CN107527511A - A kind of intelligent vehicle driving based reminding method and device - Google Patents

A kind of intelligent vehicle driving based reminding method and device Download PDF

Info

Publication number
CN107527511A
CN107527511A CN201610472493.8A CN201610472493A CN107527511A CN 107527511 A CN107527511 A CN 107527511A CN 201610472493 A CN201610472493 A CN 201610472493A CN 107527511 A CN107527511 A CN 107527511A
Authority
CN
China
Prior art keywords
traffic lights
time
region
image
state
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.)
Granted
Application number
CN201610472493.8A
Other languages
Chinese (zh)
Other versions
CN107527511B (en
Inventor
孙杰
屈凯
邝宏武
韩杰茜
朱江
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Hangzhou Hikvision Digital Technology Co Ltd
Original Assignee
Hangzhou Hikvision Digital Technology Co Ltd
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Hangzhou Hikvision Digital Technology Co Ltd filed Critical Hangzhou Hikvision Digital Technology Co Ltd
Priority to CN201610472493.8A priority Critical patent/CN107527511B/en
Publication of CN107527511A publication Critical patent/CN107527511A/en
Application granted granted Critical
Publication of CN107527511B publication Critical patent/CN107527511B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G08—SIGNALLING
    • G08G—TRAFFIC CONTROL SYSTEMS
    • G08G1/00—Traffic control systems for road vehicles
    • G08G1/09—Arrangements for giving variable traffic instructions
    • G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
    • G08G1/0967—Systems involving transmission of highway information, e.g. weather, speed limits

Landscapes

  • Life Sciences & Earth Sciences (AREA)
  • Atmospheric Sciences (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Traffic Control Systems (AREA)
  • Image Analysis (AREA)

Abstract

The embodiment of the present application discloses a kind of intelligent vehicle driving based reminding method and device, is related to technical field of information processing, applied to the train movement recording equipment installed in vehicle, wherein, the above method includes：Obtain image to be detected；Determine the target traffic lights region in described image to be detected；Obtain the state of traffic lights in the target traffic lights region；According to the state of traffic lights corresponding to the state and image to be detected of traffic lights corresponding to the first image, acquisition red light is in the very first time of current state and green light is in the second time of current state；Whether traffic lights conversion is occurred according to the very first time and second time detecting；In the case where testing result shows that traffic lights conversion occurs, driving prompting message is generated.The scheme provided using the embodiment of the present application is reminded, and improves the degree of accuracy of traffic lights transition detection, and then improve the degree of accuracy of prompting.

