CN111497741A

CN111497741A - Collision early warning method and device

Info

Publication number: CN111497741A
Application number: CN201910092951.9A
Authority: CN
Inventors: 余倩
Original assignee: Hangzhou Hikvision Digital Technology Co Ltd
Current assignee: Hangzhou Hikvision Digital Technology Co Ltd
Priority date: 2019-01-30
Filing date: 2019-01-30
Publication date: 2020-08-07
Anticipated expiration: 2039-01-30
Also published as: CN111497741B

Abstract

The invention discloses a collision early warning method and device, and belongs to the field of vehicle control. The method comprises the following steps: determining a target vehicle corresponding to a current vehicle, wherein the target vehicle is a vehicle running in front of the current vehicle; acquiring the weight of the target vehicle belonging to a vehicle on the side of an adjacent road, wherein the vehicle on the side of the adjacent road is a vehicle with a running lane adjacent to the lane where the current vehicle is located; and obtaining the probability of triggering the alarm by the target vehicle according to the weight of the target vehicle belonging to the adjacent road side vehicle. According to the method, the weight of the target vehicle running ahead and belonging to the side vehicle of the adjacent road is obtained, so that the probability of triggering the alarm of the target vehicle is obtained, the weight can influence the probability of triggering the alarm of the target vehicle to different degrees, and the method for realizing the collision early warning by detecting the side vehicle of the adjacent road can avoid the false alarm of the side vehicle of the adjacent road and improve the accuracy of the collision early warning.

Description

Collision early warning method and device

Technical Field

The invention relates to the field of vehicle control, in particular to a collision early warning method and device.

Background

Vehicle collision is a common traffic accident, and on a congested urban road, if a vehicle is too close to a vehicle in front, a collision event is easy to occur, so that a collision early warning method is urgently needed to warn a driver when a potential collision danger exists.

At present, in the related art, a video image of a front vehicle is generally acquired through a collision early warning system of the vehicle, a relative distance and a relative speed between the vehicle and the front vehicle are obtained through analysis, and collision time between the vehicle and the front vehicle, that is, time required for collision between the vehicle and the front vehicle, is obtained through calculation, the collision time is compared with an early warning time threshold, if the collision time is less than or equal to the early warning time threshold, an alarm is given to a driver, and if the collision time is greater than the early warning time threshold, the alarm is not given to the driver.

According to the technology, whether the alarm is given out is determined by calculating the collision time of the two vehicles and comparing the collision time with the early warning time threshold value, the front vehicle can be an adjacent-lane side vehicle, namely, the adjacent-lane vehicle which is closer to the current vehicle, under the condition, if the collision time of the two vehicles is smaller than or equal to the early warning time threshold value, the adjacent-lane side vehicle can trigger the alarm, but the two vehicles are not in collision danger in fact, false alarm can be caused, and therefore a collision early warning method is urgently needed to avoid the false alarm of the adjacent-lane side vehicle and improve the accuracy of collision early warning.

Disclosure of Invention

The embodiment of the invention provides a collision early warning method and device, which can solve the problem of false alarm of vehicles on the side surface of an adjacent road in the related art. The technical scheme is as follows:

in a first aspect, a collision warning method is provided, where the method includes:

determining a target vehicle corresponding to a current vehicle, wherein the target vehicle is a vehicle running in front of the current vehicle;

acquiring the weight of the target vehicle belonging to a side vehicle of an adjacent road, wherein a lane where the side vehicle of the adjacent road is located is adjacent to a lane where the current vehicle is located;

and obtaining the probability of triggering the alarm by the target vehicle according to the weight of the target vehicle belonging to the adjacent road side vehicle.

In one possible implementation manner, obtaining the probability that the target vehicle triggers an alarm according to the weight of the target vehicle belonging to the adjacent-lane side vehicle includes:

when the weight of the target vehicle belonging to the side vehicle of the adjacent road is greater than or equal to the target weight, the probability of triggering the alarm by the target vehicle is zero;

when the weight of the target vehicle belonging to the adjacent lane side vehicle is smaller than the target weight, obtaining the probability of triggering an alarm by the target vehicle according to the weight of the target vehicle belonging to the adjacent lane side vehicle and the relative motion condition of the target vehicle and the current vehicle, wherein the relative motion condition comprises at least one of the relative distance between the target vehicle and the current vehicle, the relative speed and the time interval of the collision moment of the current time distance.

In a possible implementation manner, the obtaining, according to the weight of the target vehicle belonging to the adjacent side vehicle and the relative motion condition of the target vehicle and the current vehicle, the probability that the target vehicle triggers the alarm includes:

respectively obtaining a first probability and a second probability of triggering alarm of the target vehicle according to the weight of the target vehicle belonging to the adjacent road side vehicle and the relative motion condition;

and carrying out weighted summation on the obtained first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

In one possible implementation, after obtaining the integrated probability that the target vehicle triggers the alarm, the method further includes:

when the current vehicle corresponds to a target vehicle and the comprehensive probability of triggering the alarm by the target vehicle is greater than the target probability, alarming the target vehicle;

and when the current vehicle corresponds to a plurality of target vehicles, selecting the target vehicle with the maximum comprehensive probability and greater than the target probability from the plurality of target vehicles to alarm according to the comprehensive probability of triggering alarm by the plurality of target vehicles.

