CN108417043A - A kind of detection method that high beam persistently opens - Google Patents
A kind of detection method that high beam persistently opens Download PDFInfo
- Publication number
- CN108417043A CN108417043A CN201810131747.9A CN201810131747A CN108417043A CN 108417043 A CN108417043 A CN 108417043A CN 201810131747 A CN201810131747 A CN 201810131747A CN 108417043 A CN108417043 A CN 108417043A
- Authority
- CN
- China
- Prior art keywords
- light
- high beam
- car light
- dazzling
- bright
- 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
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/017—Detecting movement of traffic to be counted or controlled identifying vehicles
- G08G1/0175—Detecting movement of traffic to be counted or controlled identifying vehicles by photographing vehicles, e.g. when violating traffic rules
Abstract
The invention discloses a kind of detection methods that high beam persistently opens, and include the following steps:Step 1, setting confidence virtual detection region step 2 judge the dazzling degree step 3 of light, judge whether the changing rule of dazzling degree meets 3.1 target following 3.2 of feature of the bright-dark cut of long-and-short distant light and track chain light changing rule 3.3 and judge whether persistently to open beneficial effects of the present invention:For this method there are the complex environment of many interference suitable for the video that traffic monitoring camera takes, step is simple, and convenient to carry out, calculation amount is small, and as a result accuracy is high, and the result provided can carry out efficient detection to the high beam persistently opened.
Description
Technical field
The present invention relates to the technical field of video processing of intelligent traffic monitoring system more particularly to a kind of vehicle at night to travel
The detection method that high beam under background persistently opens.
Background technology
Car headlamp is divided into two kinds of situations of low beam mode and high beam mode, and high beam is exactly high beam mode therein.Closely
Light lamp shines close, and high beam photograph is remote.According to traffic law, the use of dazzle light has clear stipulaties.Because far
Light lamp can be brought convenience for driving when using correct and safety, but if is also brought along greatly using incorrect
Security risk.In intelligent traffic monitoring system, for night long-and-short distant light automatic detection identify, external factor interference compared with
Few, when car light texture understands, the false drop rate of high beam detection is not high.But when vehicle condition situation is complicated, bad environments etc. it is external because
When the element huge interference of formation and when vehicle itself light texture is not clear enough, it can be made when system is detected high beam
It, i.e., can be dipped headlight flase drop at high beam, this accuracy for detecting very big reduction high beam at larger flase drop.For this purpose, this
Invention proposes a kind of detection method that the high beam of night running vehicle persistently opens, and the method is efficiently concise.
Invention content
To solve deficiency in the prior art, the object of the present invention is to provide a kind of detection sides that high beam persistently opens
Method, it is intended to solve in the prior art due in video monitoring picture night running vehicle high beam detect when, because of vehicle condition situation
Complexity causes monitored picture to contain much information, and redundancy is more, and road ponding ambient conditions is complicated, causes monitored picture to occur tight
Weight is reflective, and the problem of causing serious interference is detected to high beam.