Description

A kind of intelligent vehicle driving based reminding method and device
Technical field
The application is related to technical field of information processing, more particularly to a kind of intelligent vehicle driving based reminding method and device.
Background technology
With the rapid growth of economy and the fast development of society, the owning amount of private car raises year by year, traffic safety and Traffic congestion receives much concern.In practical application, to reduce traffic accident, road surface running vehicle efficiency is improved, it will usually to traffic The conversion of signal lamp is detected, and can so notify driver to change driving states in time.
In the prior art, when whether detection traffic lights change, typically by analyzing the image of collection, Traffic lights region therein is determined, then obtains the color of traffic lights in above-mentioned zone, and is believed according to traffic The color of signal lamp judges whether to be changed.
Under normal circumstances, it is capable of detecting when using aforesaid way whether there occurs traffic lights conversion, however, actual road Situation is more complicated on road, for example, it may be possible to can have other vehicle car lights, street lamp, advertising lamp etc., these factors are possible to The color mistake of obtained traffic lights can be caused, and then causes to detect traffic lights convert failed, accuracy in detection Low, further, when generating driving prompting message based on aforesaid way, the driving generated reminds accuracy rate low.
The content of the invention
The embodiment of the present application discloses a kind of intelligent vehicle driving based reminding method and device, to improve accuracy in detection, enters And improve the degree of accuracy of generated driving prompting message.
To reach above-mentioned purpose, the embodiment of the present application discloses a kind of intelligent vehicle driving based reminding method, methods described bag Include：
Obtain image to be detected；
Determine the target traffic lights region in described image to be detected；
Obtain the state of traffic lights in the target traffic lights region；
According to traffic lights corresponding to the state of traffic lights corresponding to the first image and described image to be detected State, obtain red light and be in the very first time of current state and green light is in the second time of current state, wherein, described the One image is：The collection moment is located at before the collection moment of described image to be detected and the figure adjacent with described image to be detected Picture；
Whether traffic lights conversion is occurred according to the very first time and second time detecting；
In the case where testing result shows that traffic lights conversion occurs, driving prompting message is generated.
It is described whether to be sent out according to the very first time and second time detecting in a kind of implementation of the application Raw traffic lights conversion, including：
Judge whether the very first time and second time meet default time relationship；
If it is, obtain the first position of red light and the second place of green light in the target traffic lights region；
Judge whether the first position and the second place meet default position relationship；
If satisfied, judge traffic lights conversion occurs.
In a kind of implementation of the application, it is in the current state of the red light：OFF state, the green light are worked as Preceding state is：It is described when judging whether the very first time and second time meet default in the case of lighting state Between relation, including：
Judge whether the very first time and second time meet following formula：
tred_dark-tgrn_light<T
Wherein, tred_darkRepresent the very first time, tgrn_lightSecond time is represented, T represents default time threshold Value.
In a kind of implementation of the application, it is described judge the first position and the second place whether meet it is pre- If position relationship, including：
Judge the first position and the second place whether meet it is at least one in relationship below：
|xred-xgrn|<D1And | yred-ygrn|<D2
|xred-xgrn|<D2And | yred-ygrn|<D1
Wherein, xred、yredRepresent abscissa, the ordinate of the first position central point, xgrn、ygrnRepresent described second Abscissa, the ordinate of place-centric point, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
In a kind of implementation of the application, the target traffic lights area determined in described image to be detected Domain, including：
Obtain the target suspect region of traffic lights in described image to be detected；
Target traffic lights region is determined from the target suspect region by default machine learning algorithm.
In a kind of implementation of the application, the target suspicion for obtaining traffic lights in described image to be detected Region, including：
Obtain the connected region in described image to be detected；
According to the geometric properties of the connected region, the target suspicion area of traffic lights in described image to be detected is obtained Domain.
In a kind of implementation of the application, it is described by default machine learning algorithm from the target suspect region Middle determination target traffic lights region, including：
The state of Current vehicle is obtained, and obtains the relative status between the target suspect region and Current vehicle；
According to the state of Current vehicle and the relative status, filtration treatment is carried out to the target suspect region；
Determine that target traffic is believed from the target suspect region after filtration treatment by default machine learning algorithm Signal lamp region.
In a kind of implementation of the application, it is described by default machine learning algorithm from the target suspect region Middle determination target traffic lights region, including：
Calculate the target suspect region respectively with suspicion in institute's detection image in the first preset period of time before current time Associated confidence between region；
By default machine learning algorithm, the target suspect region from associated confidence more than the first predetermined threshold value Middle determination target traffic lights region.
In a kind of implementation of the application, the traffic lights change detection method also includes：
Before current time traffic lights region and the target are determined in the second preset period of time in institute's detection image Associated confidence is more than the image of the second predetermined threshold value between traffic lights region；
According to the testing result for traffic lights conversion for determining image, to described image to be detected for handing over The testing result of ventilating signal lamp conversion is corrected.
To reach above-mentioned purpose, the embodiment of the present application discloses a kind of intelligent vehicle traveling reminder, described device bag Include：
Image obtains module, for obtaining image to be detected；
Area determination module, for determining the target traffic lights region in described image to be detected；
State obtains module, for obtaining the state of traffic lights in the target traffic lights region；
Time obtains module, for the state of traffic lights according to corresponding to the first image and the mapping to be checked The state of traffic lights as corresponding to, obtains that red light is in the very first time of current state and green light is in the of current state Two times, wherein, described first image is：Collection the moment be located at the collection moment of described image to be detected before and with it is described The adjacent image of image to be detected；
Transition detection module, for whether traffic lights occurring according to the very first time and second time detecting Conversion；
Generation module is reminded, in the case of showing that traffic lights conversion occurs in testing result, generation driving carries Awake information.
In a kind of implementation of the application, the transition detection module, including：
Time judging submodule, for judging whether the very first time and second time meet that the default time closes System；
Position obtains submodule, in the case of being in the judged result of the time judging submodule, obtains institute State the first position of red light and the second place of green light in target traffic lights region；
Position judgment submodule, for judging whether the first position and the second place meet that default position is closed System；
Decision sub-module is changed, in the case of being in the result of determination of the position judgment submodule, judges hair Raw traffic lights conversion.
In a kind of implementation of the application, it is in the current state of the red light：OFF state, the green light are worked as Preceding state is：In the case of lighting state, the time judging submodule, specifically for judging the very first time and described Whether two times met following formula：
tred_dark-tgrn_light<T
Wherein, tred_darkRepresent the very first time, tgrn_lightSecond time is represented, T represents default time threshold Value.
In a kind of implementation of the application, the position judgment submodule, specifically for judging the first position Whether meet with the second place at least one in relationship below：
|xred-xgrn|<D1And | yred-ygrn|<D2
|xred-xgrn|<D2And | yred-ygrn|<D1
Wherein, xred、yredRepresent abscissa, the ordinate of the first position central point, xgrn、ygrnRepresent described second Abscissa, the ordinate of place-centric point, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
In a kind of implementation of the application, the area determination module, including：
Suspect region obtains submodule, for obtaining the target suspect region of traffic lights in described image to be detected；
Region determination sub-module, for determining target from the target suspect region by default machine learning algorithm Traffic lights region.
In a kind of implementation of the application, the suspect region obtains submodule, including：
Connected region obtaining unit, for obtaining the connected region in described image to be detected；
Suspect region obtaining unit, for the geometric properties according to the connected region, obtain in described image to be detected The target suspect region of traffic lights.
In a kind of implementation of the application, the region determination sub-module, including：
First state obtaining unit, for obtaining the state of Current vehicle；
Second state obtaining unit, for obtaining the relative status between the target suspect region and Current vehicle；
Area filter unit, for the state according to Current vehicle and the relative status, to the target suspicion area Domain carries out filtration treatment；
First area determining unit, for by default machine learning algorithm from the target suspicion after filtration treatment Target traffic lights region is determined in region.
In a kind of implementation of the application, the region determination sub-module, including：
Confidence calculations subelement, when being preset respectively with before current time first for calculating the target suspect region Associated confidence in section in institute's detection image between suspect region；
Second area determining unit, for by default machine learning algorithm, it is default to be more than first from associated confidence Target traffic lights region is determined in the target suspect region of threshold value.
In a kind of implementation of the application, the traffic lights transition detecting device also includes：
Image determining module, in the second preset period of time in institute's detection image, determining that traffic is believed before current time Associated confidence is more than the image of the second predetermined threshold value between signal lamp region and the target traffic lights region；