In one possible implementation manner, the obtaining the weight of the target vehicle belonging to the adjacent side vehicle includes:

acquiring at least one item of information of distance change between the target vehicle and the current vehicle, size change rules of the target vehicle imaged in an image, wheel conditions of the target vehicle imaged in the image and the distance between the position of the target vehicle imaged in the image and the edge of the image, wherein the image is acquired by a front image sensor of the current vehicle;

and acquiring the weight of the target vehicle belonging to the adjacent road side vehicle according to the at least one item of information.

In one possible implementation manner, the obtaining, according to the at least one item of information, a weight that the target vehicle belongs to an adjacent-lane side vehicle includes:

executing a step of obtaining a weight according to each item of the at least one item of information;

and multiplying all the obtained weights by respective coefficients and then adding the obtained weights to obtain the weight of the target vehicle belonging to the adjacent lane side vehicle.

In one possible implementation, obtaining a distance change between the target vehicle and the current vehicle includes:

obtaining two-dimensional position information of the target vehicle imaged in each frame of image based on continuous multi-frame images acquired by the front-facing image sensor;

converting the two-dimensional position information imaged by the target vehicle in each frame of image into three-dimensional position information of the target vehicle in a world coordinate system;

and obtaining the distance change condition between the target vehicle and the current vehicle according to the three-dimensional position information of the target vehicle and the current vehicle at a plurality of moments, wherein the plurality of moments are the acquisition moments of the continuous multi-frame images.

In one possible implementation, acquiring a size change rule of the target vehicle imaged in the image includes:

acquiring the size of the target vehicle imaged in each frame of image based on the continuous multi-frame image acquired by the front-facing image sensor;

and obtaining the size change rule according to the size of the target vehicle imaged in each frame of image.

In one possible implementation, obtaining a distance between a position where the target vehicle is imaged in the image and an edge of the image includes:

acquiring the distance between the position imaged by the target vehicle in the frame of image and the edge of the image based on the frame of image acquired by the front-facing image sensor; or the like, or, alternatively,

and acquiring the distance between the imaging position of the target vehicle in each frame of image and the edge of the image based on the continuous multi-frame image acquired by the front-facing image sensor, and averaging all the acquired distances.

In one possible implementation, acquiring wheel conditions of the target vehicle imaged in the image includes:

acquiring the number of wheels imaged in one frame of image by the target vehicle based on the one frame of image acquired by the front-facing image sensor; or the like, or, alternatively,

and acquiring the number of wheels imaged in each frame of image of the target vehicle based on the continuous multi-frame images acquired by the front-end image sensor, and calculating the average value of the number of wheels corresponding to the continuous multi-frame images.

In a second aspect, there is provided a collision warning apparatus, the apparatus comprising:

the device comprises a determining module, a judging module and a judging module, wherein the determining module is used for determining a target vehicle corresponding to a current vehicle, and the target vehicle is a vehicle running in front of the current vehicle;

the acquisition module is used for acquiring the weight of the target vehicle belonging to the adjacent lane side vehicle, wherein the lane of the adjacent lane side vehicle is adjacent to the lane of the current vehicle;

and the alarm module is used for obtaining the probability of triggering alarm of the target vehicle according to the weight of the target vehicle belonging to the adjacent road side vehicle.

In one possible implementation, the alarm module is configured to:

In a possible implementation manner, the alarm module is configured to obtain a first probability and a second probability that the target vehicle triggers an alarm according to the weight of the target vehicle belonging to the adjacent lane side vehicle and the relative motion condition, respectively; and carrying out weighted summation on the obtained first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

In one possible implementation, the alarm module is configured to:

In one possible implementation, the obtaining module is configured to:

In a third aspect, a computer device is provided, comprising a processor and a memory; the memory is used for storing at least one instruction; the processor is configured to execute at least one instruction stored in the memory to implement the method steps of any one of the implementation manners of the first aspect.

In a fourth aspect, a computer-readable storage medium is provided, in which at least one instruction is stored, and the at least one instruction, when executed by a processor, implements the method steps of any one of the implementations of the first aspect.

The technical scheme provided by the embodiment of the invention has the beneficial effects that at least:

the method comprises the steps of obtaining the weight of a target vehicle running in front and belonging to a vehicle on the side of an adjacent road, further obtaining the probability of triggering the alarm of the target vehicle, wherein the weight can influence the probability of triggering the alarm of the target vehicle to different degrees. According to the method for realizing collision early warning by detecting the adjacent-road side vehicles, when a plurality of vehicles are driven by the front side, the vehicle postures can be distinguished according to the weight of the vehicles belonging to the adjacent-road side vehicles, the vehicles which are driven on the adjacent road and have no overlapping tail with the current vehicle are determined from the plurality of vehicles, and the vehicles are the adjacent-road side vehicles and have no collision danger, so that no alarm is given, and the false alarm of the adjacent-road side vehicles is avoided.

