CN117734680B - Blind area early warning method, system and storage medium for large vehicle - Google Patents
Blind area early warning method, system and storage medium for large vehicle Download PDFInfo
- Publication number
- CN117734680B CN117734680B CN202410087372.6A CN202410087372A CN117734680B CN 117734680 B CN117734680 B CN 117734680B CN 202410087372 A CN202410087372 A CN 202410087372A CN 117734680 B CN117734680 B CN 117734680B
- Authority
- CN
- China
- Prior art keywords
- vehicle
- blind area
- obstacle
- early warning
- future
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000012545 processing Methods 0.000 claims abstract description 27
- 230000004888 barrier function Effects 0.000 claims abstract description 10
- 238000001514 detection method Methods 0.000 claims abstract description 8
- 238000003384 imaging method Methods 0.000 claims description 16
- 230000003068 static effect Effects 0.000 claims description 4
- 230000002123 temporal effect Effects 0.000 claims description 4
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000013519 translation Methods 0.000 claims description 3
- 238000004590 computer program Methods 0.000 claims description 2
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 230000036544 posture Effects 0.000 description 3
- 230000002441 reversible effect Effects 0.000 description 3
- 239000000523 sample Substances 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 244000025254 Cannabis sativa Species 0.000 description 1
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000010813 municipal solid waste Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 210000000056 organ Anatomy 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 230000007723 transport mechanism Effects 0.000 description 1
Landscapes
- Traffic Control Systems (AREA)
Abstract
The invention discloses a large vehicle blind area early warning method, a large vehicle blind area early warning system and a storage medium, and belongs to intelligent automobile systems. The method comprises the following steps: obtaining a vehicle blind area based on the vehicle state information and the inherent parameters of the vehicle body, wherein the vehicle blind area comprises a current blind area and a future blind area; the future blind area is a blind area of the vehicle at the future moment, the future moment is obtained according to an early warning time value, and the early warning time value is a time value obtained by acquiring weather data; acquiring radar reflection data and video image data in a blind area of a vehicle based on at least one radar and at least one camera; performing space and time matching processing on radar reflection data and video image data, and classifying the video image data based on a target detection model to obtain barrier data; calculating a collision index of the obstacle in a future blind area according to the type and the state information of the obstacle data; and sending out corresponding early warning information according to the early warning strategy. The invention improves the accuracy and the early warning efficiency of the dead zone early warning.
Description
Technical Field
The invention relates to the technical field of intelligent automobile systems, in particular to a large-scale vehicle blind area early warning method, a large-scale vehicle blind area early warning system and a storage medium.
Background
Because of the dead zone on the right side of the large-sized vehicle and the internal wheel difference, the driver is limited in view when turning right, and cannot observe the traffic condition on the right side of the vehicle through the rearview mirror in time, so that collision with other vehicles and pedestrians is caused, and serious traffic accidents are caused. Therefore, with the increasingly finer and stricter requirements of safe driving, blind zone early warning has become the standard of large vehicles. The current blind area early warning system generally adopts radar ranging early warning, through hanging into reverse gear signal or through turning to the signal, the car blind area early warning system is in the mode of working always under the reverse gear mode, and under the corresponding turn signal mode of turning on, radar work 20S, radar distance reach 3 m' S range finding, alarm warning sound changes according to the different bee calling organ of barrier distance. Thus realizing safe driving and reversing.
The existing early warning system is based on radar ranging, and is poor in recognition accuracy and high in false alarm interference rate. The improvement scheme for the radar blind area detection device is that the radar detects blind area obstacles, the camera is matched to acquire video streams, vehicles and pedestrians in the blind area are identified through target identification so as to give early warning to drivers, the false alarm rate is reduced, and the accuracy and the efficiency of the blind area early warning are improved. However, the accuracy of early warning can be reduced due to the influence of weather, especially in rainy and snowy weather, the situations that pedestrians get on umbrellas and wear raincoats are greatly increased, and the target recognition of the stationary pedestrians in the blind areas is often difficult to recognize through images. In bad weather, such as haze, rain and snow, the vision of a driver is affected, vehicles or pedestrians which do not enter blind areas are easy to ignore and react, the braking distance is increased, and the probability of collision is increased.
Disclosure of Invention
The present invention aims to solve at least one of the technical problems existing in the prior art. Therefore, the invention provides a large vehicle blind area early warning method, a large vehicle blind area early warning system and a storage medium, which can improve early warning accuracy.