As a result correction verification module, the testing result for traffic lights conversion of image is determined for foundation, to described The testing result for traffic lights conversion of image to be detected is corrected.
As seen from the above, in the embodiment of the present application, train movement recording equipment obtains image to be detected, and determines image to be detected In target traffic lights region after, obtain the states of traffic signals, and the traffic lights according to corresponding to the first image The state of traffic lights corresponding to state, image to be detected, obtain red light and be at the very first time and the green light of current state In the second time of current state, then according to the above-mentioned very first time and the second time detecting, whether there occurs traffic lights turn Change, in the case where testing result shows that traffic lights conversion occurs, generate driving prompting message.As seen from the above, this Shen Please not only only account for the state of traffic lights, that is, the colouring information of light in embodiment, it is also contemplated that traffic letter Signal lamp is in the time of current state, and influence of the ambient environmental factors to testing result is reduced from time dimension, therefore, is carried High accuracy in detection, and then improve the degree of accuracy of generated driving prompting message.
Brief description of the drawings
, below will be to embodiment or existing in order to illustrate more clearly of the embodiment of the present application or technical scheme of the prior art There is the required accompanying drawing used in technology description to be briefly described, it should be apparent that, drawings in the following description are only this Some embodiments of application, for those of ordinary skill in the art, on the premise of not paying creative work, can be with Other accompanying drawings are obtained according to these accompanying drawings.
Fig. 1 is the schematic flow sheet for the first intelligent vehicle driving based reminding method that the embodiment of the present application provides；
Fig. 2 is the schematic flow sheet for second of intelligent vehicle driving based reminding method that the embodiment of the present application provides；
Fig. 3 is the structural representation for the first intelligent vehicle traveling reminder that the embodiment of the present application provides；
Fig. 4 is the structural representation for second of intelligent vehicle traveling reminder that the embodiment of the present application provides.
Embodiment
Below in conjunction with the accompanying drawing in the embodiment of the present application, the technical scheme in the embodiment of the present application is carried out clear, complete Site preparation describes, it is clear that described embodiment is only some embodiments of the present application, rather than whole embodiments.It is based on Embodiment in the application, those of ordinary skill in the art are obtained every other under the premise of creative work is not made Embodiment, belong to the scope of the application protection.
The schematic flow sheet for the first intelligent vehicle driving based reminding method that Fig. 1 provides for the embodiment of the present application, this method It can apply to the train movement recording equipment installed on vehicle.
Specifically, above-mentioned train movement recording equipment can be understood as：Drive recorder.
The above method includes：
S101：Obtain image to be detected.
In this step, above-mentioned image to be detected can be understood as image capture device gathers in real time, vehicle front scene Corresponding image.Above-mentioned image capture device is located on vehicle, for example, can be located at the front windshield center upper portion position of vehicle Put, may be located on the other positions of vehicle certainly, the application is defined not to this.Specifically, above-mentioned IMAQ is set Standby can be the image capture device that the drive recorder installed on vehicle carries, and can also be separately installed in the figure on vehicle As collecting device.
Certainly, above-mentioned image to be detected can also have ready conditions collection, for example, train movement recording equipment is according to automobile navigation Information, determine that current vehicle position is provided with the distance between crossing of traffic lights with front, when the distance is less than necessarily During numerical value, start image capture device and carry out IMAQ, and then obtain above-mentioned image to be detected.
In addition, in the case where above-mentioned image capture device is the image capture device that drive recorder carries, the application The executive agent of embodiment can be understood as：The MCU (Microcontroller Unit, micro-control unit) of drive recorder.
S102：Determine the target traffic lights region in image to be detected.
Because traffic lights have specific feature, it may for example comprise：Red light region, amber light region, green light region, and Three regions transversely or longitudinally arrange etc. in a fixed order, so, determine the target traffic signals in image to be detected During lamp region, the feature and image recognition algorithm of traffic lights can be combined, determines the target traffic in image to be detected Signal lamp region.
Specifically, it is determined that during target traffic lights region in image to be detected, it is also contemplated that detected The size in region, if although detected region meets the feature of traffic lights, but region is too small, it is believed that car Current distance is provided with the crossing of traffic lights farther out, and when vehicle is travelled to the crossing, the state of traffic lights can Change can be had occurred and that, so, driving prompting message can be generated in this case, therefore, it is possible to above-mentioned zone not made For the target traffic lights region of image to be detected.
It should be understood that complicated, diversified information, especially driving at night often be present in driving scene, field of driving a vehicle Light information is more abundant in scape, for example, ornament lamp information, billboard on the car light information of surrounding vehicles, Adjacent Buildings Light information on plaque etc., may exist from the point of view of certain shooting angle, in these information similar with traffic lights Information.In consideration of it, in a kind of implementation of the application, when determining the target traffic lights region in image to be detected, The target suspect region of traffic lights in image to be detected can be first obtained, then by default machine learning algorithm from mesh Mark and target traffic lights region is determined in suspect region.
It should be noted that the target suspect region in the image to be detected obtained can be a region, can also It is multiple regions, the application is defined not to this.
In addition, a variety of machine learning algorithms in the prior art be present, those skilled in the art can be according to having known Know and obtain, I will not elaborate.
In a kind of implementation of the application, when obtaining the target suspect region of traffic lights in image to be detected, The connected region in image to be detected can be first obtained, then further according to the geometric properties of above-mentioned connected region, is obtained to be detected The target suspect region of traffic lights in image.
Optionally, the target suspect region for obtaining traffic lights in image to be detected can be by way of BLOB clusters Obtain, specifically, carrying out two-value to image to be detected according to the colouring information according to each pixel in image to be detected first Change is handled, and then each pixel in image to be detected is marked according to binary conversion treatment result, obtains connected region.
The geometric properties of connected region can include：The length-width ratio in region, the area in region etc..Due to traffic lights Limitednumber, and its shape tool has certain limitations, so its length-width ratio generally corresponds to several special values, and IMAQ is set Object is smaller in standby acquired image, that is, the area of object in the picture is smaller, illustrate object and vehicle it Between distance it is more remote, and when the distance between traffic lights and vehicle farther out when, the switchings of traffic lights is typically to driving Influence less, based on the above situation, the geometric properties such as region length-width ratio, the region area of connected region can be combined to binaryzation The connected region obtained after processing is filtered, and is disliked so as to more be accurately obtained the target of traffic lights in image to be detected Doubt region.
For example, the region length-width ratio of the connected region obtained after binary conversion treatment is 10：1, according to traffic lights Design rule, the length-width ratio obviously do not meet the design requirement of traffic lights, can so it is determined that during target suspect region So that the connected region to be foreclosed.
For another aspect, image capture device be it is real-time carry out IMAQ, according to the tandem of acquisition time, There is stronger correlation, the scene corresponding to each image is also similar between each image.Based on this, the application's In another implementation, target traffic lights region is determined from target suspect region by default machine learning algorithm When, can first calculate target suspect region respectively with suspect region in institute's detection image in the first preset period of time before current time Between associated confidence, then by default machine learning algorithm, the mesh of the first predetermined threshold value is more than from associated confidence Mark and target traffic lights region is determined in suspect region.It is default with before current time first respectively to calculate target suspect region Associated confidence in period in institute's detection image between suspect region, the association results of each frame are obtained, and then obtain signal Lamp movement locus.Wherein, when above-mentioned associated confidence is more than the first predetermined threshold value, it is believed that be successfully associated.
Can be understood as vehicle has two states：Inactive state and motion state, and motion state can be divided into Microinching state and quick motion state etc.；Traffic lights only have a kind of state：Inactive state.No matter vehicle is in Which kind of state, state and the traffic lights of vehicle are mutually matched relative to the state of vehicle.
For example, vehicle remains static, because traffic lights are static, then traffic lights are relative to vehicle State is inactive state, and traffic lights and the distance between vehicle are constant；
Vehicle is kept in motion, due to traffic lights be it is static, then traffic lights relative to vehicle state For motion state, and moved towards direction of traffic, traffic lights and the distance between vehicle are more and more nearer.
In a kind of implementation of the application, mesh is determined from target suspect region by default machine learning algorithm When marking traffic lights region, the state of Current vehicle can be first obtained, and obtain between target suspect region and Current vehicle Relative status, according to the state of Current vehicle and above-mentioned relative status, filtration treatment is carried out to target suspect region, then Target traffic lights region is determined from the target suspect region after filtration treatment by default machine learning algorithm.
A kind of situation：It is assumed that the state of the Current vehicle obtained is inactive state, the target suspect region obtained with Relative status between Current vehicle are the state moved to direction of traffic, then it is considered that the state of Current vehicle is disliked with target Doubt region to mismatch relative to the relative status between Current vehicle, the target suspect region can be filtered out.