In addition, vehicles running on the lane and having overlapped tails with the current vehicle can be determined from a plurality of vehicles running from the front, when the vehicles are in collision danger, the alarm can be given in time, the problem that the vehicle closest to the current vehicle in the lane leaks the alarm due to the fact that the vehicle on the side of the adjacent lane is given an alarm when the distance between the vehicle on the side of the adjacent lane and the current vehicle is short is avoided, and the accuracy of collision early warning is improved.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

Fig. 1 is a flowchart of a collision warning method according to an embodiment of the present invention;

fig. 2 is a flowchart of a collision warning method according to an embodiment of the present invention;

FIG. 3 is a schematic representation of an image of a preceding vehicle provided by an embodiment of the present invention;

FIG. 4 is a schematic representation of an image of a preceding vehicle provided by an embodiment of the present invention;

fig. 5 is a schematic structural diagram of a collision warning apparatus according to an embodiment of the present invention;

fig. 6 is a schematic structural diagram of a computer device 600 according to an embodiment of the present invention.

Detailed Description

In order to make the objects, technical solutions and advantages of the present invention more apparent, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Fig. 1 is a flowchart of a collision warning method according to an embodiment of the present invention. Referring to fig. 1, the method includes:

101. and determining a target vehicle corresponding to the current vehicle, wherein the target vehicle is a vehicle running in front of the current vehicle.

102. And acquiring the weight of the target vehicle belonging to the adjacent road side vehicle, wherein the lane of the adjacent road side vehicle is adjacent to the lane of the current vehicle.

103. And obtaining the probability of triggering the alarm by the target vehicle according to the weight of the target vehicle belonging to the side vehicle of the adjacent road.

According to the method provided by the embodiment of the invention, the probability of triggering the alarm by the target vehicle is obtained by obtaining the weight of the target vehicle running ahead, which belongs to the side vehicle of the adjacent road, and the weight can influence the probability of triggering the alarm by the target vehicle to different degrees. According to the method for realizing collision early warning by detecting the adjacent-road side vehicles, when a plurality of vehicles are driven by the front side, the vehicle postures can be distinguished according to the weight of the vehicles belonging to the adjacent-road side vehicles, the vehicles which are driven on the adjacent road and have no overlapping tail with the current vehicle are determined from the plurality of vehicles, and the vehicles are the adjacent-road side vehicles and have no collision danger, so that no alarm is given, and the false alarm of the adjacent-road side vehicles is avoided.

In one possible implementation manner, obtaining the probability that the target vehicle triggers the alarm according to the weight that the target vehicle belongs to the adjacent-road side vehicle includes:

when the weight of the target vehicle belonging to the side vehicle of the adjacent lane is smaller than the target weight, obtaining the probability of triggering an alarm by the target vehicle according to the weight of the target vehicle belonging to the side vehicle of the adjacent lane and the relative motion condition of the target vehicle and the current vehicle, wherein the relative motion condition comprises at least one of the relative distance between the target vehicle and the current vehicle, the relative speed and the time interval of the collision moment of the distance at the current moment.

and carrying out weighted summation on the first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

In one possible implementation, the obtaining the weight of the target vehicle belonging to the adjacent-lane side vehicle includes:

acquiring at least one item of information of the distance change condition between the target vehicle and the current vehicle, the size change rule of the target vehicle imaged in an image, the wheel condition of the target vehicle imaged in the image and the distance between the position of the target vehicle imaged in the image and the edge of the image, wherein the image is acquired by a front image sensor of the current vehicle;

and acquiring the weight of the target vehicle belonging to the adjacent lane side vehicle according to the at least one item of information.

In one possible implementation manner, the obtaining the weight of the target vehicle belonging to the adjacent-lane side vehicle according to the at least one item of information includes:

In one possible implementation, obtaining a distance variation between the target vehicle and the current vehicle includes:

based on continuous multi-frame images acquired by the front-facing image sensor, acquiring two-dimensional position information of the target vehicle imaged in each frame of image;

acquiring the size of the target vehicle imaged in each frame of image based on the continuous multi-frame image acquired by the front-mounted image sensor;

acquiring the distance between the position imaged by the target vehicle in the frame of image and the edge of the image based on the frame of image acquired by the front-mounted image sensor; or the like, or, alternatively,

and acquiring the distance between the imaging position of the target vehicle in each frame of image and the edge of the image based on the continuous multi-frame image acquired by the front-mounted image sensor, and averaging all the acquired distances.

acquiring the number of wheels imaged in one frame of image by the target vehicle based on the one frame of image acquired by the front-mounted image sensor; or the like, or, alternatively,

All the above-mentioned optional technical solutions can be combined arbitrarily to form the optional embodiments of the present invention, and are not described herein again.

Fig. 2 is a flowchart of a collision warning method according to an embodiment of the present invention. Referring to fig. 2, the method includes:

201. and determining a target vehicle corresponding to the current vehicle, wherein the target vehicle is a vehicle running in front of the current vehicle.

In the embodiment of the invention, the current vehicle can be provided with a collision early warning system, and the collision early warning system is a system for warning and reminding a driver of the current vehicle when the vehicle in front of the current vehicle has potential danger to the current vehicle on a lane of the current vehicle. The collision early warning method provided by the embodiment of the invention can be realized by a collision early warning system of the current vehicle.