On the one hand, the embodiment of the invention provides a large-scale vehicle blind area early warning method, which comprises the following steps: obtaining a vehicle blind area based on vehicle state information and inherent parameters of a vehicle body, wherein the vehicle blind area comprises a current blind area and a future blind area; the future blind area is a blind area of a vehicle at a future moment, the future moment is obtained according to an early warning time value, and the early warning time value is a time value obtained by acquiring weather data; acquiring radar reflection data and video image data in a blind area of the vehicle based on at least one radar and at least one camera; performing space and time matching processing on the radar reflection data and the video image data, and performing classification processing on the video image data based on a target detection model to obtain barrier data; the obstacle data includes a type, status information, a time stamp, and location information; calculating the collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data; the collision index is used for evaluating the possibility of collision between the vehicle and the obstacle; sending corresponding early warning information according to an early warning strategy, including: and carrying out grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area.
The method for early warning the blind area of the large vehicle at least comprises the following beneficial effects: according to the embodiment of the invention, the radar reflection data and the video stream are acquired by installing the cameras and the probes at the various positions of the vehicle body, so that whether the blind area has an obstacle or not and the type and the state of the obstacle are determined, the early warning time is determined according to the required avoidance time under different weather, the possibility of collision when the vehicle enters the blind area at the future moment is estimated according to the state of the vehicle and the type and the state of the obstacle, early warning is performed, enough avoidance time is reserved for a driver, and the early warning efficiency is improved.
According to some embodiments of the invention, calculating the collision index of the obstacle in the future blind area according to the type and state information of the obstacle data includes: if the type is a vehicle or a pedestrian and the state is moving, a first collision coefficient is taken, the distance between the current obstacle and the vehicle at the future moment is predicted according to the state information, and a collision index is obtained according to the distance and the first collision coefficient; if the type and the vehicle or the pedestrian are stationary, a second collision coefficient is taken, and a collision index is obtained according to the distance between the current obstacle and the vehicle at the future time and the second collision coefficient; if the type is an unknown obstacle, judging whether the state of the unknown obstacle is moving, if so, taking a first collision coefficient, predicting the distance between the unknown obstacle and the vehicle at the future time according to state information, and obtaining a collision index according to the distance and the first collision coefficient.
According to some embodiments of the invention, the method further comprises: if the type is an unknown obstacle and the state is static, judging whether the weather type is a special weather type according to the weather data, and if not, obtaining a collision index of 0; and if the vehicle is of a special weather type, acquiring the distance between the unknown obstacle and the vehicle at the future time, and obtaining a collision index according to the distance and a third collision coefficient.
According to some embodiments of the invention, the performing spatial and temporal matching processing on the radar reflection data and the video image data, and performing classification processing on the video image data based on a target recognition model, to obtain obstacle data includes: according to whether the reflection data of each radar has an obstacle or not, if so, taking radar reflection data and video image data which are smaller than a certain time difference threshold value as obstacle data at the same moment; the video image data are image data with timestamp information, which are obtained through each camera and subjected to video imaging; establishing a vehicle body coordinate system and acquiring own coordinates of each camera and each radar in the vehicle body coordinate system; establishing a radar imaging coordinate system and a camera imaging coordinate system, and acquiring coordinates of the reflection points relative to the radar imaging coordinate system and coordinates of the pixel points relative to the camera imaging coordinate system; converting the coordinates of the reflection points and the coordinates of the pixel points into the vehicle body coordinate system through the rotation translation matrix, and associating the reflection points and the pixel points based on the position relation of the reflection points and the pixel points relative to the vehicle body; dividing the video image data into pixel blocks, and carrying out clustering matching on the obtained pixel blocks and radar reflection points in space; matching the speed, distance and azimuth information of the upper reflection point for the divided pixel blocks; performing target recognition on the pixel blocks based on the trained target recognition model to obtain a classification result; and obtaining barrier data according to the pixel blocks and the classification result.
According to some embodiments of the present invention, the sending the corresponding early warning information according to the early warning policy includes: dividing the early warning information into multi-stage early warning information, and sequentially dividing the early warning information into the following steps from high to low according to the dangerous degree: primary early warning, which means that a moving vehicle, pedestrian or unknown obstacle exists in the current blind area; secondary early warning, namely, a stationary vehicle or pedestrian exists in a current blind area, or a vehicle, pedestrian or unknown obstacle with a collision coefficient higher than a threshold value exists in a future blind area; three-stage early warning means that a stationary unknown obstacle exists in a current blind area or a moving vehicle, pedestrian or unknown obstacle exists in a future blind area; and four-stage early warning, which indicates that stationary vehicles, pedestrians or unknown obstacles exist in future blind areas.
According to some embodiments of the present invention, the sending the corresponding early warning information according to the early warning policy includes: acquiring road condition information, adjusting an early warning strategy according to the road condition information, and sending corresponding early warning information according to the adjusted early warning strategy; the traffic information includes, but is not limited to, congestion level and road type.
According to some embodiments of the invention, the object recognition model is yolov models, including pedestrian recognition models and vehicle recognition models.
