as other vehicles and pedestrians in front of the current driving vehicle on the driving road, and the traffic police can be identified on the road under the command of the traffic police. The control method controls the car lights, realizes different control of the car lights in different driving environments, and solves the problem that the existing control methods for the car lights cannot adapt to the changing driving environment.

The following describes in detail the control method for vehicle lights when the target information is a dynamic target, a static target and a traffic policeman.

In the case that the target information is a dynamic target, as shown in FIG. 2 , steps S201-S206 are used to realize the control of the vehicle lights.

S201, track the dynamic target based on a target tracking algorithm, and determine relative distance information between the tracked dynamic target and the traveling vehicle;

During the driving process of the vehicle, when there are other driving vehicles or pedestrians in front of the driving vehicle, when the driving vehicle is at a certain distance from other driving vehicles or pedestrians, the brake lights need to be turned on to remind the driver to slow down, and at the same time, the vehicle behind the driving vehicle needs to be activated. Driving vehicles play an early warning role to avoid rear-end collisions caused by emergency braking.

However, dynamic objects such as vehicles or pedestrians in front of the moving vehicle are moving, and the dynamic objects themselves have moving speeds. When the relative speeds between the dynamic objects and the moving vehicle are different, the distances required to activate the brake lights are different. Under normal circumstances, the greater the relative speed between the dynamic target and the moving vehicle, the longer the relative distance between the dynamic target and the moving vehicle, the brake lights are activated; the smaller the relative speed between the dynamic target and the moving vehicle, the less the dynamic target and the moving vehicle are. The brake lights are activated only when the relative distance between the moving vehicles can be slightly closer.

In order to track the dynamic target and obtain the relative distance and relative speed between the dynamic target and the moving vehicle, in this embodiment, when the target in the image information is identified as the dynamic target, the dynamic target is tracked based on the target tracking algorithm. After the dynamic target is tracked, the relative distance information between the dynamic target and the driving vehicle at this moment is determined.

Optionally, the target tracking algorithm is a KCF target tracking algorithm.

S202, calculating the difference between the relative distance information between the current moment and the previous moment, and the time difference between the current moment and the previous moment;

After determining the relative distance information between the dynamic target and the driving vehicle at different times, calculate the difference between the relative distance information at the current moment and the relative distance information at the previous moment, and the time difference between the current moment and the previous moment.

For example, if the current time is 1 minute and 30 seconds, the relative distance information between the dynamic target and the driving vehicle is determined to be 30 meters, and the previous moment is 1 minute and 20 seconds, and the relative distance information between the dynamic target and the driving vehicle is determined to be 20 meters. , then the difference between the relative distance information between the two moments is 10 meters, and the time difference between the two moments is 10 seconds.

S203, according to the time difference between the two moments and the difference between the relative distance information between the two moments, calculate and obtain the relative speed information between the dynamic target and the traveling vehicle;

Based on the formula that the product of speed and time is equal to the distance, according to the time difference between the two moments and the difference between the relative distance information between the two moments, the difference between the relative distance information is divided by the time difference to obtain the dynamic target and Relative speed information between the traveling vehicles.

For example, in step S202, the difference between the calculated relative distance information between the two moments is 10 meters, and the time difference between the two moments is 10 seconds, then, the calculated relative speed information between the dynamic target and the driving vehicle is 1 m/s.

S204, judging whether the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matched with the relative speed information;

If it is determined that the relative distance information between the dynamic target and the driving vehicle at the current moment is less than the braking distance threshold matched with the relative speed information, step S205 is executed;

If it is determined that the relative distance information between the dynamic target and the traveling vehicle at the current moment is greater than or equal to the braking distance threshold matched with the relative speed information, return to step S201;

The correspondence between the relative speed information and the braking distance threshold is stored in advance.

After calculating the relative distance information and relative speed information between the dynamic target and the driving vehicle, find the corresponding braking distance threshold according to the relative speed information in the pre-stored correspondence, and after finding the corresponding braking distance threshold, determine the current Whether the relative distance information between the dynamic target and the moving vehicle is smaller than the braking distance threshold.

When judging that the relative distance information between the dynamic target and the driving vehicle at the current moment is less than the braking distance threshold, it means that the dynamic target is relatively close to the driving vehicle and the brake lights need to be turned on; when judging the relative distance between the dynamic target and the driving vehicle at the current moment If the information is greater than or equal to the braking distance threshold, it means that the dynamic target still has a certain distance from the driving vehicle, and the process returns to step S201.