Additionally, a front-facing image sensor, such as an image sensor in a front-facing camera, may be mounted on the current vehicle for obtaining a real-time image in front of the current vehicle. Accordingly, in one possible implementation, the target vehicle refers to a vehicle that is within an image capture range of a front image sensor of the current vehicle. Accordingly, the determination process of the target vehicle includes: the collision early warning system of the current vehicle acquires images through a front-mounted image sensor of the current vehicle, and the vehicle included in the acquired images is used as a target vehicle corresponding to the current vehicle, wherein the target vehicle can be one vehicle or a plurality of vehicles.

The lane currently being traveled by the target vehicle may be the same as the lane currently being traveled by the current vehicle, which may be in danger of collision with the current vehicle. Of course, the target vehicle may be adjacent to the current lane in which the current vehicle is currently traveling, and these vehicles traveling in the adjacent lane are referred to as adjacent lane vehicles. Some vehicles in the adjacent lane are easily distinguished from the target vehicle running on the lane where the current vehicle is located, but some adjacent lane vehicles which are close to the current vehicle do not completely enter the image acquisition range (or the visual field of a camera) of the image sensor, most of the images of the vehicles in the images are the side surfaces of the vehicles, so that the vehicles cannot be distinguished from the target vehicle running on the lane where the current vehicle is located according to the position of a target single frame, the time interval of the collision moment of the current time and the like, and the vehicles are called as adjacent lane side vehicles.

These adjacent lane side vehicles have no risk of collision with the present vehicle. Therefore, when the current vehicle determines whether the target vehicle triggers the alarm, it may be determined first whether the target vehicle belongs to the adjacent-lane side vehicle, and specifically, the current vehicle may obtain the weight of the adjacent-lane side vehicle to which the target vehicle belongs through the subsequent step 202.

202. And acquiring the weight of the target vehicle belonging to the adjacent road side vehicle, wherein the lane of the adjacent road side vehicle is adjacent to the lane of the current vehicle.

In one possible implementation, the step 202 may include the following steps a and b:

step a, acquiring at least one item of information of the distance change condition between the target vehicle and the current vehicle, the size change rule of the target vehicle imaged in the image, the wheel condition of the target vehicle imaged in the image and the distance between the position of the target vehicle imaged in the image and the edge of the image, wherein the image is acquired by a front image sensor of the current vehicle.

In one possible implementation manner, the obtaining process may include, for a distance change between the target vehicle and the current vehicle: based on continuous multi-frame images acquired by the front-facing image sensor, acquiring two-dimensional position information of the target vehicle imaged in each frame of image; converting the two-dimensional position information imaged by the target vehicle in each frame of image into three-dimensional position information of the target vehicle in a world coordinate system; and obtaining the distance change condition between the target vehicle and the current vehicle according to the three-dimensional position information of the target vehicle and the current vehicle at a plurality of moments, wherein the plurality of moments are the acquisition moments of the continuous multi-frame images.

The distance change between the target vehicle and the current vehicle may include that the distance between the target vehicle and the current vehicle is larger and larger, or the distance between the target vehicle and the current vehicle is smaller and smaller, or the distance between the target vehicle and the current vehicle is kept unchanged, and the like. Wherein the distance includes a lateral distance and a longitudinal distance.

For each frame of image in the continuous multi-frame image, the current vehicle can perform vehicle detection on each frame of image to obtain two-dimensional position information of the target vehicle imaged in each frame of image, and further obtain the relative distance between the target vehicle and the current vehicle in the world coordinate system according to the corresponding relation of the positions of the tracking target in the pixel coordinate system and the world coordinate system. For continuous multi-frame images, the relative distance change condition of the target vehicle and the current vehicle in the world coordinate system can be obtained.

For the size change rule of the target vehicle imaged in the image, in one possible implementation, the acquiring process may include: acquiring the size of the target vehicle imaged in each frame of image based on the continuous multi-frame image acquired by the front-mounted image sensor; and obtaining the size change rule according to the size of the target vehicle imaged in each frame of image.

The size change rule may be that the size of the image of the target vehicle in the image is larger and larger, or the size of the image of the target vehicle in the image is smaller and smaller, or the size of the image of the target vehicle in the image remains unchanged.

For each frame of image in the continuous multi-frame image, the current vehicle can perform vehicle detection on each frame of image to obtain an imaged area of the target vehicle in each frame of image, for example, vehicle detection is performed on each frame of image through a vehicle detection model to obtain the whole imaged area of the target vehicle in each frame of image, that is, a vehicle area, the vehicle area can be marked by a target frame, and the size of the target frame is taken as the imaged size of the target vehicle in the image. And obtaining the size change rule of the target vehicle in the continuous multi-frame images according to the size of the target vehicle in each frame image.

For the distance between the position imaged by the target vehicle in the image and the edge of the image, in one possible implementation, the acquiring process may include: acquiring the distance between the position imaged by the target vehicle in the frame of image and the edge of the image based on the frame of image acquired by the front-mounted image sensor; or, based on the continuous multi-frame images acquired by the front-facing image sensor, acquiring the distance between the position imaged by the target vehicle in each frame of image and the edge of the image, and averaging all the acquired distances.