According to some embodiments of the present invention, the sending the corresponding early warning information according to the early warning policy includes: sending out early warning information according to different early warning levels, wherein the early warning information at least comprises one of the following various forms: buzzing, voice broadcasting, flashing of alarm lamps and LED projection.
Another aspect of the embodiment of the present invention provides a large vehicle blind area early warning system, including: the system comprises a first module, a second module and a third module, wherein the first module is used for obtaining a vehicle blind area based on vehicle state information and inherent parameters of a vehicle body, and the vehicle blind area comprises a current blind area and a future blind area; the future blind area is a blind area of a vehicle at a future moment, the future moment is obtained according to an early warning time value, and the early warning time value is a time value obtained by acquiring weather data; the second module is used for acquiring radar reflection data and video image data in the blind area of the vehicle based on at least one radar and at least one camera; the third module is used for carrying out space and time matching processing on the radar reflection data and the video image data, and carrying out classification processing on the video image data based on a target detection model to obtain barrier data; the obstacle data includes a type, status information, a time stamp, and location information; a fourth module, configured to calculate a collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data; the collision index is used for evaluating the possibility of collision between the vehicle and the obstacle; a fifth module, configured to send corresponding early warning information according to an early warning policy, including: and carrying out grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area.
The large-scale vehicle blind area early warning system provided by the embodiment of the invention at least comprises the following beneficial effects: according to the embodiment of the invention, the radar reflection data and the video stream are acquired by installing the cameras and the probes at the various positions of the vehicle body, so that whether the blind area has an obstacle or not and the type and the state of the obstacle are determined, the early warning time is determined according to the required avoidance time under different weather, the possibility of collision when the vehicle enters the blind area at the future moment is estimated according to the state of the vehicle and the type and the state of the obstacle, early warning is performed, enough avoidance time is reserved for a driver, and the early warning efficiency is improved.
Another aspect of the embodiments of the present invention provides a computer-readable storage medium having stored thereon a computer program which, when executed by a processor, implements the above-described large vehicle blind area warning method.
Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention.
Drawings
The foregoing and/or additional aspects and advantages of the invention will become apparent and may be better understood from the following description of embodiments taken in conjunction with the accompanying drawings in which:
FIG. 1 is a flow chart of a method according to an embodiment of the invention;
FIG. 2 is a schematic view of a large vehicle and its blind zone locations for which embodiments of the present invention are applied;
fig. 3 is a block schematic diagram of a system according to an embodiment of the invention.
Reference numerals:
The first module 100, the second module 200, the third module 300, the fourth module 400, the fifth module 500.
Detailed Description
Embodiments of the present invention are described in detail below, examples of which are illustrated in the accompanying drawings, wherein like or similar reference numerals refer to like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative only and are not to be construed as limiting the invention.
In the description of the present invention, a plurality means one or more, and a plurality means two or more, and it is understood that greater than, less than, exceeding, etc. does not include the present number, and it is understood that greater than, less than, within, etc. include the present number. The description of the first and second is for the purpose of distinguishing between technical features only and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or implicitly indicating the precedence of the technical features indicated.
Referring to fig. 2, the embodiment of the present invention is applied to a large vehicle, and the blind area of the vehicle mainly includes a head blind area, a front right view blind area, and a rear view blind area. When the vehicle is in reverse gear or turned, the vehicle is liable to collide with pedestrians or vehicles in the blind area. According to the embodiment of the invention, the radar reflection data and the video stream are acquired by installing a plurality of cameras and probes at each position of the head and the tail of the vehicle, whether the blind area has an obstacle or not is determined by processing the reflection data and the video stream, the type and the state of the obstacle are determined, and the grading early warning is carried out according to the acquired obstacle data. In order to improve early warning efficiency and avoid that a driver cannot avoid collision after an obstacle enters a blind area, the embodiment of the invention determines early warning time according to the required avoidance time under different weather, evaluates the possibility of collision when entering the blind area at future time according to the state of a vehicle and the type and state of the obstacle, and performs early warning in advance so as to avoid that the driver cannot avoid collision.
Referring to fig. 1, an embodiment of the present invention provides a method for early warning blind areas of a large vehicle, including the following steps:
S100, obtaining a vehicle blind area based on vehicle state information and inherent parameters of a vehicle body, wherein the vehicle blind area comprises a current blind area and a future blind area; the current dead zone is dynamically calculated through intrinsic parameters of the vehicle body including wheelbase, vehicle money, vehicle length, vehicle height and the like, and the future dead zone is calculated into the position, the posture and the future dead zone position of the vehicle at the future moment based on the state information of the vehicle dynamics model, such as the current speed, the posture, the steering wheel angle and the like. The future time of the embodiment of the invention is obtained according to the early warning time value, wherein the early warning time value is obtained by obtaining weather data. In some embodiments, the weather data includes current weather types, such as sunny days, cloudy days, haze, rain and snow, and the like, and different types of weather correspond to different early warning time values, where the early warning time values may be time values set by comprehensive factors, such as visibility of haze weather, high temperature, braking time of rain and snow weather, and the like.