Another situation：It is assumed that the state of the Current vehicle obtained is motion state, the target suspect region obtained Relative status between Current vehicle are the state moved to direction of traffic, then it is considered that the state and target of Current vehicle Suspect region matches relative to the relative status between Current vehicle, can not filter out the target suspect region.
Another situation：It is assumed that the state of the Current vehicle obtained is inactive state, the target suspect region obtained Relative status between Current vehicle are inactive state, then it is considered that the state of Current vehicle is relative with target suspect region Relative status matching between Current vehicle, can not filter out the target suspect region.
It should be noted that the application simply illustrates by taking above-mentioned several situations as an example, limit in practical application and not only In above-mentioned several situations.
S103：Obtain the state of traffic lights in target traffic lights region.
The color that traffic lights include lamp is different, the states of traffic lights also all difference.
For example, traffic lights only include red light and green light, then the state of traffic lights can have：Red light lights shape State, red light OFF state, green light light state, green light OFF state；
Traffic lights include red light, amber light, green light, then the state of traffic lights can have：Red light lights state, red Lamp OFF state, amber light light state, amber light OFF state, green light and light state, green light OFF state.
Specifically, when obtaining the state of traffic lights in target traffic lights region, can be handed over by analyzing target The color value of pixel obtains in ventilating signal lamp region.
S104：According to traffic signals corresponding to the state and image to be detected of traffic lights corresponding to the first image The state of lamp, acquisition red light is in the very first time of current state and green light is in the second time of current state.
Wherein, the first image is：Collection the moment be located at the collection moment of described image to be detected before and with it is described to be checked The adjacent image of altimetric image.
Image capture device acquired image can carry timestamp information, and the image before to present image enters During row detection, the states of traffic lights in each image is had been obtained for, therefore, can be according to having detected traffic in image The timestamp of the state of signal lamp, timestamp and present image, obtain red light and be at time and the green light of current state In the time of current state.
Optionally, the image before above-mentioned present image can only select that what is gathered in certain time length before current time Image, for example, 60 seconds etc., because traffic lights are continually changing, select the image in certain time length to be used as reference to obtain Red light is in the time of current state and green light is in the time of current state, be favorably improved the effective of obtained time Property.
In addition, when being detected to each image, if detecting, traffic lights are changed, and can be directed to and be handed over The timestamp of state recording present image after ventilating signal lamp transformation, in order to which the later stage carries out time calculating.
For example, red light is converted to the state of lighting from the OFF state of former frame, then can be using present image as when red The first frame risen, the timestamp of present image, that is, the time that red light lights are recorded, be in when the later stage also needs to acquisition red light When lighting the time of state, it can be obtained according to the difference of the timestamp and timestamp recorded above of detection moment image, The later stage can be so facilitated to calculate the time that red light is in the state that lights.
Similarly, the time of red light extinguishing, the time that green light lights, the time of green light extinguishing can also be recorded.
S105：Whether traffic lights conversion is occurred according to the very first time and the second time detecting.
It should be noted that traffic lights conversion can include：Red light extinguishes the conversion that green light lights, and can also wrap Include：Red light lights the conversion of green light extinguishing, and the application is not defined to the particular type of traffic lights conversion.
For only including the traffic lights of red light and green light, red light lights then green light extinguishing, and red light extinguishes then Green light lights, and both will not light simultaneously under normal circumstances, will not also extinguish simultaneously, and therefore, ideally, red light is in bright The time that the time for the state of rising and green light are in OFF state is equal, and in other words, red light is in time and the green light of OFF state Time in the state that lights is equal.In summary, that is to say, that ideally red light is in time of current state and green The time that lamp is in current state is equal.
And for the traffic lights comprising red light, amber light and green light, have Huang between red light and the conversion of green light The lighting of lamp, the extinguishing of amber light, so red light is in the time of current state and green light is in time of current state and has one Fixed interval, that is to say, that ideally red light is in the time of current state and green light is in time of current state not phase Deng.
In addition, the factor such as degree of accuracy and institute's collection image quality by image detection algorithm is influenceed, red light is detected The time that bright time or green light in current state are in current state may have error, be able to may also so bring The interval that red light is in the time of current state and green light was between the time of current state.
In view of the foregoing, the time of current state can be in by red light and green light is in the time of current state Between numerical relation judge whether traffic lights are changed.
Furthermore in addition to considering above-mentioned interval, it is also contemplated that the size of above-mentioned time, put out for example, red light is in The time for state of going out is 1 second, and the time that green light is in the state that lights also is 1 second, then it is considered that red light just extinguishes, green light is firm Just light, therefore, it is possible to determine that extinguish the conversion lighted to green light there occurs red light.
From the foregoing description it is known that before image to be detected and current time in certain time length collection image it Between there is correlation, so these images for traffic lights conversion testing result be also that there is correlation, be based on This, in a kind of implementation of the application, above-mentioned traffic lights detection method can also include：Before current time In two preset period of time in institute's detection image, associated confidence between traffic lights region and target traffic lights region is determined More than the image of the second predetermined threshold value, according to the testing result for traffic lights conversion for determining image, treated to described The testing result for traffic lights conversion of detection image is corrected.Erroneous judgement situation can be so eliminated, is greatly improved The degree of accuracy of traffic lights detection.
It should be noted that the value of above-mentioned second preset period of time and the first preset period of time can be with equal, can not also phase Deng above-mentioned second predetermined threshold value and the first predetermined threshold value can be with equal, can also be unequal, and the application is limited not to this It is fixed.
S106：In the case where testing result shows that traffic lights conversion occurs, driving prompting message is generated.
Above-mentioned driving prompting message can be sound prompting information, for example, voice reminder information or buzzing prompting message Etc..
For the traffic of reality, if Current traffic signal lamp extinguishes the conversion lighted to green light there occurs red light, Vehicle can be with normal pass, to prevent human pilot from, without traffic lights conversion is found in time, causing traffic congestion, can be with Voice reminder information is generated, prompts driver；
There occurs the conversion that when red plays green light extinguishing, if vehicle need to be closed to traffic Current traffic signal lamp, be anti- Only human pilot causes to knock into the back, collides the traffic accidents such as pedestrian, can generate voice without traffic lights conversion is found in time Prompting message, prompt driver.
The prompting message it can be seen that generation is accurately driven a vehicle, help to reduce traffic accident, improve road surface running vehicle efficiency.
As seen from the above, in above-mentioned each embodiment, train movement recording equipment obtains image to be detected, and determines mapping to be checked Behind target traffic lights region as in, the state of traffic signals, and the traffic lights according to corresponding to the first image are obtained State, the state of traffic lights corresponding to image to be detected, obtain the very first time and green light that red light is in current state The second time in current state, then according to the above-mentioned very first time and the second time detecting, whether there occurs traffic lights Conversion, in the case where testing result shows that traffic lights conversion occurs, generate driving prompting message.As seen from the above, on State the state that traffic lights are not only only accounted in each embodiment, that is, the colouring information of light, it is also contemplated that hand over Ventilating signal lamp is in the time of current state, and influence of the ambient environmental factors to testing result is reduced from time dimension, because This, improves accuracy in detection, and then improves the degree of accuracy of generated driving prompting message.
In a kind of implementation of the application, referring to Fig. 2, there is provided the stream of second of intelligent vehicle driving based reminding method Journey schematic diagram, compared with previous embodiment, in the present embodiment, whether traffic letter is occurred according to the very first time and the second time detecting Signal lamp changes (S105), including：
S105A：Judge whether the very first time and the second time meet default time relationship, if it is, performing S105B；
Specifically, the current state in red light is：OFF state, the current state of green light are：In the case of lighting state, It can judge whether the very first time and the second time meet default time relationship in the following manner：
Judge whether the very first time and the second time meet following formula：
tred_dark-tgrn_light<T
If meet above-mentioned expression formula, it can be understood as the very first time and the second time meet default time relationship.
In above-mentioned expression formula, tred_darkRepresent that the very first time, that is, red light are in the time of OFF state, tgrn_light Represent that the second time, that is, green light are in the time for the state that lights, T represents default time threshold, and T value can basis Concrete condition is set.
It is similar with the above situation, it is the state that lights in the current state of red light, the current state of green light is OFF state Situation, it can also be judged according to the expression formula similar with above-mentioned expression formula.
Specifically, judging whether the very first time and the second time meet following formula：