After the current vehicle acquires a frame of image through a front image sensor of the current vehicle, vehicle detection can be performed on the frame of image to obtain the position of the target vehicle imaged in the frame of image, and the distance between the position and the image edge of the frame of image is calculated. Of course, the current vehicle may also calculate the distance between the position where the target vehicle is imaged in each frame of image and the edge of the image for consecutive frames of images, and then calculate the average value of the distances.

For the wheel condition of the target vehicle imaged in the image, in one possible implementation, the acquiring process may include: acquiring the number of wheels imaged in one frame of image by the target vehicle based on the one frame of image acquired by the front-mounted image sensor; or acquiring the number of wheels imaged in each frame of image of the target vehicle based on the continuous multi-frame images acquired by the front-end image sensor, and calculating the average value of the number of wheels corresponding to the continuous multi-frame images.

After the current vehicle acquires a frame of image through a front image sensor of the current vehicle, vehicle detection can be performed on the frame of image to obtain a vehicle area imaged by the target vehicle in the frame of image, and wheel detection is performed on the vehicle area to obtain the number of wheels in the vehicle area. Of course, the current vehicle may also calculate the number of wheels imaged by the target vehicle in each frame of image for consecutive multiple frames of images, and then calculate the average value of the number of wheels.

And b, acquiring the weight of the target vehicle belonging to the adjacent road side vehicle according to the at least one item of information.

In the embodiment of the invention, all the information acquired by the current vehicle through the step a can be used for judging whether the target vehicle belongs to the adjacent side vehicle.

In one possible implementation, the obtaining of the weight of the target vehicle belonging to the adjacent-lane side vehicle may include: the current vehicle may perform the step of obtaining the weight according to each of the at least one item of information, respectively. And multiplying all the obtained weights by respective coefficients and then adding the obtained weights to obtain the weight of the target vehicle belonging to the adjacent lane side vehicle.

The collision early warning system of the current vehicle can obtain the weight of the target vehicle belonging to the side vehicle of the adjacent road according to the distance change condition between the target vehicle and the current vehicle, and the weight is recorded as a first weight. In the driving process of the current vehicle, the part of the adjacent road side vehicle staying in the camera view field is smaller and smaller, and the imaging position in the image accords with the movement trend gradually approaching to the edge of the image, so that the longitudinal distance between the target vehicle and the current vehicle is smaller and the transverse distance is larger and larger based on the imaging position of the target vehicle in the image, and the variation of the transverse distance is larger, namely the ratio of the variation of the transverse distance to the variation of the longitudinal distance is larger than a certain threshold value. Therefore, the first weight may be larger as the distance variation is such that the longitudinal distance and the lateral distance are larger.

The collision early warning system of the current vehicle can obtain the weight of the target vehicle belonging to the adjacent road side vehicle according to the size change rule of the target vehicle in the image, and the weight is recorded as a second weight. When a target vehicle running on a lane of the current vehicle approaches the current vehicle, the part staying in the camera view field is larger and larger, the size of an image formed by the target vehicle in the image is larger and larger, and the part staying in the camera view field of the adjacent lane side vehicle is smaller and smaller along with the approach of the adjacent lane side vehicle to the current vehicle, and the size of the image formed in the image is smaller and smaller. Therefore, the size change rule can be used to distinguish the vehicle traveling in the same lane from the vehicle on the side of the adjacent lane, and the second weight can be larger as the size of the target vehicle imaged in the image becomes smaller and smaller.

The collision early warning system of the current vehicle can obtain the weight of the target vehicle belonging to the adjacent lane side vehicle according to the distance between the position of the target vehicle imaged in the image and the edge of the image, and the weight is recorded as a third weight. The adjacent-lane side vehicle is generally located close to the edge of the image, and therefore, the third weight may be larger as the distance from the edge of the image to the position where the target vehicle is imaged in the image is smaller.

The collision early warning system of the current vehicle can obtain the weight of the target vehicle belonging to the adjacent lane side vehicle according to the condition of the wheel imaged by the target vehicle in the image, and the weight is recorded as a fourth weight. The imaging of the adjacent lane side vehicle in the image is mainly the vehicle side portion, and therefore, the fourth weight may be larger as the number of wheels is larger.

Further, the current vehicle may sum the first weight, the second weight, the third weight, and the fourth weight by multiplying the respective coefficients. Of course, the summation of the weights multiplied by the corresponding coefficients is only an example, and the current vehicle may also use other algorithms to obtain the weight of the target vehicle belonging to the adjacent lane side vehicle, which is not limited in the embodiment of the present invention.

The method for detecting the side vehicle of the adjacent road can avoid the problem of false alarm of the side vehicle of the adjacent road by comprehensively obtaining the weight of the side vehicle of the adjacent road according to the distance change condition between the target vehicle and the current vehicle, the size change rule of the image of the target vehicle in the image, the wheel condition, the distance between the image position and the edge of the image and the like, wherein the weight of the side vehicle of the adjacent road can influence the probability of triggering the alarm of the target vehicle to different degrees.

After the weight of the target vehicle belonging to the adjacent side vehicle is obtained, the collision early warning system of the current vehicle can compare the weight with the target weight, and execute subsequent different steps according to different comparison results. If the weight of the target vehicle belonging to the adjacent lane side vehicle is greater than or equal to the target weight, step 203 is executed, otherwise, step 204 is executed.