S200, radar reflection data and video image data in a blind area of the vehicle are acquired based on at least one radar and at least one camera. According to the embodiment of the invention, the video stream is acquired through the camera, and video image data is obtained through video imaging processing. The radar and the camera of the embodiment of the invention can be arranged in the middle of the vehicle head, the right side of the vehicle head, the vehicle tail and the like, and any number of radars and cameras can be arranged according to the precision requirements of obstacle recognition and early warning.
S300, performing space and time matching processing on radar reflection data and video image data, and performing classification processing on the video image data based on a target detection model to obtain barrier data; the obstacle data includes a type, status information, a time stamp, and location information; the object recognition model of the present embodiment is yolov models including a pedestrian recognition model and a vehicle recognition model. Particularly, the pedestrian recognition model of the embodiment can recognize pedestrians with different postures in different environments, and can recognize pedestrians riding bicycles.
S400, calculating a collision index of the obstacle in a future blind area according to the type and state information of the obstacle data; the collision index is used to evaluate the likelihood of a collision between a vehicle and an obstacle. Since moving pedestrians and vehicles can accelerate, decelerate and steer at any time, the possibility of collision with the vehicle at the future time can only be estimated according to the speed, the gesture and the distance of the current pedestrians and vehicles. In particular, stationary pedestrians and vehicles may move at a future time according to the pre-warning requirement, and the present embodiment predicts the possibility of collision according to the distance between the vehicle and the stationary pedestrian at the future time.
S500, corresponding early warning information is sent out according to an early warning strategy, and the method comprises the following steps: and carrying out grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area.
In some embodiments, calculating the collision index of the obstacle in the future blind zone according to the type and state information of the obstacle data comprises: if the type is a vehicle or a pedestrian and the state is moving, a first collision coefficient is taken, the distance between the current obstacle and the vehicle at the future moment is predicted according to the state information, and a collision index is obtained according to the distance and the first collision coefficient; if the type, the vehicle or the pedestrian is stationary, a second collision coefficient is taken, and a collision index is obtained according to the distance between the current obstacle and the vehicle at the future moment and the second collision coefficient; if the type is an unknown obstacle, judging whether the state of the unknown obstacle is moving, if so, taking a first collision coefficient, predicting the distance between the unknown obstacle and the vehicle at the future time according to the state information, and obtaining a collision index according to the distance and the first collision coefficient. In this embodiment, the first collision coefficient is larger than the second collision coefficient for evaluating the possibility of a collision in the future based on the current moving or stationary state of the vehicle or the pedestrian. In this embodiment, the unknown obstacle may be a motorcycle, an animal, an umbrella-supporting pedestrian or a raincoat pedestrian which cannot be identified by the current target identification model, and therefore the treatment of the moving unknown obstacle is to take the same first collision coefficient as the moving vehicle and pedestrian.
In some embodiments, the method of the embodiments of the present invention further comprises: if the type is an unknown obstacle and the state is static, judging whether the weather type is a special weather type according to weather data, and if the weather type is not the special weather type, obtaining a collision index of 0; if the vehicle is of a special weather type, the distance between the unknown obstacle and the vehicle at the future time is obtained, and the collision index is obtained according to the distance and the third collision coefficient. Generally, the stationary unknown obstacle of the embodiments of the present invention is a low-risk obstacle such as a tree, a grass, a trash can, a railing, etc. The method for warning the stationary unknown obstacle generally processes filtering the unknown obstacle so as to avoid causing a large number of false alarms. But the present embodiment performs special processing for the collision coefficient of an unknown obstacle in special weather. The special weather type of this embodiment may be a rainy or snowy weather in which the probability of a pedestrian or rider to prop up an umbrella or wear a raincoat is greatly increased, which makes recognition of a target difficult, and is likely to recognize the pedestrian as an unknown obstacle. Therefore, the present embodiment takes the third collision coefficient for the stationary unknown obstacle in special weather, based on the certain probability that the unknown obstacle is a pedestrian.
It will be appreciated that the first, second, and third crash coefficients of the embodiments of the present invention are merely used to evaluate the likelihood of a crash, and are not practical. The collision coefficient of the present embodiment and the collision threshold for judging the pre-warning level may be empirically set by a professional. The collision coefficient and the threshold value can also be evaluated and adjusted by a large amount of practical application data.