Tgrn_dark-tred_light<T
Wherein, tred_lightRepresent that the very first time, that is, red light are in the time for the state that lights, Tgrn_darkRepresent second Time, that is, green light are in the time of OFF state.
S105B：Obtain the first position of red light and the second place of green light in target traffic lights region.
Above-mentioned first position and said second position can have the region of regular shape, such as：Rectangular area, circle Region etc..
S105C：Judge whether first position and the second place meet default position relationship, if satisfied, judging to hand over Ventilating signal lamp is changed, and performs S106.
Specifically, it can judge whether first position and the second place meet default position relationship in the following manner：
Judge first position and the second place whether meet it is at least one in relationship below：
|xred-xgrn|<D1And | yred-ygrn|<D2
|xred-xgrn|<D2And | yred-ygrn|<D1
If meeting, above-mentioned expression formula is at least one, it can be understood as first position and the second place meet that default position is closed System.
Wherein, xred、yredAbscissa, the ordinate of first position central point are represented, that is, in red light region Heart point coordinates, xgrn、ygrnRepresent abscissa, the ordinate of second place central point, that is, the central point of green light region Coordinate, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
Optionally, above-mentioned D1D can be less than2。
Current state for red light is：OFF state, the current state of green light are：The situation of state is lighted, if above-mentioned The very first time, the second time meet default time relationship, and above-mentioned first position, the second place meet that default position is closed System, then it can be determined that the conversion extinguished there occurs red light and lighted to green light；
Current state for red light is：State is lighted, the current state of green light is：The situation of OFF state, if above-mentioned The very first time, the second time meet default time relationship, and above-mentioned first position, the second place meet that default position is closed System, then it can be determined that there occurs the conversion that when red plays green light extinguishing.
As seen from the above, in the present embodiment, when whether detection traffic lights change, the traffic not only considered is believed The state of signal lamp, traffic lights are in the time of current state, also further contemplate the position of traffic lights, due to by According to the design rule of traffic lights, its relative position of the lamp of a variety of colors is fixed in traffic lights, so, using this The scheme that embodiment provides, can reduce influence of the ambient environmental factors to testing result, therefore, it is possible to enter from Spatial Dimension One step improves accuracy in detection.
Corresponding with above-mentioned intelligent vehicle driving based reminding method, the embodiment of the present application additionally provides a kind of intelligent vehicle driving Alarm set.
Fig. 3 is a kind of structural representation for intelligent vehicle traveling reminder that the embodiment of the present application provides, applied to car The train movement recording equipment installed in, above-mentioned device of stating include：
Image obtains module 301, for obtaining image to be detected；
Area determination module 302, for determining the target traffic lights region in image to be detected；
State obtains module 303, for obtaining the state of traffic lights in the target traffic lights region；
Time obtains module 304, for the state of traffic lights according to corresponding to the first image and described to be detected The state of traffic lights corresponding to image, acquisition red light is in the very first time of current state and green light is in current state Second time, wherein, described first image is：Collection the moment be located at before the collection moment of described image to be detected and with institute State the adjacent image of image to be detected；
Transition detection module 305, for traffic letter whether to occur according to the very first time and second time detecting Signal lamp is changed；
Generation module 306 is reminded, in the case of showing that traffic lights conversion occurs in testing result, generates driving Prompting message.
Optionally, the area determination module 302 can include：
Suspect region obtains submodule, for obtaining the target suspect region of traffic lights in image to be detected；
Region determination sub-module, for determining target from the target suspect region by default machine learning algorithm Traffic lights region.
Optionally, the suspect region obtains submodule and can included：
Connected region obtaining unit, for obtaining the connected region in image to be detected；
Suspect region obtaining unit, for the geometric properties according to the connected region, obtain in described image to be detected The target suspect region of traffic lights.
Optionally, the region determination sub-module can include：
First state obtaining unit, for obtaining the state of Current vehicle；
Second state obtaining unit, for obtaining the relative status between the target suspect region and Current vehicle；
Area filter unit, for the state according to Current vehicle and the relative status, to the target suspicion area Domain carries out filtration treatment；
First area determining unit, for by default machine learning algorithm from the target suspicion after filtration treatment Traffic lights region is determined in region.
Optionally, the region determination sub-module, including：
Confidence calculations subelement, when being preset respectively with before current time first for calculating the target suspect region Associated confidence in section in institute's detection image between suspect region；
Second area determining unit, for by default machine learning algorithm, it is default to be more than first from associated confidence Target traffic lights region is determined in the target suspect region of threshold value.
Optionally, the traffic lights transition detecting device can also include：
Image determining module, in the second preset period of time in institute's detection image, determining that traffic is believed before current time Associated confidence is more than the image of the second predetermined threshold value between signal lamp region and the target traffic lights region；
As a result correction verification module, the testing result for traffic lights conversion of image is determined for foundation, to described The testing result for traffic lights conversion of image to be detected is corrected.
As seen from the above, in above-mentioned each embodiment, train movement recording equipment obtains image to be detected, and determines mapping to be checked Behind target traffic lights region as in, the state of traffic signals, and the traffic lights according to corresponding to the first image are obtained State, the state of traffic lights corresponding to image to be detected, obtain the very first time and green light that red light is in current state The second time in current state, then according to the above-mentioned very first time and the second time detecting, whether there occurs traffic lights Conversion, in the case where testing result shows that traffic lights conversion occurs, generate driving prompting message.As seen from the above, on State the state that traffic lights are not only only accounted in each embodiment, that is, the colouring information of light, it is also contemplated that hand over Ventilating signal lamp is in the time of current state, and influence of the ambient environmental factors to testing result is reduced from time dimension, because This, improves accuracy in detection, and then improves the degree of accuracy of generated driving prompting message.
In a kind of implementation of the application, referring to Fig. 4, there is provided the knot of second of intelligent vehicle traveling reminder Structure schematic diagram, wherein, the transition detection module 305 includes：
Time judging submodule 305A, during for judging whether the very first time and second time meet default Between relation；
Position obtains submodule 305B, for the situation for being yes in the judged result of the time judging submodule 404A Under, obtain the first position of red light and the second place of green light in the target traffic lights region；
Position judgment submodule 305C, for judging whether the first position and the second place meet default position Put relation；
Decision sub-module 305D is changed, for the situation for being yes in the result of determination of the position judgment submodule 404C Under, judge traffic lights conversion occurs.
Optionally, it is in the current state of the red light：OFF state, the current state of the green light are：Light state In the case of, the time judging submodule, specifically for judge the very first time and second time whether meet with Lower expression formula：
tred_dark-tgrn_light<T
Wherein, tred_darkRepresent the very first time, tgrn_lightSecond time is represented, T represents default time threshold Value.
Optionally, the position judgment submodule, specifically for judging whether are the first position and the second place Meet at least one in relationship below：
|xred-xgrn|<D1And | yred-ygrn|<D2
|xred-xgrn|<D2And | yred-ygrn|<D1
Wherein, xred、yredRepresent abscissa, the ordinate of the first position central point, xgrn、ygrnRepresent described second Abscissa, the ordinate of place-centric point, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
As seen from the above, in the present embodiment, when whether detection traffic lights change, the traffic not only considered is believed The state of signal lamp, traffic lights are in the time of current state, also further contemplate the position of traffic lights, due to by According to the design rule of traffic lights, its relative position of the lamp of a variety of colors is fixed in traffic lights, so, using this The scheme that embodiment provides, can reduce influence of the ambient environmental factors to testing result, therefore, it is possible to enter from Spatial Dimension One step improves accuracy in detection.
For device embodiment, because it is substantially similar to embodiment of the method, so description is fairly simple, it is related Part illustrates referring to the part of embodiment of the method.
It should be noted that herein, such as first and second or the like relational terms are used merely to a reality Body or operation make a distinction with another entity or operation, and not necessarily require or imply and deposited between these entities or operation In any this actual relation or order.Moreover, term " comprising ", "comprising" or its any other variant are intended to Nonexcludability includes, so that process, method, article or equipment including a series of elements not only will including those Element, but also the other element including being not expressly set out, or it is this process, method, article or equipment also to include Intrinsic key element.In the absence of more restrictions, the key element limited by sentence "including a ...", it is not excluded that Other identical element also be present in process, method, article or equipment including the key element.
Can one of ordinary skill in the art will appreciate that realizing that all or part of step in above method embodiment is To instruct the hardware of correlation to complete by program, described program can be stored in computer read/write memory medium, The storage medium designated herein obtained, such as：ROM/RAM, magnetic disc, CD etc..
The preferred embodiment of the application is the foregoing is only, is not intended to limit the protection domain of the application.It is all Any modification, equivalent substitution and improvements made within spirit herein and principle etc., it is all contained in the protection domain of the application It is interior.