203. When the weight of the target vehicle belonging to the adjacent road side vehicle is greater than or equal to the target weight, the probability of triggering the alarm by the target vehicle is zero.

In the embodiment of the invention, the weight of the target vehicle belonging to the side vehicle of the adjacent road can influence the probability of triggering the alarm of the target vehicle to different degrees, and if the weight of the target vehicle belonging to the side vehicle of the adjacent road is higher, the probability of triggering the alarm of the target vehicle is lower. The collision early warning system of the target vehicle can be preset with a target weight, when the weight of the target vehicle belonging to the adjacent lane side vehicle reaches the target weight, the target vehicle is represented to belong to the adjacent lane side vehicle, and under the condition, the probability that the target vehicle triggers the alarm is zero, and the collision early warning system can not give an alarm for the target vehicle.

Obviously, if there is only one target vehicle and the weight of the one target vehicle belonging to the adjacent lane side vehicle is greater than or equal to the target weight, the collision warning system does not issue an alarm signal. If there are multiple target vehicles and the weights of all the vehicles are greater than or equal to the target weights, the collision warning system will not send out the warning signal, and if the weight of any one of the multiple target vehicles is less than the target weight, the collision warning system may obtain the probability that the vehicle triggers the warning in step 204.

204. And when the weight of the target vehicle belonging to the side vehicle of the adjacent road is smaller than the target weight, obtaining the probability of triggering the alarm by the target vehicle according to the weight of the target vehicle belonging to the side vehicle of the adjacent road and the relative motion condition of the target vehicle and the current vehicle.

In the embodiment of the invention, when the weight of the target vehicle belonging to the adjacent lane side vehicle is smaller than the target weight, the target vehicle may not belong to the adjacent lane side vehicle, and under the condition, the collision early warning system of the target vehicle can judge whether the target vehicle needs to be alarmed or not by combining the weight of the target vehicle belonging to the adjacent lane side vehicle and other information influencing alarming.

Wherein, the relative motion situation of the target vehicle and the current vehicle can comprise at least one of a relative distance, a relative speed and a time interval between the current moment and the collision moment. For example, the current vehicle may obtain the relative distance and the relative speed between the current vehicle and the target vehicle through analysis of an image acquired by the front-mounted image sensor, and may further obtain the time interval between the current time and the time when the current vehicle collides with the target vehicle through calculation of the relative distance and the relative speed.

In a possible implementation manner, obtaining the probability that the target vehicle triggers the alarm according to the weight that the target vehicle belongs to the adjacent lane side vehicle and the relative motion condition of the target vehicle and the current vehicle comprises: and respectively obtaining a first probability and a second probability of triggering the alarm by the target vehicle according to the weight of the target vehicle belonging to the adjacent road side vehicle and the relative motion condition of the target vehicle and the current vehicle, and carrying out weighted summation on the first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

The method comprises the steps that a current vehicle can obtain a probability (first probability) of triggering alarm of a target vehicle according to the weight of the target vehicle belonging to a vehicle on the side of an adjacent road in a preset probability calculation mode, obtain a probability (second probability) of triggering alarm of the target vehicle according to the relative motion condition of the target vehicle and the current vehicle, then carry out weighted summation on the two probabilities to obtain a comprehensive probability, and take the comprehensive probability as the probability of triggering alarm of the target vehicle.

Aiming at the condition that the weight of a target vehicle corresponding to the current vehicle, which belongs to a side vehicle on the adjacent road, is smaller than the target weight, the collision early warning system of the current vehicle can obtain the comprehensive probability of triggering the alarm of the target vehicle by combining the weight of the target vehicle, which belongs to the side vehicle on the adjacent road, and the relative motion condition of the target vehicle and the current vehicle, and when the comprehensive probability of triggering the alarm of the target vehicle is larger than the target probability, the alarm is given to the target vehicle;

aiming at the condition that the weights of a plurality of target vehicles corresponding to the current vehicle, which belong to the adjacent lane side vehicle, are smaller than the target weights, the collision early warning system of the current vehicle can obtain the comprehensive probability of triggering alarm of each target vehicle by combining the weight of each target vehicle, which belongs to the adjacent lane side vehicle, and the relative motion condition of each target vehicle and the current vehicle, so that the target vehicle needing alarm is selected to alarm. For example, according to the integrated probability of the target vehicles triggering the alarm, the target vehicle with the highest integrated probability and larger than the target probability in the target vehicles is selected to trigger the alarm.

For example, for a plurality of target vehicles, after the collision warning system of the current vehicle obtains the comprehensive probability of each target vehicle in the plurality of target vehicles, if only one target vehicle with the comprehensive probability greater than the target probability is in the plurality of target vehicles, the collision warning system alarms the target vehicle. If a plurality of target vehicles with the comprehensive probability larger than the target probability exist in the plurality of target vehicles, selecting one target vehicle with the maximum comprehensive probability from the plurality of target vehicles to alarm, and if target vehicles with the same comprehensive probability exist, selecting a target vehicle with a high probability determined according to the weight of the vehicle belonging to the side of the adjacent road (namely, a target vehicle with a low weight belonging to the vehicle belonging to the side of the adjacent road) to alarm.