In some embodiments, performing spatial and temporal matching processing on the radar reflection data and the video image data, and classifying the video image data based on the object recognition model, the obtaining obstacle data includes: according to whether the reflection data of each radar has an obstacle or not, if so, taking radar reflection data and video image data which are smaller than a certain time difference threshold value as obstacle data at the same moment; the video image data is image data with time stamp information which is obtained through video imaging by each camera. Establishing a vehicle body coordinate system and acquiring own coordinates of each camera and each radar in the vehicle body coordinate system; and establishing a radar imaging coordinate system and a camera imaging coordinate system, and acquiring coordinates of the reflection points relative to the radar imaging coordinate system and coordinates of the pixel points relative to the camera imaging coordinate system. And converting the coordinates of the reflection points and the coordinates of the pixel points into a vehicle body coordinate system through rotating the translation matrix, and associating the reflection points and the pixel points based on the position relation of the reflection points and the pixel points relative to the vehicle body. Dividing the video image data into pixel blocks, and carrying out clustering matching on the obtained pixel blocks and radar reflection points in space; the speed, distance and azimuth information of the upper reflection point are matched for the divided pixel blocks. And carrying out target recognition on the pixel blocks based on the trained target recognition model to obtain a classification result. And obtaining barrier data according to the pixel blocks and the classification result.
In some embodiments, sending the corresponding early warning information according to the early warning policy includes: dividing the early warning information into multi-stage early warning information, and sequentially dividing the early warning information into the following steps from high to low according to the dangerous degree: primary early warning, which means that a moving vehicle, pedestrian or unknown obstacle exists in the current blind area; secondary early warning, namely, a stationary vehicle or pedestrian exists in a current blind area, or a vehicle, pedestrian or unknown obstacle with a collision coefficient higher than a threshold value exists in a future blind area; three-stage early warning means that a stationary unknown obstacle exists in a current blind area or a moving vehicle, pedestrian or unknown obstacle exists in a future blind area; and four-stage early warning, which indicates that stationary vehicles, pedestrians or unknown obstacles exist in future blind areas.
In some embodiments, sending the corresponding early warning information according to the early warning policy includes: acquiring road condition information, adjusting an early warning strategy according to the road condition information, and sending corresponding early warning information according to the adjusted early warning strategy; the traffic information includes, but is not limited to, congestion level and road type. According to the embodiment of the invention, the early warning strategy is adjusted according to the road condition information, for example, on a road surface with serious congestion, the collision is easy to occur in a dead zone of a vehicle, and the early warning level is improved for low-speed or stationary pedestrians or vehicles at the moment. For example, on a highway, pedestrians are greatly reduced, the possibility that the target recognition model erroneously recognizes the pedestrians as unknown obstacles is greatly reduced, and the early warning level of the unknown obstacles can be correspondingly reduced.
In some embodiments, sending the corresponding early warning information according to the early warning policy includes: and sending out early warning information according to different early warning levels, wherein the early warning information at least comprises one of the following various forms: buzzing, voice broadcasting, flashing of alarm lamps and LED projection. It will be appreciated that the magnitude and frequency of the beeping sound of the present embodiment may vary with the distance of the vehicle from the obstacle, with the distance being closer the more loud and the frequency being higher. The voice broadcasting of the embodiment can generate corresponding voice information according to the direction and the type of the obstacle.
The invention also provides embodiments of the system corresponding to the previous embodiments. For system embodiments, reference is made to the description of method embodiments for the relevant points, since they essentially correspond to the method embodiments.
Referring to fig. 3, an embodiment of the present invention provides a large vehicle blind area early warning system, including: a first module 100 for obtaining a vehicle blind area including a current blind area and a future blind area based on vehicle state information and a vehicle body intrinsic parameter; the future blind area is the blind area of the vehicle at the future moment, the future moment is obtained according to the early warning time value, and the early warning time value is the time value obtained by obtaining weather data. A second module 200 for acquiring radar reflection data and video image data in a blind zone of the vehicle based on at least one radar and at least one camera. A third module 300, configured to perform spatial and temporal matching processing on the radar reflection data and the video image data, and perform classification processing on the video image data based on the target detection model, so as to obtain obstacle data; the obstacle data includes a type, status information, a time stamp, and location information. A fourth module 400, configured to calculate a collision index of the obstacle in a future blind area according to the type and state information of the obstacle data; the collision index is used to evaluate the likelihood of a collision between a vehicle and an obstacle. A fifth module 500, configured to send corresponding early warning information according to an early warning policy, including: and carrying out grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area.
Although specific embodiments are described herein, those of ordinary skill in the art will recognize that many other modifications or alternative embodiments are also within the scope of the present disclosure. For example, any of the functions and/or processing capabilities described in connection with a particular device or component may be performed by any other device or component. In addition, while various exemplary implementations and architectures have been described in terms of embodiments of the present disclosure, those of ordinary skill in the art will recognize that many other modifications to the exemplary implementations and architectures described herein are also within the scope of the present disclosure.