Claims (18)

  1. The based reminding method 1. a kind of intelligent vehicle is driven a vehicle, it is characterised in that methods described includes：
    Obtain image to be detected；
    Determine the target traffic lights region in described image to be detected；
    Obtain the state of traffic lights in the target traffic lights region；
    According to the shape of traffic lights corresponding to the state of traffic lights corresponding to the first image and described image to be detected State, acquisition red light is in the very first time of current state and green light is in the second time of current state, wherein, first figure As being：The collection moment is located at before the collection moment of described image to be detected and the image adjacent with described image to be detected；
    Whether traffic lights conversion is occurred according to the very first time and second time detecting；
    In the case where testing result shows that traffic lights conversion occurs, driving prompting message is generated.
  2. 2. according to the method for claim 1, it is characterised in that described to be examined according to the very first time and second time Whether survey occurs traffic lights conversion, including：
    Judge whether the very first time and second time meet default time relationship；
    If it is, obtain the first position of red light and the second place of green light in the target traffic lights region；
    Judge whether the first position and the second place meet default position relationship；
    If satisfied, judge traffic lights conversion occurs.
  3. 3. according to the method for claim 2, it is characterised in that be in the current state of the red light：OFF state, it is described The current state of green light is：It is described to judge whether the very first time and second time meet in the case of lighting state Default time relationship, including：
    Judge whether the very first time and second time meet following formula：
    tred_dark-tgrn_light<T
    Wherein, tred_darkRepresent the very first time, tgrn_lightSecond time is represented, T represents default time threshold.
  4. 4. according to the method for claim 2, it is characterised in that the judgement first position and the second place are It is no to meet default position relationship, including：
    Judge the first position and the second place whether meet it is at least one in relationship below：
    |xred-xgrn|<D1And | yred-ygrn|<D2
    |xred-xgrn|<D2And | yred-ygrn|<D1
    Wherein, xred、yredRepresent abscissa, the ordinate of the first position central point, xgrn、ygrnRepresent the second place Abscissa, the ordinate of central point, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
  5. 5. according to the method for claim 1, it is characterised in that the target traffic letter determined in described image to be detected Signal lamp region, including：
    Obtain the target suspect region of traffic lights in described image to be detected；
    Target traffic lights region is determined from the target suspect region by default machine learning algorithm.
  6. 6. according to the method for claim 5, it is characterised in that described to obtain traffic lights in described image to be detected Target suspect region, including：
    Obtain the connected region in described image to be detected；
    According to the geometric properties of the connected region, the target suspect region of traffic lights in described image to be detected is obtained.
  7. 7. the method according to claim 5 or 6, it is characterised in that it is described by default machine learning algorithm from described Target traffic lights region is determined in target suspect region, including：
    The state of Current vehicle is obtained, and obtains the relative status between the target suspect region and Current vehicle；
    According to the state of Current vehicle and the relative status, filtration treatment is carried out to the target suspect region；
    Target traffic lights are determined from the target suspect region after filtration treatment by default machine learning algorithm Region.
  8. 8. the method according to claim 5 or 6, it is characterised in that it is described by default machine learning algorithm from described Target traffic lights region is determined in target suspect region, including：
    Calculate the target suspect region respectively with suspect region in institute's detection image in the first preset period of time before current time Between associated confidence；
    By default machine learning algorithm, it is more than from associated confidence in the target suspect region of the first predetermined threshold value really Set the goal traffic lights region.
  9. 9. according to the method for claim 1, it is characterised in that methods described also includes：
    Before current time in the second preset period of time in institute's detection image, traffic lights region and the target traffic are determined Associated confidence is more than the image of the second predetermined threshold value between signal lamp region；
    According to the testing result for traffic lights conversion for determining image, described image to be detected is believed for traffic The testing result of signal lamp conversion is corrected.
  10. 10. a kind of intelligent vehicle traveling reminder, it is characterised in that described device includes：
    Image obtains module, for obtaining image to be detected；
    Area determination module, for determining the target traffic lights region in described image to be detected；
    State obtains module, for obtaining the state of traffic lights in the target traffic lights region；
    Time obtains module, for the state of traffic lights according to corresponding to the first image and described image to be detected pair The state for the traffic lights answered, obtain red light and be in the very first time of current state and when green light is in the second of current state Between, wherein, described first image is：Collection the moment be located at the collection moment of described image to be detected before and with it is described to be checked The adjacent image of altimetric image；
    Transition detection module, for turning according to whether the very first time and second time detecting occur traffic lights Change；
    Generation module is reminded, in the case of showing that traffic lights conversion occurs in testing result, letter is reminded in generation driving Breath.
  11. 11. device according to claim 10, it is characterised in that the transition detection module, including：
    Time judging submodule, for judging whether the very first time and second time meet default time relationship；
    Position obtains submodule, in the case of being in the judged result of the time judging submodule, obtains the mesh Mark the first position of red light and the second place of green light in traffic lights region；
    Position judgment submodule, for judging whether the first position and the second place meet default position relationship；
    Decision sub-module is changed, in the case of being in the result of determination of the position judgment submodule, judgement is handed over Ventilating signal lamp is changed.
  12. 12. device according to claim 11, it is characterised in that be in the current state of the red light：OFF state, institute The current state for stating green light is：In the case of lighting state, the time judging submodule, during specifically for judging described first Between and second time whether meet following formula：
    tred_dark-tgrn_light<T
    Wherein, tred_darkRepresent the very first time, tgrn_lightSecond time is represented, T represents default time threshold.
  13. 13. device according to claim 11, it is characterised in that the position judgment submodule, specifically for judging State first position and the second place whether meet it is at least one in relationship below：
    |xred-xgrn|<D1And | yred-ygrn|<D2
    |xred-xgrn|<D2And | yred-ygrn|<D1
    Wherein, xred、yredRepresent abscissa, the ordinate of the first position central point, xgrn、ygrnRepresent the second place Abscissa, the ordinate of central point, D1Represent the first predeterminated position threshold value, D2Represent the second predeterminated position threshold value.
  14. 14. device according to claim 10, it is characterised in that the area determination module, including：
    Suspect region obtains submodule, for obtaining the target suspect region of traffic lights in described image to be detected；
    Region determination sub-module, for determining target traffic from the target suspect region by default machine learning algorithm Signal lamp region.
  15. 15. device according to claim 14, it is characterised in that the suspect region obtains submodule, including：
    Connected region obtaining unit, for obtaining the connected region in described image to be detected；
    Suspect region obtaining unit, for the geometric properties according to the connected region, obtain traffic in described image to be detected The target suspect region of signal lamp.
  16. 16. the device according to claims 14 or 15, it is characterised in that the region determination sub-module, including：
    First state obtaining unit, for obtaining the state of Current vehicle；
    Second state obtaining unit, for obtaining the relative status between the target suspect region and Current vehicle；
    Area filter unit, for the state according to Current vehicle and the relative status, the target suspect region is entered Row filtration treatment；
    First area determining unit, for by default machine learning algorithm from the target suspect region after filtration treatment Middle determination target traffic lights region.
  17. 17. the device according to claims 14 or 15, it is characterised in that the region determination sub-module, including：
    Confidence calculations subelement, for calculate the target suspect region respectively with before current time in the first preset period of time Associated confidence in institute's detection image between suspect region；
    Second area determining unit, for by default machine learning algorithm, being more than the first predetermined threshold value from associated confidence The target suspect region in determine target traffic lights regions.
  18. 18. device according to claim 10, it is characterised in that described device also includes：
    Image determining module, in the second preset period of time in institute's detection image, determining traffic lights before current time Associated confidence is more than the image of the second predetermined threshold value between region and the target traffic lights region；
    As a result correction verification module, the testing result for traffic lights conversion of image is determined for foundation, to described to be checked The testing result for traffic lights conversion of altimetric image is corrected.
CN201610472493.8A 2016-06-22 2016-06-22 Intelligent vehicle driving reminding method and device Active CN107527511B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201610472493.8A CN107527511B (en) 2016-06-22 2016-06-22 Intelligent vehicle driving reminding method and device