For example, if the combined probability that two target vehicles in the plurality of target vehicles trigger the alarm is the same and is greater than the combined probability that other target vehicles trigger the alarm, taking the two target vehicles as a first vehicle and a second vehicle as an example, if the first vehicle has a smaller weight of belonging to the side vehicle of the adjacent lane than the second vehicle, the first vehicle is determined to be the vehicle which needs to trigger the alarm currently.

It should be noted that, the foregoing steps 203 to 204 are one possible implementation manner of obtaining the probability that the target vehicle triggers the alarm according to the weight of the target vehicle belonging to the adjacent-lane side vehicle. The alarm is not triggered when the weight of the target vehicle belonging to the adjacent side vehicle is large, false alarm of the adjacent side vehicle can be avoided, when the weight of the target vehicle belonging to the adjacent side vehicle is small, the weight of the target vehicle belonging to the adjacent side vehicle and other information influencing alarm probability are combined to determine whether to trigger the alarm, the alarm performance of a collision early-warning system is optimized, the reliability and accuracy of the alarm can be guaranteed, and the user experience is improved.

In the related art, whether collision early warning is performed is determined only by calculating a time interval between the current time and the time of collision, as shown in fig. 3 and 4, the images shown in fig. 3 and 4 may be images of vehicles ahead, which are collected by a front image sensor of the current vehicle, as shown in fig. 3, the vehicles B and a are both tails of vehicles, which are both vehicles overlapping with the current vehicle (host vehicle), and both the vehicles B and a are traveling in the lane (the lane where the current vehicle is located), so that the vehicle closest to the current vehicle in the lane is the vehicle B, and if the time interval between the current time and the time of collision between the current vehicle and the vehicle B is less than an early warning time threshold value, the early warning of collision between the current vehicle and the current vehicle is triggered. However, as shown in fig. 4, the vehicle B is a vehicle on the side of the adjacent lane, the target frame is the side of the vehicle B, the tail of the vehicle B does not overlap with the current vehicle, and there is no danger of collision, at this time, it should be determined that the target a is the vehicle closest to the current vehicle on the lane, and it is determined whether to alarm according to the time interval between the current time and the time when the current vehicle collides with the vehicle a, but since the vehicle posture is not distinguished when detecting the vehicle ahead in the related art, the target frame information obtained in the two cases shown in fig. 3 and fig. 4 is completely consistent, so that in the case shown in fig. 4, the related art still determines that the vehicle closest to the current vehicle on the lane is the vehicle B, and still calculates the time interval between the current time and the time when the current vehicle collides with the vehicle B, if the calculated time interval is smaller than the early warning, the early warning of the collision of the front vehicle is triggered, so that the false alarm of the vehicle B is caused, the problem of the false alarm of an adjacent road is introduced, and meanwhile, the false alarm of the target A is caused if the vehicle A should give an alarm.

In the embodiment of the invention, whether collision early warning is carried out or not is determined by acquiring the weight of the vehicle which runs ahead and belongs to the side vehicle of the adjacent lane, and for the two conditions shown in fig. 3 and 4, the weight of the vehicle A and the weight of the vehicle B which belong to the side vehicle of the adjacent lane are acquired, so that under the condition shown in fig. 4, the weight of the vehicle B which belongs to the side vehicle of the adjacent lane is judged to be larger, the vehicle is considered to be the side vehicle of the adjacent lane, no collision danger exists, and no alarm is carried out, so that the problem of false alarm of the adjacent lane is avoided.

Fig. 5 is a schematic structural diagram of a collision warning apparatus according to an embodiment of the present invention. Referring to fig. 5, the apparatus includes:

the determining module 501 is configured to determine a target vehicle corresponding to a current vehicle, where the target vehicle is a vehicle traveling ahead of the current vehicle;

an obtaining module 502, configured to obtain a weight that the target vehicle belongs to a vehicle on a side of an adjacent lane, where a lane where the vehicle on the side of the adjacent lane is located is adjacent to a lane where the current vehicle is located;

and the alarm module 503 is configured to obtain the probability that the target vehicle triggers an alarm according to the weight of the target vehicle belonging to the vehicle on the side of the adjacent lane.

In one possible implementation, the alarm module 503 is configured to:

when the weight of the target vehicle belonging to the side vehicle of the adjacent lane is smaller than the target weight, obtaining the probability of triggering the alarm of the target vehicle according to the weight of the target vehicle belonging to the side vehicle of the adjacent lane and the relative motion condition of the target vehicle and the current vehicle, wherein the relative motion condition comprises at least one of the relative distance between the target vehicle and the current vehicle, the relative speed and the time interval of the collision moment of the distance at the current moment.

In a possible implementation manner, the alarm module is configured to obtain a first probability and a second probability that the target vehicle triggers an alarm according to the weight of the target vehicle belonging to the adjacent lane side vehicle and the relative motion condition; and carrying out weighted summation on the first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

In one possible implementation, the alarm module is configured to:

and when the current vehicle corresponds to a plurality of target vehicles, selecting the target vehicle with the maximum comprehensive probability and greater than the target probability from the plurality of target vehicles for alarming according to the comprehensive probability of triggering the alarm by the plurality of target vehicles.