The above described apparatus embodiments are merely illustrative, wherein the units illustrated as separate components may or may not be physically separate, i.e. may be located in one place, or may be distributed over a plurality of network elements. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
Those of ordinary skill in the art will appreciate that all or some of the steps, systems, and methods disclosed above may be implemented as software, firmware, hardware, and suitable combinations thereof. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, digital signal processor, or microprocessor, or as hardware, or as an integrated circuit, such as an application specific integrated circuit. Such software may be distributed on computer readable media, which may include computer storage media (or non-transitory media) and communication media (or transitory media). The term computer storage media includes both volatile and nonvolatile, removable and non-removable media implemented in any method or technology for storage of information such as computer readable instructions, data structures, program modules or other data, as known to those skilled in the art. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital Versatile Disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to store the desired information and which can be accessed by a computer. Furthermore, as is well known to those of ordinary skill in the art, communication media typically embodies computer readable instructions, data structures, program modules or other data in a modulated data signal such as a carrier wave or other transport mechanism and includes any information delivery media.
While the preferred embodiment of the present application has been described in detail, the present application is not limited to the above embodiment, and various equivalent modifications and substitutions can be made by those skilled in the art without departing from the spirit of the present application, and these equivalent modifications and substitutions are intended to be included in the scope of the present application as defined in the appended claims.
Claims (8)
1. The method for warning the blind area of the large vehicle is characterized by comprising the following steps of:
Obtaining a vehicle blind area based on vehicle state information and inherent parameters of a vehicle body, wherein the vehicle blind area comprises a current blind area and a future blind area; the future blind area is a blind area of a vehicle at a future moment, the future moment is obtained according to an early warning time value, and the early warning time value is a time value obtained by acquiring weather data;
acquiring radar reflection data and video image data in a blind area of the vehicle based on at least one radar and at least one camera;
Performing space and time matching processing on the radar reflection data and the video image data, and performing classification processing on the video image data based on a target detection model to obtain barrier data; the obstacle data includes a type, status information, a time stamp, and location information;
Calculating the collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data; the collision index is used for evaluating the possibility of collision between the vehicle and the obstacle;
Sending corresponding early warning information according to an early warning strategy, including: performing grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area;
the calculating the collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data comprises the following steps:
if the type is a vehicle or a pedestrian and the state is moving, a first collision coefficient is taken, the distance between the current obstacle and the vehicle at the future moment is predicted according to the state information, and a collision index is obtained according to the distance and the first collision coefficient;
If the type and the vehicle or the pedestrian are stationary, a second collision coefficient is taken, and a collision index is obtained according to the distance between the current obstacle and the vehicle at the future time and the second collision coefficient;
if the type is an unknown obstacle, judging whether the state of the unknown obstacle is moving, if so, taking a first collision coefficient, predicting the distance between the unknown obstacle and a vehicle at the future time according to state information, and obtaining a collision index according to the distance and the first collision coefficient;
The method further comprises the steps of: if the type is an unknown obstacle and the state is static, judging whether the weather type is a special weather type according to the weather data, and if not, obtaining a collision index of 0; and if the vehicle is of a special weather type, acquiring the distance between the unknown obstacle and the vehicle at the future time, and obtaining a collision index according to the distance and a third collision coefficient.
2. The method of claim 1, wherein the performing spatial and temporal matching on the radar reflection data and the video image data, and performing classification on the video image data based on a target recognition model, to obtain the obstacle data comprises:
According to whether the reflection data of each radar has an obstacle or not, if so, taking radar reflection data and video image data which are smaller than a certain time difference threshold value as obstacle data at the same moment; the video image data are image data with timestamp information, which are obtained through each camera and subjected to video imaging;
Establishing a vehicle body coordinate system and acquiring own coordinates of each camera and each radar in the vehicle body coordinate system; establishing a radar imaging coordinate system and a camera imaging coordinate system, and acquiring coordinates of the reflection points relative to the radar imaging coordinate system and coordinates of the pixel points relative to the camera imaging coordinate system;
converting the coordinates of the reflection points and the coordinates of the pixel points into the vehicle body coordinate system through the rotation translation matrix, and associating the reflection points and the pixel points based on the position relation of the reflection points and the pixel points relative to the vehicle body;
Dividing the video image data into pixel blocks, and carrying out clustering matching on the obtained pixel blocks and radar reflection points in space; matching the speed, distance and azimuth information of the upper reflection point for the divided pixel blocks;
performing target recognition on the pixel blocks based on the trained target recognition model to obtain a classification result;
And obtaining barrier data according to the pixel blocks and the classification result.
