CN109740412A - A kind of signal lamp failure detection method based on computer vision - Google Patents
A kind of signal lamp failure detection method based on computer vision Download PDFInfo
- Publication number
- CN109740412A CN109740412A CN201811334431.6A CN201811334431A CN109740412A CN 109740412 A CN109740412 A CN 109740412A CN 201811334431 A CN201811334431 A CN 201811334431A CN 109740412 A CN109740412 A CN 109740412A
- Authority
- CN
- China
- Prior art keywords
- signal lamp
- indicate
- region
- backboard
- indicates
- 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
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of signal lamp failure detection methods based on computer vision, it is characterized in that, include the following steps: step 1): obtaining image from the monitor camera of existing traffic intersection and carry out signal lamp calibration, obtain signal lamp regional ensemble L={ li=(RLi,YLi,BLi,TPi) | i=1,2,3 ..., n }, step 2): signal lamp state is carried out to continue sampling identification, and maintains ordered set the S={ (rs of stateij,ysij,bsij,tpij,tj) | i=1,2 ..., n；J=1,2 ..., m }；Step 3): signal lamp failure is judged according to set S, and is W by the failure logging of i-th of signal lamp backboardi, step 4): if Wi≠ null, then by WiIt is sent to administrative staff；It goes to step 2)；The beneficial effects of the present invention are: without being analyzed using present monitor camera acquired image, realizing the detection of signal lamp failure, have many advantages, such as real-time, convenience in the case where being transformed to signal lamp itself.
Description
Technical field
The present invention relates to computer image processing technology fields, and in particular to a kind of signal lamp detection side of view-based access control model
Method.
Background technique
It is the important link of intelligent transportation system that whether signal lamp working properly, it will affect traffic safety and is
The convenience of people's normal traffic.How the working condition of real time detection signal lamp, and Real-time Feedback signal lamp failure for protect
Card traffic is normally of great significance.
In order to solve the problems, such as that signal lamp failure detects, domestic and international academia, industry propose many schemes, at present absolutely
Most of most successful modes are fault detection module to be added in signal lamp product, but this method has intrinsic defect,
For example reliability, can not then detect, etc. with signal lamp high integration, signal lamp power-off, for this purpose, many be used for signal lamp shape
The computer vision methods of state detection are suggested, wherein the technical solution being closer to the present invention includes: application for a patent for invention
Number: 201510763022.8, patent name: a kind of self-identifying failure traffic signal lamp system discloses a kind of self-identifying failure
Traffic signal lamp system, including traffic lights, the traffic lights can issue traffic signals；The system also includes adopt
Collect image collecting device, controller and the wireless signal transmitting device of running red light for vehicle, controller is controlled, when by the letter
When more than continuous 10 Che Zhongyou given thresholds of signal lamp running red light for vehicle；Application number of invention patent:
201810315357.7, title: traffic lights detection recognition method and system based on unmanned platform propose a kind of base
It is based on HSI component map in the traffic lights detection recognition method of unmanned platform, passes through the training of sample candidate region feature
Classifier is classified using traffic lights region of the classifier to acquisition, obtains the status information of traffic lights；Invention
Number of patent application: 201810174371.X, title: a kind of dividing method of traffic lights discloses a kind of traffic lights
Dividing method, in the traffic lights detection in auxiliary drivings, the application such as automatic Pilot and identifying, due to segmentation threshold
The problem of the segmentation inaccuracy of signal lamp caused by unreasonable gives and a kind of carries out signal lamp segmentation using one-dimensional Gauss model
Method；Application number of invention patent: CN201810298435.7, title: a kind of traffic signal light fault detection method and system, it is public
A kind of traffic signal light fault detection method and system have been opened, it is whether full according to flicker frequency of the main signal light within each period
Whether the default flicker frequency of foot or color change meet pre-set color variation, judge main signal light whether failure；If main signal light
Failure then closes main signal light, while starting auxiliary pilot lamp and issuing failure alarm signal；Application number of invention patent:
CN201810126429.3, title: a kind of method of intelligent recognition circle traffic signal light condition discloses a kind of intelligent recognition
The method of round traffic signal light condition；Application number of invention patent: CN201611154612.1, title: traffic lights are reset