To achieve the above object, the technical scheme is that:A kind of detection method that high beam persistently opens, including with
Lower step:
Step 1, setting confidence virtual detection region
In order to which the light size for ensureing detected is suitable, tracking is stablized, and excludes unnecessary interference, in road area
Set confidence virtual detection line cf and bg in adeh, i.e., only detection vehicle drives to the car light of confidence virtual detection region bcfg and opens
Open situation;
Step 2 judges the dazzling degree of light
According to the design principle of headlamp, setting bright-dark cut L:
I.e. when the motor vehicle for opening high beam does not reach bright-dark cut L, prison, which claps camera, can receive transmitting
The light gone out, collected light are dazzling hot spots;
When motor line crosses bright-dark cut L, prison claps camera and does not receive the light that high beam is launched then, only
It is in not dazzling hot spot only there was only reflected beam portion, collected light;
When open dipped headlight motor vehicle before and after driving to bright-dark cut L, can not possibly have irradiation light enter take the photograph
As head, collected car light is not dazzling hot spot in whole driving process;
By judging the size of dazzling degree, specifically judge whether be high beam method it is as follows:
For the gray-scale map of car light image, high threshold th1When can correspond to a car light area S1, Low threshold th2When can correspond to
One car light area S2, the dazzling degree of the car light is may determine that according to the size of the ratio K in formula 1:
Threshold value th is to judge dazzling threshold value, it is believed that when K is more than th, car light is dazzling, and car light is not dazzling when being less than th;Threshold
Value th is empirical value, between preferably 0.5 to 1.5, can be adjusted according to actual environment;
Step 3 judges whether the changing rule of dazzling degree meets the feature of the bright-dark cut of long-and-short distant light
3.1 target following
When the headlamp of motor vehicle enters cf confidence detection lines, the company by video frame will be carried out to the car light of the motor vehicle
Continuous tracking, until the headlamp of the motor vehicle walks out confidence detection line bg;
3.2 tracking chain light changing rules
During tracking, while the calculating of dazzling degree K is carried out to the car light, at the end of tracking statistics with
Track chain light changing rule;The changing rule that chain is tracked for a complete high beam, typically before bright-dark cut L
It is more than threshold value th for K, is that K is less than threshold value th after bright-dark cut L;
3.3 judge whether persistently to open
When actually judging whether high beam persistently opens, it can be determined, that is, be thought in light and shade according to certain ratio
The K that the car light in the total video frame of n% is had more than before the L of line of demarcation is more than threshold value th, is had more than after bright-dark cut L
The K of the car light is less than threshold value th in the total video frame of m%, and the total video frame number at least up to hands over rule to assert and persistently opens distance light
The number of seconds of lamp and the product of shooting frame rate;If meeting this changing rule thinks that the car light persistently opens high beam;Equally, if
Car light variation is unsatisfactory for above-mentioned rule and then not thinks it is that high beam persistently opens;In addition, if the dazzling degree variation of car light does not have
Light and shade change procedure is then also considered as not being that high beam persistently opens.
Preferably, in the 3.3 of the step 3, n takes 90, m to take 95.
Beneficial effects of the present invention:There are many interference suitable for the video that traffic monitoring camera takes for this method
Complex environment, step is simple, convenient to carry out, and calculation amount is small, and as a result accuracy is high, and the result provided can be remote to what is persistently opened
Light lamp carries out efficient detection.
Description of the drawings
Fig. 1 is the confidence virtual detection line being arranged on road.
Fig. 2 is bright-dark cut schematic diagram.
Fig. 3 is schematic diagram of the car light gray scale facula area in high-low threshold value.
Fig. 4 is the dazzling degree changing rule figure that motor vehicle far-reaching lamp persistently opens.
Specific implementation mode
A kind of detection method that high beam persistently opens, includes the following steps:
Step 1, setting confidence virtual detection region
Implement high beam persistently to open the first step of detection to be exactly to set confidence virtual detection line, as shown in Figure 1.In order to protect
The detected light size of card is suitable, and tracking is stablized, and excludes unnecessary interference, and this method is virtually examined using setting confidence
The method of survey line;Adeh is road area in Fig. 1, and motor vehicle is travelled according to arrow direction, and cf and bg are exactly the confidence void set
Quasi- detection line only detects the car light unlatching situation that vehicle drives to confidence virtual detection region bcfg;
Step 2 judges the dazzling degree of light
According to the design principle of headlamp, Fig. 2 cathetus L is bright-dark cut illustrated position (the also straight line in corresponding diagram 1
L), when opening the motor vehicle (No. 1 and No. 2 positions in figure) of high beam without reaching bright-dark cut L, prison bat camera can be with
The light launched is received, thus collected light is dazzling hot spot;When motor vehicle is run over (in figure on No. 3 positions)
When bright-dark cut L, prison claps camera and does not receive the light that high beam is launched then, only only reflected beam portion, therefore
Collected light is in not dazzling hot spot;When open dipped headlight motor vehicle before and after driving to bright-dark cut L, not
There may be irradiation light to enter camera, therefore collected car light is not dazzling hot spot in whole driving process;
So the size by judging dazzling degree, it is possible to determine that whether the car light in the frame figure is high beam;Specifically sentence
Disconnected method is as follows:
For the gray-scale map of car light image, high threshold th1When can correspond to a car light area S1(the inner circle O of Fig. 31), low threshold
Value th2When can correspond to a car light area S2(the outer circle O of Fig. 32), so may determine that according to the size of the ratio K in formula 1
The dazzling degree of the car light;
Threshold value th is to judge dazzling threshold value, it is believed that when K is more than th, car light is dazzling, and car light is not dazzling when being less than th.Threshold
Value th is empirical value, between preferably 0.5 to 1.5, can be adjusted according to actual environment;
Step 3 judges whether the changing rule of dazzling degree meets the feature of the bright-dark cut of long-and-short distant light
3.1 target following
When the headlamp of motor vehicle enters cf confidence detection lines, the company by video frame will be carried out to the car light of the motor vehicle
Continuous tracking, until the headlamp of the motor vehicle walks out confidence detection line bg;
3.2 tracking chain light changing rules
During tracking, while the calculating of dazzling degree K is carried out to the car light, at the end of tracking statistics with
Track chain light changing rule.The changing rule that chain is tracked for a complete high beam, typically before bright-dark cut L
It is more than threshold value th for K, is that K is less than threshold value th after bright-dark cut L.
3.3 judge whether persistently to open
It when actually judging whether high beam persistently opens, can be determined according to certain ratio, it is believed that in light and shade point
The K that the car light in 90% total video frame is had more than before boundary line L is more than threshold value th, is had more than after bright-dark cut L
The K of the car light is less than threshold value th in 95% total video frame, and the total video frame number at least up to friendship rule assert lasting open far
The number of seconds of light lamp and the product of shooting frame rate;If meeting this changing rule thinks that the car light persistently opens high beam;High beam
The dazzling degree changing rule persistently opened is as shown in Figure 4.
Equally, not think it is that high beam persistently opens if car light variation is unsatisfactory for above-mentioned rule;In addition, if car light
Dazzling degree changes no light and shade change procedure and is then also considered as not being that high beam persistently opens.
The invention can preferably detect the high beam persistently opened, and detection algorithm is simple, and efficiently, calculation amount is small.
The foregoing is merely illustrative of the preferred embodiments of the present invention, is not intended to limit the invention, all essences in the present invention
Any modification, equivalent replacement or improvement etc., should all be included in the protection scope of the present invention made by within refreshing and principle.
Claims (2)
1. a kind of detection method that high beam persistently opens, which is characterized in that include the following steps:
Step 1, setting confidence virtual detection region
In order to which the light size for ensureing detected is suitable, tracking is stablized, and excludes unnecessary interference, in road area adeh
Interior setting confidence virtual detection line cf and bg only detect the car light unlatching feelings that vehicle drives to confidence virtual detection region bcfg
Condition;
Step 2 judges the dazzling degree of light
According to the design principle of headlamp, setting bright-dark cut L:
I.e. when the motor vehicle for opening high beam does not reach bright-dark cut L, prison, which is clapped camera and can be received, to be launched
Light, collected light are dazzling hot spots;
When motor line crosses bright-dark cut L, prison claps camera and does not receive the light that high beam is launched then, only
It is in not dazzling hot spot to have reflected beam portion, collected light;
When open dipped headlight motor vehicle before and after driving to bright-dark cut L, can not possibly there is irradiation light to enter camera,
Collected car light is not dazzling hot spot in whole driving process;
By judging the size of dazzling degree, specifically judge whether be high beam method it is as follows:
For the gray-scale map of car light image, high threshold th1When can correspond to a car light area S1, Low threshold th2When can correspond to one
Car light area S2, the dazzling degree of the car light is may determine that according to the size of the ratio K in formula 1:
Threshold value th is to judge dazzling threshold value, it is believed that when K is more than th, car light is dazzling, and car light is not dazzling when being less than th;Threshold value th
It is empirical value, between preferably 0.5 to 1.5, can be adjusted according to actual environment;
Step 3 judges whether the changing rule of dazzling degree meets the feature of the bright-dark cut of long-and-short distant light
3.1 target following
When the headlamp of motor vehicle enters cf confidence detection lines, by the car light of the motor vehicle carry out by video frame it is continuous with
Track, until the headlamp of the motor vehicle walks out confidence detection line bg;
3.2 tracking chain light changing rules
During tracking, while the calculating of dazzling degree K is carried out to the car light, for statistical trace chain at the end of tracking
Light changing rule;The changing rule that chain is tracked for a complete high beam is K typically before bright-dark cut L
More than threshold value th, it is less than threshold value th after bright-dark cut L for K;
3.3 judge whether persistently to open
It when actually judging whether high beam persistently opens, can be determined according to certain ratio, that is, think to demarcate in light and shade
The K that the car light in the total video frame of n% is had more than before line L is more than threshold value th, has more than m%'s after bright-dark cut L
The K of the car light is less than threshold value th in total video frame, and the total video frame number, which is at least up to, hands over rule identification persistently to open high beam
The product of number of seconds and shooting frame rate;If meeting this changing rule thinks that the car light persistently opens high beam;
Equally, not think it is that high beam persistently opens if car light variation is unsatisfactory for above-mentioned rule;In addition, if car light is dazzling
Degree changes no light and shade change procedure and is then also considered as not being that high beam persistently opens.
2. a kind of detection method that high beam persistently opens according to claim 1, which is characterized in that the 3.3 of the step 3
In, n takes 90, m to take 95.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810131747.9A CN108417043B (en) | 2018-02-09 | 2018-02-09 | Detection method for continuous starting of high beam |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810131747.9A CN108417043B (en) | 2018-02-09 | 2018-02-09 | Detection method for continuous starting of high beam |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108417043A true CN108417043A (en) | 2018-08-17 |
CN108417043B CN108417043B (en) | 2020-07-14 |
Family
ID=63127057
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810131747.9A Active CN108417043B (en) | 2018-02-09 | 2018-02-09 | Detection method for continuous starting of high beam |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108417043B (en) |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110646173A (en) * | 2019-09-30 | 2020-01-03 | 青岛讯极科技有限公司 | Detection method for continuous opening of high beam |
CN110823362A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | High beam detection system and detection method thereof |
CN110823533A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Non-standard light detection system and detection method thereof |
CN110823363A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Automobile light detection system and detection method thereof |
CN113628447A (en) * | 2020-05-06 | 2021-11-09 | 杭州海康威视数字技术股份有限公司 | High beam light starting detection method, device, equipment and system |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007272785A (en) * | 2006-03-31 | 2007-10-18 | Matsushita Electric Ind Co Ltd | Image processing type vehicle detector and image processing type vehicle detection method |
CN102298845A (en) * | 2011-08-29 | 2011-12-28 | 安徽超远信息技术有限公司 | Far-beam light video detection method and system |
CN103295399A (en) * | 2013-05-14 | 2013-09-11 | 西安理工大学 | On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics |
CN104008650A (en) * | 2014-05-20 | 2014-08-27 | 宁辉 | Motor vehicle high beam monitoring device and identification method thereof |
CN105184227A (en) * | 2015-08-13 | 2015-12-23 | 安徽超远信息技术有限公司 | Traffic signal control system with automobile high beam light detection function |
CN205388828U (en) * | 2016-01-20 | 2016-07-20 | 王威立 | Motor vehicle far -reaching headlamp is violating regulations to be detected and snapshot system |
CN103413135B (en) * | 2013-07-31 | 2016-08-10 | 东软集团股份有限公司 | A kind of vehicle head lamp bright dark border line detecting method, device and system |
CN107169422A (en) * | 2017-04-21 | 2017-09-15 | 浙江安谐智能科技有限公司 | The method of discrimination of high beam open and-shut mode based on headlamp radiation direction |
CN107563265A (en) * | 2016-06-30 | 2018-01-09 | 杭州海康威视数字技术股份有限公司 | A kind of high beam detection method and device |
-
2018
- 2018-02-09 CN CN201810131747.9A patent/CN108417043B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP2007272785A (en) * | 2006-03-31 | 2007-10-18 | Matsushita Electric Ind Co Ltd | Image processing type vehicle detector and image processing type vehicle detection method |
CN102298845A (en) * | 2011-08-29 | 2011-12-28 | 安徽超远信息技术有限公司 | Far-beam light video detection method and system |
CN103295399A (en) * | 2013-05-14 | 2013-09-11 | 西安理工大学 | On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics |
CN103413135B (en) * | 2013-07-31 | 2016-08-10 | 东软集团股份有限公司 | A kind of vehicle head lamp bright dark border line detecting method, device and system |
CN104008650A (en) * | 2014-05-20 | 2014-08-27 | 宁辉 | Motor vehicle high beam monitoring device and identification method thereof |
CN105184227A (en) * | 2015-08-13 | 2015-12-23 | 安徽超远信息技术有限公司 | Traffic signal control system with automobile high beam light detection function |
CN205388828U (en) * | 2016-01-20 | 2016-07-20 | 王威立 | Motor vehicle far -reaching headlamp is violating regulations to be detected and snapshot system |
CN107563265A (en) * | 2016-06-30 | 2018-01-09 | 杭州海康威视数字技术股份有限公司 | A kind of high beam detection method and device |
CN107169422A (en) * | 2017-04-21 | 2017-09-15 | 浙江安谐智能科技有限公司 | The method of discrimination of high beam open and-shut mode based on headlamp radiation direction |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110823362A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | High beam detection system and detection method thereof |
CN110823533A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Non-standard light detection system and detection method thereof |
CN110823363A (en) * | 2019-09-10 | 2020-02-21 | 李东华 | Automobile light detection system and detection method thereof |
CN110646173A (en) * | 2019-09-30 | 2020-01-03 | 青岛讯极科技有限公司 | Detection method for continuous opening of high beam |
CN113628447A (en) * | 2020-05-06 | 2021-11-09 | 杭州海康威视数字技术股份有限公司 | High beam light starting detection method, device, equipment and system |
CN113628447B (en) * | 2020-05-06 | 2022-12-23 | 杭州海康威视数字技术股份有限公司 | High beam light starting detection method, device, equipment and system |
Also Published As
Publication number | Publication date |
---|---|
CN108417043B (en) | 2020-07-14 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108417043A (en) | A kind of detection method that high beam persistently opens | |
Sina et al. | Vehicle counting and speed measurement using headlight detection | |
US8995723B2 (en) | Detecting and recognizing traffic signs | |
CN105678316B (en) | Active drive manner based on multi-information fusion | |
CN104340109B (en) | Driver assistance system and its operation method | |
CN105931467B (en) | A kind of method and device tracking target | |
CN107808392B (en) | Automatic tracking and positioning method and system for security check vehicle in open scene | |
CN108230254A (en) | A kind of full lane line automatic testing method of the high-speed transit of adaptive scene switching | |
CN102712282A (en) | Method for controlling a headlight arrangement for a vehicle and such a headlight arrangement | |
CN103400111A (en) | Method for detecting fire accident on expressway or in tunnel based on video detection technology | |
CN108357418A (en) | A kind of front truck driving intention analysis method based on taillight identification | |
CN107909012B (en) | Real-time vehicle tracking detection method and device based on disparity map | |
CN205768952U (en) | Night intelligence auxiliary driving device | |
CN110450706A (en) | A kind of adaptive distance light lamp control system and image processing algorithm | |
CN108230690A (en) | A kind of high beam based on convolutional neural networks continues the method for discrimination of opening | |
CN106096504A (en) | A kind of model recognizing method based on unmanned aerial vehicle onboard platform | |
CN109887276B (en) | Night traffic jam detection method based on fusion of foreground extraction and deep learning | |
CN103295399A (en) | On-state judging method of headlights on full beam of night-driving cars based on morphological characteristics | |
CN102712284A (en) | Method for controlling a headlight arrangement for a vehicle and such a headlight arrangement | |
CN108163014A (en) | A kind of engine drivers in locomotive depot Fu Zhu lookout method for early warning and device | |
CN111619324A (en) | Intelligent anti-dazzling method and system for sight tracking automobile | |
CN108834283A (en) | A kind of light that vehicle enters tunnel shifts to an earlier date control system for identifying | |
CN113052159A (en) | Image identification method, device, equipment and computer storage medium | |
Wang et al. | Road edge detection in all weather and illumination via driving video mining | |
CN109508659A (en) | A kind of face identification system and method for crossing |
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 |