S205. Turn on the brake light.

When the vehicle is driving near the intersection, there are traffic light signs at the intersection, thereby recognizing static objects. When the target information is a static target, the control method for the lights is similar to the control method for the lights when the dynamic target is used. The difference is that the static target is static and its position does not change. Referring to Figure 3, steps S301-S303 are used. Realize the control of the lights.

S301. Calculate relative distance information between the static target and the traveling vehicle;

S302, judging whether the relative distance information between the static target and the traveling vehicle is less than a static braking distance threshold;

If it is judged that the relative position information is less than the static braking distance threshold, it means that the driving vehicle is already very close to the static target, and step S303 is executed to remind the driver to slow down and at the same time play a warning role for the driving vehicle behind the driving vehicle , to avoid rear-end problems caused by emergency braking;

If it is determined that the relative position information is greater than or equal to the static braking distance threshold, it means that the driving vehicle still has a certain distance from the static target, and the normal driving state can be maintained, and the process returns to step S301.

S303. Turn on the brake light.

In the scene of traffic intervention, the traffic police need to direct the traffic. When the traffic police is identified, the gesture actions of the traffic police need to be recognized, and the corresponding lights are controlled according to the gesture actions of the traffic police.

Referring to FIG. 4 , steps S401-S404 are used to realize the control of the vehicle lights.

S401 , estimating the posture of the traffic policeman to obtain an estimation result;

Based on the human body attitude estimation algorithm, the attitude of the traffic police is estimated.

Optionally, the human body pose estimation algorithm in this embodiment is the OpenPose pose estimation algorithm.

S402, extracting the human skeleton point of the traffic policeman from the estimation result;

S403. After processing the human skeleton points by using a pre-established spatiotemporal graph convolution network, the semantic information of the traffic police posture is obtained;

In this embodiment, a spatiotemporal graph convolution network is pre-established, and the spatiotemporal graph convolution network is used for behavior recognition, for example, a spatiotemporal graph convolution network is established based on the ST-GCN behavior recognition algorithm.

The extracted human skeleton points are used as the input of the spatiotemporal graph convolutional network, and the inputted human skeleton points are recognized by the spatiotemporal graph convolutional network, and the recognized behaviors are output to obtain the semantic information of the traffic police posture.

The semantic information output in this embodiment includes stop, decelerate, turn left, turn right, and the like.

S404. According to the semantic information, control the vehicle lights matching the semantic information.

If the semantic information is stop or decelerate, turn on the brake light to remind the driver to brake or decelerate, and at the same time act as a warning to the vehicles behind;

If the semantic information is a left turn, turn on the left turn light to remind the driver to turn left, and at the same time play an early warning role for the vehicles behind;

If the semantic information is a right turn, the right turn light will be turned on to remind the driver to turn right, and at the same time, it will serve as a warning to the vehicles behind.

The above describes in detail the method of controlling the lights in the dynamic target scene, the static target scene and the traffic police scene. In addition to the lights, it is also necessary to use high beams and low beams. For scenes with low light, on the basis of the above disclosed control method for vehicle lights, it also includes:

S501, according to the collected image information, determine whether the driving environment is a night environment;

If it is determined that it is a night environment, step S502 is executed;

If it is judged that it is not a night environment, exit the process.

In this embodiment, it is determined whether the driving environment is a nighttime environment according to the grayscale of the collected image information.

S502, turn on the high beam;

In the night environment, turning on the high beams allows the driver to observe the environment in a large area around the driving vehicle. However, there are other vehicles or pedestrians on the opposite side of the driving vehicle, and turning on the high beams will affect other vehicles or pedestrians. In order to avoid affecting other vehicles or pedestrians, steps S503-S506 are adopted in this embodiment to realize the recognition of When there is a dynamic target, the switching between high beam and low beam is automatically controlled, which improves driving safety.

In this embodiment, the execution sequence of steps S501-S502 may be executed before any step after step S101. The execution sequence of steps S503-S506 may be executed before any step after step S102.

S503, determine whether the target information includes a dynamic target;

If it is determined that the target information includes a dynamic target, step S504 is performed;

If it is determined that the target information does not include a dynamic target, step S506 is performed;

The dynamic targets here are vehicles, pedestrians, etc.