Position method and device, about a kind of traffic lights method for relocating and device, according at least two frame short exposure monitoring images and
The initial information of the preconfigured target traffic lights, determines the position offset of the target traffic lights；Root
According to the initial information of the target traffic lights and the position offset of the target traffic lights, the mesh is relocated
Mark position of the traffic lights in monitoring image；Application number of invention patent: CN201711220616.X, title: one kind is based on
The traffic lights recognition methods that conspicuousness calculates discloses a kind of traffic lights recognition methods calculated based on conspicuousness,
Traffic lights identify after realizing signal lamp calibration deflection；Application number of invention patent: CN201810179893.9, title: a kind of
Signal lamp failure detection system based on Internet of Things is related to a kind of signal lamp failure detection system based on Internet of Things；Invention is special
Sharp application number: CN201711143943.X, title: one using computer vision and deep learning carry out traffic lights detection and
Classification, discloses a kind of method for one or more traffic lights that detect and classify；Application number of invention patent:
CN201610941081.4, title: count down traffic signal lamp condition detection method and the monitoring system based on the method provide
A kind of count down traffic signal lamp condition detection method based on video and the monitoring system based on the method；Patent of invention Shen
Please number: CN201711322127.5, title: intersection identifier marking based on computer vision and signal lamp Intellisense side
Method discloses a kind of intersection identifier marking based on computer vision and signal lamp Intellisense method；Utility model patent
Application number: CN201720824130.6 title: a kind of traffic lights intelligent identifying system discloses a kind of traffic lights intelligence
Energy identifying system, carries out detection identification by traffic light signal of the visual identifying system to road ahead, according to the information of detection
Can vehicle be prejudged by junction ahead, and voice prompting driver, if appropriate for continuing to move ahead, auxiliary driver safety drives
Pass through crossroad；Application number of invention patent: CN201710852496.9, a kind of title: traffic lights image processing method
And traffic lights image processing apparatus, disclose a kind of traffic lights image processing method and traffic lights image procossing
Device, to restore the color and shape of abnormal traffic signal lamp, avoid red light and amber light erroneous detection；Application number of invention patent:
CN201711157204.6, title: a kind of road traffic signal lamp situation intelligent monitor system discloses a kind of road traffic letter
Signal lamp situation intelligent monitor system, using survey number, testing the speed monitors three kinds of information collection modes to traffic information progress with signal lamp
Acquisition；Application number of invention patent: CN201711195384.7, a kind of title: identification of the traffic lights based on multi-categorizer
Method is related to a kind of recognition methods of traffic lights based on multi-categorizer；Application number of invention patent:
CN201710774141.2, title: a kind of signal lamp working condition detection system and method disclose a kind of signal lamp work shape
State detection system and method, for solving the problems, such as that signal lamp detection is easy to appear missing inspection and erroneous detection；Application number of invention patent:
CN201610940701.2, title: a kind of signal lamp duration detection method based on deep learning discloses a kind of based on depth
The signal lamp duration detection method of study, detects signal lamp duration in real time；Application number of invention patent:
CN201710284682.7, title: a kind of traffic signal light condition based reminding method and mobile terminal provide a kind of traffic signals
Lamp state based reminding method and mobile terminal, are related to field of computer technology；Application number of invention patent: CN201710245522.1,
Title: a kind of traffic lights recognizer based on convolutional network proposes that a kind of traffic lights based on convolutional network are known
Other algorithm.
In conclusion Current protocols go out together mainly for detection of signal lamp state, for red greenish-yellow any the two with bright
Frequency issues can not detect, in addition, they are all that the module integrated using signal lamp is detected, can not utilize existing equipment,
For this purpose, the present invention proposes a kind of signal lamp failure detection method based on computer vision.
Summary of the invention
Aiming at the problems existing in the prior art, the present invention provides solve the problems, such as that the detection of above-mentioned signal lamp failure exists
A kind of signal lamp failure detection method based on computer vision.