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201610472493.8A CN107527511B (en) 2016-06-22 2016-06-22 Intelligent vehicle driving reminding method and device

Publications (2)

Publication Number Publication Date
CN107527511A true CN107527511A (en) 2017-12-29
CN107527511B CN107527511B (en) 2020-10-09

Family

ID=60735549

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201610472493.8A Active CN107527511B (en) 2016-06-22 2016-06-22 Intelligent vehicle driving reminding method and device

Country Status (1)

Country Link
CN (1) CN107527511B (en)

Cited By (13)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109389079A (en) * 2018-09-30 2019-02-26 无锡职业技术学院 A kind of traffic lights recognition methods
CN109544955A (en) * 2018-12-26 2019-03-29 广州小鹏汽车科技有限公司 A kind of state acquiring method and system of traffic lights
WO2020001223A1 (en) * 2018-06-29 2020-01-02 北京市商汤科技开发有限公司 Method and device for traffic signal detection and intelligent driving, vehicle, and electronic device
CN111178119A (en) * 2018-11-13 2020-05-19 北京市商汤科技开发有限公司 Intersection state detection method and device, electronic equipment and vehicle
CN111723625A (en) * 2019-03-22 2020-09-29 上海海拉电子有限公司 Traffic light image recognition processing method and device, auxiliary traffic system and storage medium
CN112016344A (en) * 2019-05-28 2020-12-01 深圳市商汤科技有限公司 State detection method and device of signal indicator lamp and driving control method and device
CN112348879A (en) * 2020-10-30 2021-02-09 深圳市优必选科技股份有限公司 Vehicle operation control method and device, electronic equipment and storage medium
CN112885126A (en) * 2019-11-30 2021-06-01 华为技术有限公司 Driving prompting method and related device
CN113129617A (en) * 2019-12-30 2021-07-16 华为技术有限公司 Driving prompting method and device
CN113723229A (en) * 2021-08-16 2021-11-30 浙江大华技术股份有限公司 Signal lamp detection method and device and computer readable storage medium
CN113808117A (en) * 2021-09-24 2021-12-17 北京市商汤科技开发有限公司 Lamp detection method, device, equipment and storage medium
CN113947917A (en) * 2021-09-23 2022-01-18 联想(北京)有限公司 Prompting method and device of signal display device, equipment and storage medium
CN114419926A (en) * 2022-01-24 2022-04-29 浙江海康智联科技有限公司 Intersection left-turning auxiliary early warning system and method based on vehicle-road cooperation