S504, judging whether the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold;

If it is determined that the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, step S505 is executed;

If it is determined that the relative distance information between the dynamic target and the driving vehicle is greater than or equal to the switching distance threshold, step S506 is executed;

S505, if the high beam is currently turned on, then the high beam is switched to the low beam; if the currently turned on is the low beam, the low beam is maintained, and the process returns to step S503;

In the initial state, the driving vehicle turns on the high beams in the night environment. In this case, if the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, it means that the high beams turned on by the driving vehicle will affect the dynamic target and affect the vision of the dynamic target. For low beam operation.

After the high beam is switched to the low beam, since the driving vehicle is in the process of turning on the low beam, the driver's field of vision is small, so there is a need to switch the low beam to the high beam.

In this embodiment, after switching to the low beam, the step of judging whether the target information includes a dynamic target and the subsequent steps are returned to.

By executing the step of judging whether the target information includes a dynamic target, it can be determined whether there is a dynamic target on the opposite side of the driving vehicle. If it is judged that the target information includes a dynamic target, an operation of judging whether the relative distance information between the dynamic target and the traveling vehicle is smaller than the switching distance threshold is performed.

If the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, since the current vehicle is turning on the low beam headlights, keep the low beam headlights turned on;

If the relative distance information between the dynamic target and the driving vehicle is greater than or equal to the switching distance threshold, it means that the dynamic target is at a certain distance from the driving vehicle. Therefore, the high beams can be turned on, so switching from low beams to high beams is performed. operation.

S506 , if the high beam is currently turned on, keep the high beam, and if the currently turned on is low beam, switch the low beam to the high beam.

It can be seen from the above technical solutions that in this embodiment, the image information around the driving vehicle can be obtained, and the image information can be identified to identify different targets existing in different driving environments. other vehicles, pedestrians and other dynamic targets, according to the relative distance and relative speed between the dynamic target and the current driving vehicle, when the relative distance between the dynamic target and the current driving vehicle is less than the braking distance threshold at a certain relative speed, turn on the brake In the environment of an intersection, static targets including traffic lights and relevant traffic signs can be identified, then when the distance between the current driving vehicle and the static target is less than the static braking distance threshold, the brake light is turned on; The traffic police can be identified on the road, then the posture of the traffic police is recognized to determine whether the semantic information of the traffic police posture is to stop, slow down, turn left or turn right, and according to different semantic information, the lights that match the semantic information are determined. At the same time, it can also identify whether it is night according to the image information. If it is night, on the basis of identifying the target information, it is also necessary to switch control between the high beam and the low beam according to the target information. It can be seen that the present embodiment realizes different control of the vehicle lights under different driving environments, and solves the problem that the existing control methods for the vehicle lights cannot adapt to the changing driving environment.

In other embodiments, before identifying the collected image information, a preprocessing operation may also be included;

The preprocessing operation includes filtering the image information, and the filtering process is to use median filtering to filter the graphic information to suppress image noise; and then perform histogram equalization processing on the filtered image information to obtain a image information. Subsequently, the preprocessed image information is identified.

Corresponding to the above-mentioned control method for a vehicle lamp, the present embodiment also provides a control system for a vehicle lamp. Referring to FIG. 6 , the control system for the vehicle lamp includes:

Data acquisition module 601 and machine vision algorithm module 602;

a data collection module 601, used for collecting image information around the driving vehicle;

The machine vision algorithm module 602 is used to identify the image information and obtain target information; the target information includes at least one of the following: a dynamic target, a static target, and a traffic police; Take control.

It can be seen from the above technical solutions that in this embodiment, the image information around the driving vehicle is acquired, and the image information can be identified to identify different targets existing in different driving environments. The static targets of traffic lights and related traffic signs can identify dynamic targets such as other vehicles and pedestrians in front of the current driving vehicle on the driving road, and the traffic police can be identified on the road under the command of the traffic police. The control method controls the car lights, realizes different control of the car lights in different driving environments, and solves the problem that the existing control methods for the car lights cannot adapt to the changing driving environment.

In other embodiments, the machine vision algorithm module 602 includes:

Distance calculation unit, speed calculation unit, brake judgment unit and control unit;

The distance calculation unit is configured to track the dynamic target based on a target tracking algorithm, and calculate the relative distance information between the dynamic target tracked at the current moment and the traveling vehicle;

The speed calculation unit is used to calculate the difference between the relative distance information between the current moment and the previous moment, and the time difference between the current moment and the previous moment; according to the time difference between the two moments and the time difference between the two moments The difference between the relative distance information of , calculates the relative speed information between the dynamic target and the driving vehicle;

The braking judgment unit is used for judging whether the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matched with the relative speed information;

The control unit is configured to turn on the brake light if it is determined that the relative distance information between the dynamic target and the traveling vehicle at the current moment is less than the braking distance threshold matched with the relative speed information.