Technical scheme is as follows:
A kind of signal lamp failure detection method based on computer vision, which comprises the steps of:
Step 1): image is obtained from the monitor camera of existing traffic intersection and carries out signal lamp calibration, obtains signal lamp
Regional ensemble L={ li=(RLi,YLi,BLi,TPi) | i=1,2,3 ..., n }, wherein liIndicate i-th of signal lamp backboard, n
Indicate signal lamp backboard quantity, RLiIndicate the red signal rectangular area of i-th of signal lamp backplane region, YLiIndicate i-th of letter
The steady yellow rectangular area of signal lamp backplane region, BLiIndicate the green light signals rectangular area of i-th of signal lamp backplane region,
TPiIndicate i-th of signal lamp type, TPi∈ TPC and TPi≠ null, TPC indicate signal lamp type set and TPC=null,
Circle lamp, left-hand rotation arrow lamp, arrow lamp of keeping straight on, right-hand rotation arrow lamp }, null indicates null value；
Step 2): signal lamp state is carried out to continue sampling identification, and maintains ordered set the S={ (rs of stateij,
ysij,bsij,tpij,tj) | i=1,2 ..., n；J=1,2 ..., m }；Wherein, rsijIndicate i-th signal lamp backboard red light portion
Divide in j-th of time tjWhen state, rsij∈ { 1,0 }, rsijIndicate bright for 1,0 indicates to go out, ysijIndicate i-th of signal lamp
Backboard amber light part is in j-th of time tjWhen state, ysij∈ { 1,0 }, ysijIndicate bright for 1,0 indicates to go out, bsijIndicate the
I signal lamp backboard green light part is in j-th of time tjWhen state, bsij∈ { 1,0 }, bsijIndicate bright for 1,0 indicates to go out,
tpijIndicate i-th of the signal lamp backboard actually recognized in j-th of time tjWhen class signal type, tpij∈ TPC, tjIt indicates
J-th of acquisition time, m indicate the quantity of acquisition；
Step 3): signal lamp failure is judged according to set S, and is W by the failure logging of i-th of signal lamp backboardi, specifically
Are as follows: to each signal lamp backboard li, W is seti=null；If formula (1) meets, Wi=i-th signal lamp backboard red light brightness
It is abnormal；If formula (2) meets, Wi=i-th signal lamp type is abnormal；If any one satisfaction of formula (3), (4) or (5), Wi=the
I signal lamp on/off frequency anomaly；
Wherein, RT0Indicate the time that red light lights in a signal lamp cycle, unit is the second；BT0It indicates to believe at one
The time that green light lights in the signal lamp period, unit are the second；δ expression adjustment factor, δ ∈ (0,1.0]；RTP0It indicates in T0Period
The minimum number of interior signal lamp type mistake；I () indicates indicative function, if the parameter expression of the function is true, letter
Number return value is 1, is otherwise 0；RSB0Indicate that signal lamp backboard red light, amber light or any two in green light region light simultaneously
Permission frequency threshold value；
Step 4): if Wi≠ null, then by WiIt is sent to administrative staff；It goes to step 2).
A kind of signal lamp failure detection method based on computer vision, which is characterized in that the step 2) tool
Body are as follows:
Step 2.1): to current time tk, RL is intercepted to acquired imageiRegion, if the red component in the region is average
Gray value is less than R0, then rs is recordedik=0 and tpikOtherwise=null identifies the signal lamp type in the region simultaneously using HOG+SVM
It is denoted as tpik, and record rsik=1, tpik∈TPC；YL is intercepted to acquired imageiRegion, if the average gray in the region
Value is less than Y0, then ys is recordedik=0, otherwise, record ysik=1；BL is intercepted to acquired imageiRegion, if the region is flat
Equal gray value is less than B0, then bs is recordedik=0, otherwise, record bsik=1；Wherein, R0Indicate red light region gray threshold, Y0Table
Show amber light region gray threshold, B0Indicate green light region gray threshold；
Step 2.2): if tm-t1<T0, then by (rsik,ysik,bsik,tpik,tk) it is added to the tail portion of set S, it goes to step
2)；Otherwise, by (rsi1,ysi1,bsi1,tpi1,t1) deleted from set S, while by (rsik,ysik,bsik,tpik,tk) addition
To the tail portion of set S；Wherein, T0Indicate the minimum time section that signal lamp state persistently samples, unit is the second；
The beneficial effects of the present invention are: without utilizing present monitoring in the case where being transformed to signal lamp itself
Video camera acquired image is analyzed, and realizes the detection of signal lamp failure, has many advantages, such as real-time, convenience.
Detailed description of the invention
Fig. 1 is signal lamp region labeling schematic diagram of the present invention.
Specific implementation method
A specific embodiment of the invention is elaborated below with reference to embodiment.