3. The method of claim 1, wherein the sending the corresponding pre-warning information according to the pre-warning policy comprises:
dividing the early warning information into multi-stage early warning information, and sequentially dividing the early warning information into the following steps from high to low according to the dangerous degree:
Primary early warning, which means that a moving vehicle, pedestrian or unknown obstacle exists in the current blind area;
Secondary early warning, namely, a stationary vehicle or pedestrian exists in a current blind area, or a vehicle, pedestrian or unknown obstacle with a collision coefficient higher than a threshold value exists in a future blind area;
three-stage early warning means that a stationary unknown obstacle exists in a current blind area or a moving vehicle, pedestrian or unknown obstacle exists in a future blind area;
and four-stage early warning, which indicates that stationary vehicles, pedestrians or unknown obstacles exist in future blind areas.
4. The method of claim 3, wherein the sending the corresponding pre-warning information according to the pre-warning strategy comprises:
acquiring road condition information, adjusting an early warning strategy according to the road condition information, and sending corresponding early warning information according to the adjusted early warning strategy;
The traffic information includes, but is not limited to, congestion level and road type.
5. The large vehicle blind area warning method according to claim 2, wherein the target recognition model is yolov models including a pedestrian recognition model and a vehicle recognition model.
6. The method of claim 1, wherein the sending the corresponding pre-warning information according to the pre-warning policy comprises: sending out early warning information according to different early warning levels, wherein the early warning information at least comprises one of the following various forms: buzzing, voice broadcasting, flashing of alarm lamps and LED projection.
7. A large vehicle blind zone warning system, comprising:
the system comprises a first module, a second module and a third module, wherein the first module is used for obtaining a vehicle blind area based on vehicle state information and inherent parameters of a vehicle body, and the vehicle blind area comprises a current blind area and a future blind area; the future blind area is a blind area of a vehicle at a future moment, the future moment is obtained according to an early warning time value, and the early warning time value is a time value obtained by acquiring weather data;
the second module is used for acquiring radar reflection data and video image data in the blind area of the vehicle based on at least one radar and at least one camera;
The third module is used for carrying out space and time matching processing on the radar reflection data and the video image data, and carrying out classification processing on the video image data based on a target detection model to obtain barrier data; the obstacle data includes a type, status information, a time stamp, and location information;
A fourth module, configured to calculate a collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data; the collision index is used for evaluating the possibility of collision between the vehicle and the obstacle;
the calculating the collision index of the obstacle in the future blind area according to the type and the state information of the obstacle data comprises the following steps:
if the type is a vehicle or a pedestrian and the state is moving, a first collision coefficient is taken, the distance between the current obstacle and the vehicle at the future moment is predicted according to the state information, and a collision index is obtained according to the distance and the first collision coefficient;
If the type and the vehicle or the pedestrian are stationary, a second collision coefficient is taken, and a collision index is obtained according to the distance between the current obstacle and the vehicle at the future time and the second collision coefficient;
if the type is an unknown obstacle, judging whether the state of the unknown obstacle is moving, if so, taking a first collision coefficient, predicting the distance between the unknown obstacle and a vehicle at the future time according to state information, and obtaining a collision index according to the distance and the first collision coefficient;
If the type is an unknown obstacle and the state is static, judging whether the weather type is a special weather type according to the weather data, and if not, obtaining a collision index of 0; if the vehicle is of a special weather type, acquiring the distance between the unknown obstacle and the vehicle at the future time, and acquiring a collision index according to the distance and a third collision coefficient;
A fifth module, configured to send corresponding early warning information according to an early warning policy, including: and carrying out grading early warning processing according to the type and state of the obstacle in the current blind area, the type and state of the obstacle in the future blind area and the collision index of the obstacle in the future blind area.