Optionally, the distance calculation unit is further configured to calculate relative distance information between the static target and the traveling vehicle;

The braking determination unit is further configured to determine whether the relative distance information between the static target and the traveling vehicle is less than a static braking distance threshold;

The control unit is further configured to turn on the brake light if it is determined that the relative distance information between the static target and the traveling vehicle is less than the static braking distance threshold.

In other embodiments, the machine vision algorithm module 602 further includes:

an environment judgment unit, a target judgment unit and a switch judgment unit;

The environment judging unit is used for judging whether the driving environment is a night environment according to the collected image information;

the target judging unit, configured to judge whether the target information includes a dynamic target if it is judged to be a night environment;

The switching judgment unit is configured to judge whether the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold when judging that the target information includes a dynamic target;

The control unit is configured to turn on the high beams when the environment judgment unit judges that it is a night environment; and when judging that the relative distance information between the dynamic target and the driving vehicle is less than the switching distance threshold, if the currently turned on is far If the high beam is switched to the low beam, if the low beam is currently turned on, keep the low beam.

In other embodiments, the machine vision algorithm module 602 includes:

Attitude estimation unit, extraction unit and processing unit;

The attitude estimation unit is used for estimating the attitude of the traffic policeman to obtain an estimation result;

the extraction unit, used for extracting the human skeleton points of the traffic policeman from the estimation result;

The processing unit is configured to obtain semantic information of the traffic police posture after processing the human skeleton points by using a pre-established spatiotemporal graph convolutional network;

The control unit is further configured to control the vehicle lights matching the semantic information according to the semantic information.

In other embodiments, on the basis of the data acquisition module 601 and the machine vision algorithm module 602, the control system further includes:

preprocessing module 603;

The preprocessing module 603 is configured to perform filtering processing on the image information, and perform histogram equalization processing on the image information after filtering processing to obtain image information after equalization processing.

It can be seen from the above technical solutions that in this embodiment, the image information around the driving vehicle can be obtained, and the image information can be identified to identify different targets existing in different driving environments. other vehicles, pedestrians and other dynamic targets, according to the relative distance and relative speed between the dynamic target and the current driving vehicle, when the relative distance between the dynamic target and the current driving vehicle is less than the braking distance threshold at a certain relative speed, turn on the brake In the environment of an intersection, static targets including traffic lights and relevant traffic signs can be identified, then when the distance between the current driving vehicle and the static target is less than the static braking distance threshold, the brake light is turned on; The traffic police can be identified on the road, then the posture of the traffic police is recognized to determine whether the semantic information of the traffic police posture is to stop, slow down, turn left or turn right, and according to different semantic information, the lights that match the semantic information are determined. At the same time, it can also identify whether it is night according to the image information. If it is night, on the basis of identifying the target information, it is also necessary to switch control between the high beam and the low beam according to the target information. It can be seen that the present embodiment realizes different control of the vehicle lights under different driving environments, and solves the problem that the existing control methods for the vehicle lights cannot adapt to the changing driving environment.

The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same and similar parts between the various embodiments can be referred to each other. As for the apparatus provided in the embodiment, since it corresponds to the method provided in the embodiment, the description is relatively simple, and for related parts, please refer to the description in the method section.

It should be noted that, herein, the terms "comprising", "comprising" or any other variation thereof are intended to encompass non-exclusive inclusion, such that a process, method, article or device comprising a series of elements includes not only those elements, It also includes other elements not expressly listed or inherent to such a process, method, article or apparatus. Without further limitation, an element qualified by the phrase "comprising a..." does not preclude the presence of additional identical elements in a process, method, article or apparatus that includes the element.

The above description of the disclosed embodiments enables any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

The above are only the preferred embodiments of the present invention. It should be pointed out that for those skilled in the art, without departing from the principles of the present invention, several improvements and modifications can be made. It should be regarded as the protection scope of the present invention.

Claims (10)

1. A control method for a vehicle lamp, characterized by comprising:

collecting image information around a running vehicle;

identifying the image information to obtain target information; the target information includes at least one of: dynamic targets, static targets, traffic police;

and controlling the vehicle lamp according to the control mode corresponding to the target information.