A kind of signal lamp failure detection method based on computer vision, the specific steps are as follows:
Step 1): image is obtained from the monitor camera of existing traffic intersection and carries out signal lamp calibration, obtains signal lamp
Regional ensemble L={ li=(RLi,YLi,BLi,TPi) | i=1,2,3 ..., n }, wherein liIndicate i-th of signal lamp backboard, n
Indicate signal lamp backboard quantity, RLiIndicate the red signal rectangular area of i-th of signal lamp backplane region, YLiIndicate i-th of letter
The steady yellow rectangular area of signal lamp backplane region, BLiIndicate the green light signals rectangular area of i-th of signal lamp backplane region,
TPiIndicate i-th of signal lamp type, TPi∈ TPC and TPi≠ null, TPC indicate signal lamp type set and TPC=null,
Circle lamp, left-hand rotation arrow lamp, arrow lamp of keeping straight on, right-hand rotation arrow lamp }, null indicates null value；In the present embodiment, as shown in Figure 1, L
Include three signal lamp backboards: l1=(RL1,YL1,BL1, left-hand rotation arrow lamp), l2=(RL2,YL2,BL2, arrow lamp of keeping straight on),
l3=(RL3,YL3,BL3, right-hand rotation arrow lamp)；
Step 2): signal lamp state is carried out to continue sampling identification, and maintains ordered set the S={ (rs of stateij,
ysij,bsij,tpij,tj) | i=1,2 ..., n；J=1,2 ..., m }, wherein rsijIndicate i-th signal lamp backboard red light portion
Divide in j-th of time tjWhen state, rsij∈ { 1,0 }, rsijIndicate bright for 1,0 indicates to go out, ysijIndicate i-th of signal lamp
Backboard amber light part is in j-th of time tjWhen state, ysij∈ { 1,0 }, ysijIndicate bright for 1,0 indicates to go out, bsijIndicate the
I signal lamp backboard green light part is in j-th of time tjWhen state, bsij∈ { 1,0 }, bsijIndicate bright for 1,0 indicates to go out,
tpijIndicate i-th of the signal lamp backboard actually recognized in j-th of time tjWhen class signal type, tpij∈ TPC, tjIt indicates
J-th of acquisition time, m indicate the quantity of acquisition；
Specifically: step 2.1): to current time tk, RL is intercepted to acquired imageiRegion, if the red in the region
Component average gray value is less than R0, then rs is recordedik=0 and tpikOtherwise=null identifies the signal in the region using HOG+SVM
Lamp type is simultaneously denoted as tpik, and record rsik=1, tpik∈TPC；YL is intercepted to acquired imageiRegion, if the region
Average gray value is less than Y0, then ys is recordedik=0, otherwise, record ysik=1；BL is intercepted to acquired imageiRegion, if should
The average gray value in region is less than B0, then bs is recordedik=0, otherwise, record bsik=1；Wherein, R0Indicate red light area grayscale threshold
Value, Y0Indicate amber light region gray threshold, B0Indicate green light region gray threshold；In the present embodiment, R0=150, Y0=120,
B0=120；
Step 2.2): if tm-t1<T0, then by (rsik,ysik,bsik,tpik,tk) it is added to the tail portion of set S, it goes to step
2)；Otherwise, by (rsi1,ysi1,bsi1,tpi1,t1) deleted from set S, while by (rsik,ysik,bsik,tpik,tk) addition
To the tail portion of set S；Wherein, T0Indicate the minimum time section that signal lamp state persistently samples, unit is the second；In the present embodiment
In, T0=300；
Step 3): signal lamp failure is judged according to set S, and is W by the failure logging of i-th of signal lamp backboardi, specifically
Are as follows: to each signal lamp backboard li, W is seti=null；If formula (1) meets, Wi=i-th signal lamp backboard red light brightness
It is abnormal；If formula (2) meets, Wi=i-th signal lamp type is abnormal；If any one satisfaction of formula (3), (4) or (5), Wi=the
I signal lamp on/off frequency anomaly；
Wherein, RT0Indicate the time that red light lights in a signal lamp cycle, unit is the second；BT0It indicates to believe at one
The time that green light lights in the signal lamp period, unit are the second；δ expression adjustment factor, δ ∈ (0,1.0]；RTP0It indicates in T0Period
The minimum number of interior signal lamp type mistake；I () indicates indicative function, if the parameter expression of the function is true, letter
Number return value is 1, is otherwise 0；RSB0Indicate that signal lamp backboard red light, amber light or any two in green light region light simultaneously
Permission frequency threshold value；In the present embodiment, RT0=35, BT0=28, δ=0.8, RTP0=10, RSB0=30；
Step 4): if Wi≠ null, then by WiIt is sent to administrative staff；It goes to step 2).