In one possible implementation, the obtaining module 502 is configured to:

In one possible implementation, the obtaining module is configured to:

In one possible implementation, the obtaining module 502 is configured to:

In the embodiment of the invention, the probability of triggering the alarm by the target vehicle is obtained by obtaining the weight of the target vehicle running ahead, which belongs to the side vehicle of the adjacent road, and the weight can influence the probability of triggering the alarm by the target vehicle to different degrees. According to the method for realizing collision early warning by detecting the adjacent-road side vehicles, when a plurality of vehicles are driven by the front side, the vehicle postures can be distinguished according to the weight of the vehicles belonging to the adjacent-road side vehicles, the vehicles which are driven on the adjacent road and have no overlapping tail with the current vehicle are determined from the plurality of vehicles, and the vehicles are the adjacent-road side vehicles and have no collision danger, so that no alarm is given, and the false alarm of the adjacent-road side vehicles is avoided.

It should be noted that: the collision warning device provided in the above embodiment is exemplified by only dividing the functional modules in collision warning, and in practical applications, the function distribution may be completed by different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above-described functions. In addition, the collision early warning device and the collision early warning method provided by the above embodiments belong to the same concept, and specific implementation processes thereof are detailed in the method embodiments and are not described herein again.

Fig. 6 is a schematic structural diagram of a computer device 600 according to an embodiment of the present invention, where the computer device 600 may be configured in a vehicle for executing the collision warning method provided in the foregoing embodiments. The computer device 600 may have a relatively large difference due to different configurations or performances, and may include one or more processors (CPUs) 601 and one or more memories 602, where the memory 602 stores at least one instruction, and the at least one instruction is loaded and executed by the processor 601 to implement the collision warning method provided by the above-mentioned method embodiments. Certainly, the computer device may further have components such as a wired or wireless network interface, a keyboard, and an input/output interface, so as to perform input and output, and the computer device may further include other components for implementing the functions of the device, which is not described herein again.

In an exemplary embodiment, a computer-readable storage medium, such as a memory, storing at least one instruction, which when executed by a processor, implements the collision warning method in the above embodiments, is also provided. For example, the computer-readable storage medium may be a Read-Only Memory (ROM), a Random Access Memory (RAM), a Compact Disc Read-Only Memory (CD-ROM), a magnetic tape, a floppy disk, an optical data storage device, and the like.

It will be understood by those skilled in the art that all or part of the steps for implementing the above embodiments may be implemented by hardware, or may be implemented by a program instructing relevant hardware, where the program may be stored in a computer-readable storage medium, and the above-mentioned storage medium may be a read-only memory, a magnetic disk or an optical disk, etc.

The above description is only exemplary of the present invention and should not be taken as limiting the invention, as any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A collision warning method, comprising:

2. The method of claim 1, wherein obtaining the probability that the target vehicle triggers an alarm based on the weight of the target vehicle belonging to an adjacent-lane side vehicle comprises:

3. The method according to claim 2, wherein the obtaining the probability that the target vehicle triggers the alarm according to the weight that the target vehicle belongs to the adjacent lane side vehicle and the relative motion situation of the target vehicle and the current vehicle comprises:

4. The method of claim 3, wherein after obtaining the composite probability that the target vehicle will trigger an alert, the method further comprises:

5. The method of claim 1, wherein the obtaining the weight of the target vehicle belonging to the adjacent lane side vehicle comprises:

6. The method of claim 5, wherein the obtaining the weight of the target vehicle belonging to the adjacent-lane side vehicle according to the at least one item of information comprises:

7. The method of claim 5, wherein obtaining a change in distance between the target vehicle and the current vehicle comprises:

8. The method of claim 5, wherein obtaining a dimensional change law of the target vehicle imaged in the image comprises:

9. The method of claim 5, wherein obtaining a distance from an image edge to a location where the target vehicle is imaged in the image comprises:

10. The method of claim 5, wherein obtaining the wheel condition of the target vehicle imaged in the image comprises:

11. A collision warning apparatus, characterized in that the apparatus comprises:

12. The apparatus of claim 11, wherein the alarm module is configured to:

13. The device of claim 12, wherein the alarm module is configured to obtain a first probability and a second probability that the target vehicle triggers an alarm according to the weight of the target vehicle belonging to the adjacent side vehicle and the relative motion condition, respectively; and carrying out weighted summation on the first probability and the second probability to obtain the comprehensive probability of triggering the alarm by the target vehicle.

14. The apparatus of claim 13, wherein the alarm module is configured to:

15. The apparatus of claim 11, wherein the obtaining module is configured to:

and acquiring the weight of the target vehicle belonging to the adjacent road vehicle according to the at least one item of information.

16. The apparatus of claim 15, wherein the obtaining module is configured to:

17. The apparatus of claim 15, wherein the obtaining module is configured to:

18. The apparatus of claim 15, wherein the obtaining module is configured to:

19. The apparatus of claim 15, wherein the obtaining module is configured to:

20. The apparatus of claim 15, wherein the obtaining module is configured to:

21. A computer device comprising a processor and a memory; the memory is used for storing at least one instruction; the processor, configured to execute at least one instruction stored on the memory to implement the method steps of any of claims 1-10.

22. A computer-readable storage medium, having stored therein at least one instruction, which when executed by a processor, performs the method steps of any one of claims 1-10.