2. The control method according to claim 1, wherein the target information is a dynamic target, and the controlling the vehicle lamp according to the control mode corresponding to the target information comprises:

tracking the dynamic target based on a target tracking algorithm, and calculating the relative distance information between the dynamic target tracked at the current moment and the running vehicle;

calculating the difference of the relative distance information between the current time and the previous time and the time difference between the current time and the previous time;

calculating to obtain the relative speed information between the dynamic target and the running vehicle according to the time difference between the two moments and the difference of the relative distance information between the two moments;

judging whether the relative distance information between the dynamic target and the running vehicle at the current moment is smaller than a brake distance threshold value matched with the relative speed information;

and if the relative distance information between the dynamic target and the running vehicle at the current moment is judged to be smaller than the brake distance threshold matched with the relative speed information, starting a brake lamp.

3. The control method according to claim 1, wherein the target information is a static target, and the controlling the vehicle lamp according to the control mode corresponding to the target information comprises:

calculating relative distance information between the static target and the running vehicle;

judging whether the relative distance information between the static target and the running vehicle is smaller than a static braking distance threshold value or not;

and if the relative distance information between the static target and the running vehicle is judged to be smaller than the static braking distance threshold value, starting a brake lamp.

4. The control method according to claim 1, wherein if the target information is a traffic police, the controlling a vehicle light according to the control mode corresponding to the target information comprises:

estimating the attitude of the traffic police to obtain an estimation result;

extracting human skeleton points of the traffic police from the estimation result;

processing the human skeleton points by utilizing a pre-established space-time graph convolution network to obtain semantic information of the traffic police gesture;

and controlling the car light matched with the semantic information according to the semantic information.

5. The control method according to any one of claims 1 to 4, characterized by further comprising:

judging whether the driving environment is a night environment or not according to the acquired image information;

if the environment is judged to be the night environment, the high beam is turned on;

judging whether the target information comprises a dynamic target or not;

if the target information comprises the dynamic target, judging whether the relative distance information between the dynamic target and the running vehicle is smaller than a switching distance threshold value;

if the relative distance information between the dynamic target and the running vehicle is judged to be smaller than the switching distance threshold value, the high beam is switched to the low beam if the high beam is switched on currently, and the low beam is kept if the low beam is switched on currently.

6. The control method according to claim 5, further comprising, before said identifying the image information to obtain the target information:

and carrying out filtering processing on the image information, and carrying out histogram equalization processing on the image information after the filtering processing to obtain equalized image information.

7. A control system for a vehicle lamp, comprising:

the data acquisition module is used for acquiring image information around a running vehicle;

the machine vision algorithm module is used for identifying the image information to obtain target information; the target information includes at least one of: dynamic targets, static targets, traffic police; and controlling the vehicle lamp according to the control mode corresponding to the target information.

8. The control system of claim 7, wherein the machine vision algorithm module comprises:

the distance calculation unit is used for tracking the dynamic target based on a target tracking algorithm and calculating the relative distance information between the dynamic target tracked at the current moment and the running vehicle;

a speed calculation unit for calculating a difference between the relative distance information between the current time and the previous time and a time difference between the current time and the previous time; calculating to obtain the relative speed information between the dynamic target and the running vehicle according to the time difference between the two moments and the difference of the relative distance information between the two moments;

the brake judging unit is used for judging whether the relative distance information between the dynamic target and the running vehicle at the current moment is smaller than a brake distance threshold value matched with the relative speed information;

and the control unit is used for turning on a brake lamp if the relative distance information between the dynamic target and the running vehicle at the current moment is judged to be smaller than the brake distance threshold matched with the relative speed information.

9. The control system of claim 7 or 8, wherein the machine vision algorithm module further comprises:

the environment judging unit is used for judging whether the driving environment is a night environment or not according to the acquired image information;

a target judgment unit, configured to judge whether the target information includes a dynamic target if the target information is judged to be a night environment;

a switching determination unit configured to determine whether relative distance information between the dynamic target and the traveling vehicle is smaller than a switching distance threshold value when it is determined that the target information includes the dynamic target;

the control unit is used for starting the high beam when the environment judgment unit judges that the environment is a night environment; and when the relative distance information between the dynamic target and the running vehicle is judged to be smaller than the switching distance threshold value, if the current turning-on is the high beam, the high beam is switched to the dipped headlight, and if the current turning-on is the dipped headlight, the dipped headlight is kept.

10. The control system of claim 9, further comprising:

and the preprocessing module is used for carrying out filtering processing on the image information and carrying out histogram equalization processing on the image information after the filtering processing to obtain the image information after the equalization processing.

Images