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101739808A (en) * 2008-11-05 2010-06-16 倚天资讯股份有限公司 Stay-to-move reminder for vehicles and reminding method
US20110095907A1 (en) * 2009-10-23 2011-04-28 Fuji Jukogyo Kabushuki Kaisha Right-turn driving support apparatus
CN102908154A (en) * 2012-11-07 2013-02-06 北京工业大学 Method for detecting start reaction time of driver based on machine vision
CN104050447A (en) * 2014-06-05 2014-09-17 奇瑞汽车股份有限公司 Traffic light identification method and device
CN104916150A (en) * 2015-05-12 2015-09-16 三星电子（中国）研发中心 Traffic safety presentation method applied to electronic device and electronic device

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN101739808A (en) * 2008-11-05 2010-06-16 倚天资讯股份有限公司 Stay-to-move reminder for vehicles and reminding method
US20110095907A1 (en) * 2009-10-23 2011-04-28 Fuji Jukogyo Kabushuki Kaisha Right-turn driving support apparatus
CN102908154A (en) * 2012-11-07 2013-02-06 北京工业大学 Method for detecting start reaction time of driver based on machine vision
CN104050447A (en) * 2014-06-05 2014-09-17 奇瑞汽车股份有限公司 Traffic light identification method and device
CN104916150A (en) * 2015-05-12 2015-09-16 三星电子（中国）研发中心 Traffic safety presentation method applied to electronic device and electronic device

Cited By (20)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2020001223A1 (en) * 2018-06-29 2020-01-02 北京市商汤科技开发有限公司 Method and device for traffic signal detection and intelligent driving, vehicle, and electronic device
CN110660254A (en) * 2018-06-29 2020-01-07 北京市商汤科技开发有限公司 Traffic signal lamp detection and intelligent driving method and device, vehicle and electronic equipment
CN109389079A (en) * 2018-09-30 2019-02-26 无锡职业技术学院 A kind of traffic lights recognition methods
CN109389079B (en) * 2018-09-30 2022-02-15 无锡职业技术学院 Traffic signal lamp identification method
CN111178119A (en) * 2018-11-13 2020-05-19 北京市商汤科技开发有限公司 Intersection state detection method and device, electronic equipment and vehicle
CN109544955A (en) * 2018-12-26 2019-03-29 广州小鹏汽车科技有限公司 A kind of state acquiring method and system of traffic lights
CN111723625A (en) * 2019-03-22 2020-09-29 上海海拉电子有限公司 Traffic light image recognition processing method and device, auxiliary traffic system and storage medium
CN111723625B (en) * 2019-03-22 2023-08-08 上海海拉电子有限公司 Traffic light image recognition processing method and device, auxiliary traffic system and storage medium
CN112016344A (en) * 2019-05-28 2020-12-01 深圳市商汤科技有限公司 State detection method and device of signal indicator lamp and driving control method and device
CN112885126A (en) * 2019-11-30 2021-06-01 华为技术有限公司 Driving prompting method and related device
CN113129617A (en) * 2019-12-30 2021-07-16 华为技术有限公司 Driving prompting method and device
CN112348879B (en) * 2020-10-30 2023-12-19 深圳市优必选科技股份有限公司 Vehicle operation control method and device, electronic equipment and storage medium
CN112348879A (en) * 2020-10-30 2021-02-09 深圳市优必选科技股份有限公司 Vehicle operation control method and device, electronic equipment and storage medium
CN113723229A (en) * 2021-08-16 2021-11-30 浙江大华技术股份有限公司 Signal lamp detection method and device and computer readable storage medium
CN113947917A (en) * 2021-09-23 2022-01-18 联想(北京)有限公司 Prompting method and device of signal display device, equipment and storage medium
CN113947917B (en) * 2021-09-23 2023-10-24 联想(北京)有限公司 Prompting method and device for signal display device, equipment and storage medium
CN113808117A (en) * 2021-09-24 2021-12-17 北京市商汤科技开发有限公司 Lamp detection method, device, equipment and storage medium
CN113808117B (en) * 2021-09-24 2024-05-21 北京市商汤科技开发有限公司 Lamp detection method, device, equipment and storage medium
CN114419926A (en) * 2022-01-24 2022-04-29 浙江海康智联科技有限公司 Intersection left-turning auxiliary early warning system and method based on vehicle-road cooperation
CN114419926B (en) * 2022-01-24 2022-12-20 浙江海康智联科技有限公司 Intersection left-turning auxiliary early warning system and method based on vehicle-road cooperation

Also Published As

Publication number Publication date
CN107527511B (en) 2020-10-09

Similar Documents

Publication Publication Date Title
CN107527511A (en) A kind of intelligent vehicle driving based reminding method and device
CN103400111B (en) Method for detecting fire accident on expressway or in tunnel based on video detection technology
CN107066953A (en) It is a kind of towards the vehicle cab recognition of monitor video, tracking and antidote and device
CN102897086B (en) Back car driving information detecting and prompting method and system
CN104182756B (en) Method for detecting barriers in front of vehicles on basis of monocular vision
CN106679633B (en) A kind of vehicle-mounted distance-finding system base and method
CN105844257A (en) Early warning system based on machine vision driving-in-fog road denoter missing and early warning method
CN108037505A (en) A kind of night front vehicles detection method and system
CN108357418A (en) A kind of front truck driving intention analysis method based on taillight identification
CN101984478A (en) Abnormal S-type driving warning method based on binocular vision lane marking detection
CN107891808A (en) Driving based reminding method, device and vehicle
JP2020194263A (en) Accident analysis device, accident analysis method, and program
JP4814816B2 (en) Accident occurrence prediction simulation apparatus, method and program, safety system evaluation apparatus and accident alarm apparatus
CN107886034A (en) Driving based reminding method, device and vehicle
CN109887276B (en) Night traffic jam detection method based on fusion of foreground extraction and deep learning
CN106254827A (en) A kind of group&#39;s mist Intelligent Recognition method for early warning and device thereof
CN107909012A (en) A kind of real-time vehicle tracking detection method and device based on disparity map
CN110458050A (en) Vehicle based on Vehicular video cuts detection method and device
JP2020194528A (en) Accident analysis device, accident analysis method, and program
CN104972972A (en) Driving safety assisting method and system and electronic equipment
CN106203273A (en) The lane detection system of multiple features fusion, method and senior drive assist system
CN104512329A (en) Reversing safety auxiliary method, reversing safety auxiliary device and automobile
CN109895694B (en) Lane departure early warning method and device and vehicle
CN113076821A (en) Event detection method and device
CN109421598A (en) A kind of vehicle DAS (Driver Assistant System), vehicle early warning method and device

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
GR01 Patent grant