8. A computer readable storage medium having stored thereon a computer program which, when executed by a processor, implements the method of any of claims 1 to 6.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410087372.6A CN117734680B (en) | 2024-01-22 | 2024-01-22 | Blind area early warning method, system and storage medium for large vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410087372.6A CN117734680B (en) | 2024-01-22 | 2024-01-22 | Blind area early warning method, system and storage medium for large vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117734680A CN117734680A (en) | 2024-03-22 |
| CN117734680B true CN117734680B (en) | 2024-06-07 |
Family
ID=90277918
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410087372.6A Active CN117734680B (en) | 2024-01-22 | 2024-01-22 | Blind area early warning method, system and storage medium for large vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117734680B (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104494542A (en) * | 2014-12-22 | 2015-04-08 | 合肥晟泰克汽车电子有限公司 | Dynamic monitoring system and method for dead zone of vehicle |
| CN112216097A (en) * | 2019-07-09 | 2021-01-12 | 华为技术有限公司 | Method and device for detecting blind area of vehicle |
| CN113276769A (en) * | 2021-04-29 | 2021-08-20 | 深圳技术大学 | Vehicle blind area anti-collision early warning system and method |
| WO2021196145A1 (en) * | 2020-04-02 | 2021-10-07 | 华为技术有限公司 | Vehicle blind spot recognition method, automatic driving assistant system and smart driving vehicle comprising same |
| CN115546758A (en) * | 2022-09-14 | 2022-12-30 | 湖北三峡职业技术学院 | Method for predicting environmental collision hidden danger of construction blind area of excavator |
| CN116061807A (en) * | 2023-03-06 | 2023-05-05 | 北京理工大学前沿技术研究院 | Blind area early warning method and system based on vehicle-road information fusion |
| CN116118776A (en) * | 2023-01-03 | 2023-05-16 | 重庆长安汽车股份有限公司 | Safety control method and device for vehicle, electronic equipment and storage medium |
| CN116645831A (en) * | 2023-05-29 | 2023-08-25 | 安徽泰然信息技术有限公司 | Traffic blind area detection and early warning system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US9216688B2 (en) * | 2013-03-15 | 2015-12-22 | Adil Ansari | System and method for blindzone object detection |
-
2024
- 2024-01-22 CN CN202410087372.6A patent/CN117734680B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104494542A (en) * | 2014-12-22 | 2015-04-08 | 合肥晟泰克汽车电子有限公司 | Dynamic monitoring system and method for dead zone of vehicle |
| CN112216097A (en) * | 2019-07-09 | 2021-01-12 | 华为技术有限公司 | Method and device for detecting blind area of vehicle |
| WO2021004077A1 (en) * | 2019-07-09 | 2021-01-14 | 华为技术有限公司 | Method and apparatus for detecting blind areas of vehicle |
| WO2021196145A1 (en) * | 2020-04-02 | 2021-10-07 | 华为技术有限公司 | Vehicle blind spot recognition method, automatic driving assistant system and smart driving vehicle comprising same |
| CN113276769A (en) * | 2021-04-29 | 2021-08-20 | 深圳技术大学 | Vehicle blind area anti-collision early warning system and method |
| CN115546758A (en) * | 2022-09-14 | 2022-12-30 | 湖北三峡职业技术学院 | Method for predicting environmental collision hidden danger of construction blind area of excavator |
| CN116118776A (en) * | 2023-01-03 | 2023-05-16 | 重庆长安汽车股份有限公司 | Safety control method and device for vehicle, electronic equipment and storage medium |
| CN116061807A (en) * | 2023-03-06 | 2023-05-05 | 北京理工大学前沿技术研究院 | Blind area early warning method and system based on vehicle-road information fusion |
| CN116645831A (en) * | 2023-05-29 | 2023-08-25 | 安徽泰然信息技术有限公司 | Traffic blind area detection and early warning system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117734680A (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113276769B (en) | Vehicle blind area anti-collision early warning system and method | |
| CN113998034B (en) | Rider assistance system and method | |
| CN106652468B (en) | The detection and from vehicle violation early warning alarm set and method in violation of rules and regulations of road vehicle front truck | |
| EP3298604B1 (en) | Safety system for a vehicle to detect and warn of a potential collision | |
| US7366325B2 (en) | Moving object detection using low illumination depth capable computer vision | |
| EP2383713B1 (en) | Detecting and recognizing traffic signs | |
| CN112172663A (en) | Danger alarm method based on door opening and related equipment | |
| US20110211071A1 (en) | Method and system for evaluating brightness values in sensor images of image-evaluating adaptive cruise control systems, especially with respect to day/night distinction | |
| CN110660271B (en) | Identification of conditions of importance for driving over greater distances | |
| CN108694363A (en) | The method and apparatus that the pedestrian of vehicle periphery is detected | |
| CN114932902B (en) | Ghost probe early warning avoiding method and system based on Internet of vehicles technology | |
| Choi et al. | Cut-in vehicle warning system exploiting multiple rotational images of SVM cameras | |
| CN117734680B (en) | Blind area early warning method, system and storage medium for large vehicle | |
| CN110647863A (en) | Visual signal acquisition and analysis system for intelligent driving | |
| CN111753593A (en) | Real-time detection method, system and device for riding vehicle of vehicle-mounted all-round system | |
| CN117082332A (en) | Automatic configuration of camera settings using radar | |
| CN116424313A (en) | Vehicle and control method thereof | |
| JP7402753B2 (en) | Safety support system and in-vehicle camera image analysis method | |
| CN115123237A (en) | Intelligent lane changing auxiliary system easy to install | |
| CN115639562B (en) | Commercial vehicle blind spot detection system, braking method, commercial vehicle and storage medium | |
| CN115384403B (en) | Driving distraction early warning device and method for predicting front vehicle braking | |
| RU2809334C1 (en) | Unmanned vehicle and method for controlling its motion | |
| WO2024144436A1 (en) | Device and method for detecting objects | |
| US20230382408A1 (en) | Monitoring system and method for monitoring | |
| WO2024144437A1 (en) | Autonomous vehicle and method for controlling the movement thereof |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant |