CN107993470A - Count down traffic signal lamp condition detection method and the monitoring system based on the method - Google Patents
Count down traffic signal lamp condition detection method and the monitoring system based on the method Download PDFInfo
- Publication number
- CN107993470A CN107993470A CN201610941081.4A CN201610941081A CN107993470A CN 107993470 A CN107993470 A CN 107993470A CN 201610941081 A CN201610941081 A CN 201610941081A CN 107993470 A CN107993470 A CN 107993470A
- Authority
- CN
- China
- Prior art keywords
- signal lamp
- color
- character
- count down
- video
- 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.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/097—Supervising of traffic control systems, e.g. by giving an alarm if two crossing streets have green light simultaneously
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Image Analysis (AREA)
Abstract
The present invention provides a kind of count down traffic signal lamp condition detection method based on video and the monitoring system based on the method.This method obtains the signal lamp section in present frame first, and then colour recognition is carried out to determine whether color mistake, then logical operation is carried out to the signal lamp color of each phase in crossing to determine whether color conflict, then countdown character picture is pre-processed and uses fuzzy PCA methods identification character, judge that character has zero defect, whether there is Lothrus apterus in the chaotic, time of phasetophase to the sequential of each phase.The monitoring system includes camera unit, industrial control computer, the network equipment and computer detection system, industrial control computer receives the image sequence of camera unit shooting through cable, pass through the state of the real-time marker lamp of the method for machine vision, computer supervisory control system is sent information to when signal lamp failure, vehicle supervision department is taken appropriate measures in time, reduces the generation of traffic accident.
Description
Technical field
Invention is related to intelligent traffic monitoring system, and in particular to a kind of count down traffic signal lamp state based on video is known
Other method and the monitoring system based on this method.
Background technology
Traffic lights have very important effect in traffic dispersion and management, to more and more crowded city road
Road, traffic lights are even more indispensable, they are well-regulated logical in causing the wagon flow in city, the stream of people in a limited space
OK, vehicle flowrate and traffic safety be ensure that to greatest extent.But due to working under rugged environment outdoors, traffic signals
Lamp is easily damaged, simultaneously because traffic lights are dispersed in each crossing in city, its state is difficult to be obtained in time by manual method
Take, so as to cause vehicle supervision department to obtain the state of signal lamp in real time and traffic is managed, work as signal lamp failure
When, traffic congestion is easily caused, or even cause traffic accident.
Current existing signal lamp state detecting system is had the method that is detected based on hardware circuit and based on video
Detection method.The former by being detected to the electric current inside signal lamp or semaphore, voltage, such method because of signal lamp inside
Structure is different and uses different implementations, and without universality, in addition such method judges according to the value of electric current, voltage
The working status of signal lamp, does not have identity to the sequential confusion of countdown signal lamp；The solid color that the latter passes through signal lamp
The state of signal lamp is identified in the color characteristic in space, shape facility, texture feature etc., but since traffic lights are in
Open air, is vulnerable to illumination, the influence of day-night change, features described above, particularly color characteristic, it is difficult to effectively portray the spy of signal
Point, so as to influence detection result, at the same time these methods are only the state of marker lamp, not to countdown character into
Row identification.
The content of the invention
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of count down traffic signal lamp based on video
State identification method.
Present invention also offers a kind of signal lamp monitoring system based on the method.
The recognition methods of countdown signal lamp state and monitoring system provided by the invention based on video can identify traffic
Signal lamp color is incorrect, the display of phase color conflict, count down time is wrong, sequential is incorrect and phasetophase time conflict
These states, and these status informations can be sent to vehicle supervision department in time.
In order to achieve the above object, the technical solution adopted by the present invention is：Count down traffic signal lamp shape based on video
State recognition methods obtains the signal lamp section in present frame first, and then carries out colour recognition to determine whether color mistake, connects
And logical operation is carried out to determine whether color conflict to the signal lamp color of each phase in crossing, then to countdown character picture
Pre-processed and use fuzzy PCA methods identification character, judge character have zero defect, each phase sequential whether chaotic, phase
Between time have Lothrus apterus；Monitoring system based on the method include camera unit, industrial control computer, the network equipment and
Computer supervisory control system, industrial control computer receive the image sequence of camera unit shooting through cable, pass through machine vision
The real-time marker lamp of method state, send information to computer supervisory control system when signal lamp failure.
Further, the count down traffic signal lamp state identification method based on video, comprises the following steps：
The first step, obtains the current frame image of pending signal lamp；
Second step, obtains signal lamp section in present frame；
3rd step, marker lamp color attribute；
4th step, determines whether color mistake, if the color outside signal lamp presentation is red, green, yellow, pipe is issued by error message
Reason center；
5th step, if without color mistake, carries out logical operation, to determine whether phasetophase to the signal lamp color of each phase in crossing
Color conflict, if having conflict if will send information to administrative center, otherwise carry out countdown character recognition；
6th step, pre-processes signal lamp countdown character picture；
7th step, identifies countdown character by fuzzy PCA recognition methods and judges that character whether there is defect, if character exists
Defect then transmits information to administrative center, otherwise carries out sequential judgement；
8th step, the recognition result of two frame adjacent to same signal lamp carries out judging whether sequential confusion, if sequential is chaotic
Egrabage is then sent to administrative center, is otherwise carried out in next step；
9th step, collaboration judges the count down time of each phase at same crossing with the presence or absence of conflict, will punching if there is conflict
Prominent information is sent to administrative center, otherwise returns to the first step.
Further, foregoing marker lamp color attribute is that two color space of joint RGB and HSI is realized, it is wrapped
Include the following steps：
(1) the three-component maximum difference of R, G, B of each pixel is calculated in rgb space), according to maximum difference by picture
Element is divided into colored and two class pixel of achromaticity, even, pixelIt is colored, is otherwise achromaticity；
(2) ratio of colour element in signal lamp section is calculatedIf<T 2 , then it is assumed that current demand signal lamp face
Color mistake；Otherwise signal lamp interval graph picture is transformed into HSI spaces；
(3) histogram of the H components of colour element is calculated, counts centered on peak value, be less than given threshold value with peak value deviationT 3
Domain color average, according to domain color average to canonical red（H=0）, yellow（H=0.1667）, green（H=0.3333）Away from
From the color for judging signal lamp is red, yellow, and green or other colors, if other colors, then it is assumed that current demand signal lamp color is wrong
By mistake.
Further, foregoing fuzzy PCA recognition methods process is as follows：
(1) sample set of countdown character group is collected, their fuzziness is calculated and subset is divided into according to fuzziness；
(2) the PCA subspace projection matrixes of each character group are generated respectively to each subset；
(3) calculate the fuzziness of current character group to be identified and select corresponding subset accordingly, and then calculate respectively to be identified
Character group each subspace projection residual errors distance, obtain least residual distance and distance be less than given threshold value subclass be know
Not as a result, being character errors if minimum range is more than given threshold value.
The count down traffic signal lamp monitoring system, includes camera unit, industrial control computer, the network equipment
And computer detection system.
The camera unit is installed on the Traffic signal post at crossing or on portal frame.
In the consumer unit of the industrial control computer installation by the road.
The camera unit is connected with industrial control computer by cable.
The industrial control computer is connected by the network equipment with computer detection system.
The industrial control computer receives the image sequence of camera unit shooting, is known by the method for machine vision
Level signal lamp state, and computer detection system is passed to by network unit.
The camera unit includes multiple high-definition cameras for being respectively used to collection crossing all directions information, for adopting
Collect motor vehicle signal lamp, the non-motor vehicle signal lamp of respective direction.
The count down traffic signal lamp monitoring system, an industrial control computer detect all letters at a crossing
Signal lamp state, realizes the detection of each signal lamp color, countdown sequential, phase color and time conflict.
Compared with prior art, the beneficial effects of the present invention are：
(1) by combining two color space of RGB and HSI, using the colored domain color attribute of method identification of statistics with histogram, have
The problem of signal lamp under different illumination conditions is reflective, color is extensive is overcome to effect, improves the accuracy of colour recognition；
(2) character group is divided into according to fog-level by different subsets using fuzzy PCA recognition methods, and passes through principal component point
Analysis method generates projection matrix, and the feature of corresponding character group is portrayed with this, finally selects phase according to the fuzziness of character to be identified
The subset answered is identified, and overcomes the problem of discrimination is not high caused by the image obtained under different image-forming conditions obscures；
(3) count down traffic signal lamp monitoring system is by camera unit, industrial control computer, the network equipment and computer
Detecting system forms, and the detection of all signal lamp states at crossing, including each signal are realized by an industrial control computer
The detection of lamp color, countdown sequential, phase color and time conflict, in addition, video camera can also be used to gathering vehicle flowrate, violating the regulations
Other traffic informations such as information, realize the greatest benefit of system；
(4) the count down traffic signal lamp state identification method and monitoring system based on video described in, realize crossing signals
Fault message simultaneously can be notified vehicle supervision department in time by the automatic identification of lamp, it is taken appropriate measures and relieved traffic congestion,
Traffic safety is effectively guaranteed, realizes road rate to greatest extent.
Brief description of the drawings
Attached drawing 1 is the system construction drawing of the present invention.
Attached drawing 2 is the image of video camera shooting.
Attached drawing 3 is the flow chart of the count down traffic signal lamp state identification method based on video of the present invention.
Attached drawing 4 is the signal lamp section manually obtained.
Embodiment
The present invention is described in further details with reference to the accompanying drawings and detailed description.
As shown in Figure 1, the count down traffic signal lamp condition monitoring system based on video of the present invention, includes video camera list
Member, industrial control computer, the network equipment and computer detection system.
The camera unit includes multiple high-definition cameras, is separately mounted on the Traffic signal post at crossing or portal frame
On, for gather to signal information.
In the consumer unit of the industrial control computer installation by the road.
The camera unit is connected with industrial control computer by cable.
The industrial control computer is connected by the network equipment with computer detection system.
The industrial control computer receives the image sequence of camera unit shooting, is identified by the method for machine vision
Signal lamp state, and computer detection system is passed to by network unit.
Image that the camera acquisition arrives as shown in Fig. 2, due to video camera be installed on to lamp stand or gantry
On frame, signal lamp is smaller in the picture, and signal lamp may have it is multiple, in addition to signal lamp, since video camera obtains crossing
Panorama, other traffic informations at crossing are further comprises in image, therefore video camera obtains video information and monitored for other
System uses, and realizes the greatest benefit of hardware system.
As shown in figure 3, the count down traffic signal lamp state identification method based on video, comprises the following steps:
The first step, obtains the current frame image of pending signal lamp；
Second step, obtains signal lamp section in present frame, since video camera and signal location are relatively fixed, in video sequence
In row image, the position of signal lamp is fixed, in order to reduce the treating capacity of information, improves efficiency, signal lamp section is by artificial
Delimit, as shown in figure 4, how many signal lamp just draws how many a sections in figure；
3rd step, marker lamp color attribute；
4th step, determines whether color mistake, if the color outside signal lamp presentation is red, green, yellow, pipe is issued by error message
Reason center；
5th step, if without color mistake, carries out logical operation, to determine whether phasetophase to the signal lamp color of each phase in crossing
Color conflict, if having conflict if will send information to administrative center, otherwise carry out countdown character recognition；
6th step, pre-processes signal lamp countdown character picture；
7th step, identifies countdown character by fuzzy PCA recognition methods and judges that character whether there is defect, if character exists
Defect then transmits information to administrative center, otherwise carries out sequential judgement；
8th step, the recognition result of two frame adjacent to same signal lamp carries out judging whether sequential confusion, if sequential is chaotic
Egrabage is then sent to administrative center, is otherwise carried out in next step；
9th step, collaboration judges the count down time of each phase at same crossing with the presence or absence of conflict, will punching if there is conflict
Prominent information is sent to administrative center, otherwise returns to the first step.
Wherein, marker lamp color attribute is that two color space of joint RGB and HSI realizes that it comprises the following steps：
(1) the three-component maximum difference of R, G, B of each pixel is calculated in rgb space), i.e.,
Then pixel is divided into by colored and two class pixel of achromaticity according to maximum difference, i.e.,
Represent pixelIt is colored, is otherwise achromaticity pixel,T 1 For the threshold value of setting；
(2) ratio of colour element in signal lamp section is calculated
M × n is the size in signal lamp section, if<T2, then it is assumed that current demand signal lamp color mistake, T2For setting
Threshold value；Otherwise signal lamp interval graph picture is transformed into HSI spaces；
(3) histogram of the H components of colour element is calculated
Ask the peak value of colour element histogram correspondingMHValue：
Calculate withMHCentered on, withMHDeviation is less than given threshold valueT 3 Domain color average：
Then domain color average is calculated respectively to canonical red（H=0）, yellow（H=0.1667）, green（H=0.3333）Away from
From：
Order
If ,Then the color of signal lamp isminDistCorresponding color, if, then signal lamp
Color be reddish yellow it is green outside color, it is believed that current demand signal lamp color mistake.
Wherein, phase color logic judgment process is:
The signal lamp color for making all directions four direction is respectivelyIf
Then think that current phase color is correct, otherwise there are phase color conflict.
Wherein, the preprocessing process of signal lamp countdown character picture is included：Countdown word group symbol image is carried out big
Small normalized, it is all 25*30 pixels to make character group image size；Character group image after normalization is carried out at gray processing
Reason.
Wherein, it is as follows that PCA recognition methods processes are obscured：
(1) sample set of countdown character group is collected, the fuzziness of sample is calculated and subset is divided into according to fuzziness；
(2) the PCA subspace projection matrixes of each character group are generated respectively to each subset；
(3) calculate the fuzziness of current character group to be identified and select corresponding subset accordingly, and then calculate respectively to be identified
Character group each subspace projection residual errors distance, obtain least residual distance and distance be less than given threshold value subclass be know
Not as a result, being character errors if minimum range is more than given threshold value.
Further, the calculating process of fuzziness is as follows：
To imageI(containKA pixel), orderV ij Represent pixel(i,j)'sk×kNeighborhood,p(i,j) expression pixel (i,j) ash
Degree, neighborhoodV ij Gray feature be：
WhereinM(i,j) andm(i,j) it is neighborhoodV ij Interior maximum gray scale and minimal gray, whenM(i,j)= m(i,j) when,
Order：
WhereinD 2 >D 1 >0For variance threshold values,D(i,j) beV ij Gray variance
Order：, whereinTFor Triangle Module, it is defined asT(a,b）=a∧b
Calculate non fuzziness cardinality：
ImageIFuzzinessFuzzyDFor：
FuzzinessFuzzyDValue range be [0,1], value it is bigger, represent image it is fuzzyyer.
Further, the calculating of PCA subspace projections matrix and assorting process are as follows：
(1) character sample image column is lined up into characteristic vector, calculates the covariance matrix of subclassR i (i=1,2,…,K)；
(2) asked with svd algorithmR i Feature vector(j=1,2,…,p) and characteristic value；
(3) characteristic value is arranged in decreasing order, order：
Wherein 0<r<1, mTo make the minimum positive integer that above formula is set up, then the dimension of subspace ism；
Subspace is：；
The eigenmatrix of subspace is：；
The projection matrix of subspace is：
(4) projection residual errors distance of the character to be identified in each subspace is calculated：
Least residual distance is：
Obtain least residual class be：
IfThen character to be identified is theiClass, otherwise character errors.
After the character recognition of each phase comes out, the judgement whether sequential confusion in phase, the sequential of each phase are carried out
Deterministic process is：
Order PointNot Wei the count down time value that identifies of present frame and former frame, if, then
Sequential is correct, and otherwise sequential is chaotic in phase.
Phasetophase timing conflict deterministic process is：
OrderThe count down time value of respectively current four phases in all directions, if in the presence of
One of following state, then be phase time conflict
Wherein For the amber light duration of setting.
Claims (6)
1. the count down traffic signal lamp state identification method based on video, it is characterised in that, comprise the steps of:
The first step, obtains the current frame image of pending signal lamp；
Second step, obtains signal lamp section in present frame；
3rd step, marker lamp color attribute；
4th step, determines whether color mistake, if the color outside signal lamp presentation is red, green, yellow, pipe is issued by error message
Reason center；
5th step, if without color mistake, carries out logical operation, to determine whether phase to the signal lamp color of each phase in crossing
Between color conflict, if having conflict if will send information to administrative center, otherwise carry out countdown character recognition；
6th step, pre-processes signal lamp countdown character picture；
7th step, identifies countdown character by fuzzy PCA recognition methods and judges that character whether there is defect, if character exists
Defect then transmits information to administrative center, otherwise carries out sequential judgement；
8th step, the recognition result of two frame adjacent to same signal lamp carries out judging whether sequential confusion, if sequential is chaotic
Egrabage is then sent to administrative center, is otherwise carried out in next step；
9th step, collaboration judges the count down time of each phase at same crossing with the presence or absence of conflict, will punching if there is conflict
Prominent information is sent to administrative center, otherwise returns to the first step；
Wherein, the marker lamp color attribute is that two color space of joint RGB and HSI realizes that it further comprises
The following steps：
(1) the three-component maximum difference of R, G, B of each pixel is calculated in rgb space, is divided into pixel according to maximum difference
Colored and two class pixel of achromaticity；
(2) ratio of colour element in signal lamp section is calculated, if the ratio of colour element is less than given threshold value, then it is assumed that
Current demand signal lamp color mistake, is otherwise transformed into HSI spaces by signal lamp interval graph picture；
(3) histogram of the H components of colour element is calculated, counts centered on peak value, be less than given threshold value with peak value deviationT 3
Domain color average, according to domain color average to canonical red（H=0）, yellow（H=0.1667）, green（H=0.3333）Away from
From the color for judging signal lamp is red, yellow, and green or other colors, if other colors, then it is assumed that current demand signal lamp color is wrong
By mistake；
Wherein, the fuzzy PCA recognition methods process is as follows：
(1) sample set of countdown character group is collected, their fuzziness is calculated and subset is divided into according to fuzziness；
(2) the PCA subspace projection matrixes of each character group are generated respectively to each subset；
(3) calculate the fuzziness of current character group to be identified and select corresponding subset accordingly, and then calculate respectively to be identified
Character group each subspace projection residual errors distance, obtain least residual distance and distance be less than given threshold value subclass be know
Other result.
2. the count down traffic signal lamp state identification method according to claim 1 based on video, it is characterised in that institute
Color to be identified is divided into red, yellow, and green and other colors, effectively overcomed by the color characteristic stated according to colored domain color
The problem of signal lamp under different illumination conditions is reflective, color is extensive.
3. the count down traffic signal lamp state identification method according to claim 1 based on video, it is characterised in that institute
Character group is divided into different subsets by the fuzzy PCA recognition methods stated according to fog-level, and is given birth to by principal component analytical method
Into projection matrix, the feature of corresponding character group is portrayed with this, corresponding subset is finally selected according to the fuzziness of character to be identified
It is identified, overcomes the problem of discrimination is not high caused by the image obtained under different image-forming conditions obscures.
4. a kind of count down traffic signal lamp condition monitoring system based on video, it is characterised in that include camera unit, work
Industry control computer, the network equipment and computer supervisory control system, the camera unit be installed on crossing Traffic signal post on or
On portal frame, in the consumer unit of the industrial control computer installation by the road, the camera unit and Industry Control meter
Calculation machine is connected by cable, and the industrial control computer is connected by the network equipment with computer supervisory control system, the industry
Control computer receives the image sequence of camera unit shooting, by the method marker lamp state of machine vision, and leads to
Cross network unit and pass to computer supervisory control system.
5. the count down traffic signal lamp condition monitoring system based on video as claimed in claim 4, it is characterised in that shooting
Machine unit includes multiple high-definition cameras for being respectively used to collection crossing all directions information, for gathering the signal of respective direction
Lamp.
6. the count down traffic signal lamp condition monitoring system based on video as claimed in claim 4, it is characterised in that one
Industrial control computer detects all signal lamp states at a crossing, realizes each signal lamp color, countdown sequential, phase color
With the detection of time conflict.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610941081.4A CN107993470A (en) | 2016-10-26 | 2016-10-26 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610941081.4A CN107993470A (en) | 2016-10-26 | 2016-10-26 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107993470A true CN107993470A (en) | 2018-05-04 |
Family
ID=62028515
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610941081.4A Pending CN107993470A (en) | 2016-10-26 | 2016-10-26 | Count down traffic signal lamp condition detection method and the monitoring system based on the method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107993470A (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108538068A (en) * | 2018-06-04 | 2018-09-14 | 太仓迪米克斯节能服务有限公司 | A kind of traffic reminding method and its system based on countdown length |
| CN109767637A (en) * | 2019-02-28 | 2019-05-17 | 杭州飞步科技有限公司 | The method and apparatus of the identification of countdown signal lamp and processing |
| CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
| CN110826456A (en) * | 2019-10-31 | 2020-02-21 | 青岛海信网络科技股份有限公司 | Countdown board fault detection method and system |
| CN111882910A (en) * | 2020-06-15 | 2020-11-03 | 太原市高远时代科技有限公司 | High-accuracy traffic signal lamp fault detection method and system |
| CN112101272A (en) * | 2020-09-23 | 2020-12-18 | 北京百度网讯科技有限公司 | Traffic light detection method and device, computer storage medium and road side equipment |
| CN112991791A (en) * | 2019-12-13 | 2021-06-18 | 上海商汤临港智能科技有限公司 | Traffic information identification and intelligent driving method, device, equipment and storage medium |
| CN113065466A (en) * | 2021-04-01 | 2021-07-02 | 安徽嘻哈网络技术有限公司 | Traffic light detection system for driving training based on deep learning |
| CN113593261A (en) * | 2021-06-30 | 2021-11-02 | 高新兴科技集团股份有限公司 | Signal lamp countdown processing method, device, system, equipment and storage medium |
| CN114898582A (en) * | 2022-05-26 | 2022-08-12 | 浙江中控信息产业股份有限公司 | Intelligent traffic signal lamp |
| CN113065466B (en) * | 2021-04-01 | 2024-06-04 | 安徽嘻哈网络技术有限公司 | Deep learning-based traffic light detection system for driving training |
-
2016
- 2016-10-26 CN CN201610941081.4A patent/CN107993470A/en active Pending
Cited By (15)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108538068A (en) * | 2018-06-04 | 2018-09-14 | 太仓迪米克斯节能服务有限公司 | A kind of traffic reminding method and its system based on countdown length |
| CN109767637A (en) * | 2019-02-28 | 2019-05-17 | 杭州飞步科技有限公司 | The method and apparatus of the identification of countdown signal lamp and processing |
| CN109767637B (en) * | 2019-02-28 | 2021-08-10 | 杭州飞步科技有限公司 | Method and device for identifying and processing countdown signal lamp |
| CN110782692A (en) * | 2019-10-31 | 2020-02-11 | 青岛海信网络科技股份有限公司 | Signal lamp fault detection method and system |
| CN110826456A (en) * | 2019-10-31 | 2020-02-21 | 青岛海信网络科技股份有限公司 | Countdown board fault detection method and system |
| CN112991791A (en) * | 2019-12-13 | 2021-06-18 | 上海商汤临港智能科技有限公司 | Traffic information identification and intelligent driving method, device, equipment and storage medium |
| CN112991791B (en) * | 2019-12-13 | 2022-07-26 | 上海商汤临港智能科技有限公司 | Traffic information identification and intelligent driving method, device, equipment and storage medium |
| CN111882910A (en) * | 2020-06-15 | 2020-11-03 | 太原市高远时代科技有限公司 | High-accuracy traffic signal lamp fault detection method and system |
| CN112101272A (en) * | 2020-09-23 | 2020-12-18 | 北京百度网讯科技有限公司 | Traffic light detection method and device, computer storage medium and road side equipment |
| CN112101272B (en) * | 2020-09-23 | 2024-05-14 | 阿波罗智联(北京)科技有限公司 | Traffic light detection method, device, computer storage medium and road side equipment |
| CN113065466A (en) * | 2021-04-01 | 2021-07-02 | 安徽嘻哈网络技术有限公司 | Traffic light detection system for driving training based on deep learning |
| CN113065466B (en) * | 2021-04-01 | 2024-06-04 | 安徽嘻哈网络技术有限公司 | Deep learning-based traffic light detection system for driving training |
| CN113593261A (en) * | 2021-06-30 | 2021-11-02 | 高新兴科技集团股份有限公司 | Signal lamp countdown processing method, device, system, equipment and storage medium |
| CN113593261B (en) * | 2021-06-30 | 2022-07-12 | 高新兴科技集团股份有限公司 | Signal lamp countdown processing method, device, system, equipment and storage medium |
| CN114898582A (en) * | 2022-05-26 | 2022-08-12 | 浙江中控信息产业股份有限公司 | Intelligent traffic signal lamp |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107993470A (en) | Count down traffic signal lamp condition detection method and the monitoring system based on the method | |
| CN109636777A (en) | A kind of fault detection method of traffic lights, system and storage medium | |
| CN110197589B (en) | Deep learning-based red light violation detection method | |
| CN109064768B (en) | One kind being used for traffic signals facility staff realtime monitoring system and method | |
| CN104331710B (en) | On off state identifying system | |
| CN105744232A (en) | Method for preventing power transmission line from being externally broken through video based on behaviour analysis technology | |
| CN109484935A (en) | A kind of lift car monitoring method, apparatus and system | |
| CN103456024B (en) | A kind of moving target gets over line determination methods | |
| CN111882910B (en) | Traffic signal lamp fault detection method and system | |
| CN107103330A (en) | A kind of LED status recognition methods and device | |
| CN111626139A (en) | Accurate detection method for fault information of IT equipment in machine room | |
| CN102496282A (en) | Traffic intersection signal light state identification method based on RGB color transformation | |
| CN104780361B (en) | A kind of quality evaluating method of city video monitoring system | |
| US10824885B2 (en) | Method and apparatus for detecting braking behavior of front vehicle of autonomous vehicle | |
| CN111654692B (en) | Camera detection method and device, camera and storage medium | |
| CN103236181A (en) | Traffic signal lamp state monitoring system and method based on machine vision | |
| CN112487891B (en) | Visual intelligent dynamic identification model construction method applied to electric power operation site | |
| CN110175533A (en) | Overpass traffic condition method of real-time, device, terminal and storage medium | |
| CN109446926A (en) | A kind of traffic monitoring method and device, electronic equipment and storage medium | |
| CN114495011A (en) | Non-motor vehicle and pedestrian illegal intrusion identification method based on target detection, storage medium and computer equipment | |
| CN103324957A (en) | Identification method and identification device of state of signal lamps | |
| CN103106400B (en) | A kind of method for detecting human face and device | |
| CN116682162A (en) | Robot detection algorithm based on real-time video stream | |
| CN110826456A (en) | Countdown board fault detection method and system | |
| CN104091402A (en) | Distinguishing system and distinguishing method of multi-state alarm color-changing lamp of 24V power supply cabinet |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20180504 |
|
| WD01 | Invention patent application deemed withdrawn after publication |