Content described in this specification embodiment is only enumerating to the way of realization of inventive concept, protection of the invention
Range should not be construed as being limited to the specific forms stated in the embodiments, and protection scope of the present invention is also and in this field skill
Art personnel conceive according to the present invention it is conceivable that equivalent technologies mean.
Claims (2)
1. a kind of signal lamp failure detection method based on computer vision, which comprises the steps of:
Step 1): image is obtained from the monitor camera of existing traffic intersection and carries out signal lamp calibration, obtains signal lamp region
Set L={ li=(RLi,YLi,BLi,TPi) | i=1,2,3 ..., n }, wherein liIndicate that i-th of signal lamp backboard, n indicate letter
Signal lamp backboard quantity, RLiIndicate the red signal rectangular area of i-th of signal lamp backplane region, YLiIndicate i-th of signal lamp back
The steady yellow rectangular area in plate region, BLiIndicate the green light signals rectangular area of i-th of signal lamp backplane region, TPiIt indicates
I-th of signal lamp type, TPi∈ TPC and TPi≠ null, TPC indicate signal lamp type set and { null, circle lamp are left by TPC=
Turn arrow lamp, arrow lamp of keeping straight on, right-hand rotation arrow lamp }, null indicates null value；
Step 2): signal lamp state is carried out to continue sampling identification, and maintains ordered set the S={ (rs of stateij,ysij,
bsij,tpij,tj) | i=1,2 ..., n；J=1,2 ..., m }；Wherein, rsijIndicate i-th of signal lamp backboard red light part
J time tjWhen state, rsij∈ { 1,0 }, rsijIndicate bright for 1,0 indicates to go out, ysijIndicate i-th of signal lamp backboard amber light
Part is in j-th of time tjWhen state, ysij∈ { 1,0 }, ysijIndicate bright for 1,0 indicates to go out, bsijIndicate i-th of signal lamp
Backboard green light part is in j-th of time tjWhen state, bsij∈ { 1,0 }, bsijIndicate bright for 1,0 indicates to go out, tpijIndicate real
I-th of signal lamp backboard that border recognizes is in j-th of time tjWhen class signal type, tpij∈ TPC, tjIndicate j-th of acquisition
Time, m indicate the quantity of acquisition；
Step 3): signal lamp failure is judged according to set S, and is W by the failure logging of i-th of signal lamp backboardi, specifically: it is right
Each signal lamp backboard li, W is seti=null；If formula (1) meets, Wi=i-th signal lamp backboard red light brightness is abnormal；If
Formula (2) meets, then Wi=i-th signal lamp type is abnormal；If any one satisfaction of formula (3), (4) or (5), Wi=i-th signal
Lamp on/off frequency anomaly；
Wherein, RT0Indicate the time that red light lights in a signal lamp cycle, unit is the second；BT0It indicates in a signal lamp
The time that green light lights in period, unit are the second；δ expression adjustment factor, δ ∈ (0,1.0]；RTP0It indicates in T0Letter in period
The minimum number of signal lamp type mistake；I () indicates indicative function, if the parameter expression of the function is true, function is returned
Returning value is 1, is otherwise 0；RSB0Indicate the permission that signal lamp backboard red light, amber light or any two in green light region light simultaneously
Frequency threshold value；
Step 4): if Wi≠ null, then by WiIt is sent to administrative staff；It goes to step 2).
2. a kind of signal lamp failure detection method based on computer vision according to claim 1, which is characterized in that institute
It states in step 2) specifically:
Step 2.1): to current time tk, RL is intercepted to acquired imageiRegion, if the red component average gray in the region
Value is less than R0, then rs is recordedik=0 and tpikOtherwise=null identifies the signal lamp type in the region using HOG+SVM and is denoted as
tpik, and record rsik=1, tpik∈TPC；YL is intercepted to acquired imageiRegion, if the average gray value in the region is less than
Y0, then ys is recordedik=0, otherwise, record ysik=1；BL is intercepted to acquired imageiRegion, if the average gray in the region
Value is less than B0, then bs is recordedik=0, otherwise, record bsik=1；Wherein, R0Indicate red light region gray threshold, Y0Indicate amber light
Area grayscale threshold value, B0Indicate green light region gray threshold；
Step 2.2): if tm-t1<T0, then by (rsik,ysik,bsik,tpik,tk) it is added to the tail portion of set S, it goes to step 2)；It is no
Then, by (rsi1,ysi1,bsi1,tpi1,t1) deleted from set S, while by (rsik,ysik,bsik,tpik,tk) it is added to set
The tail portion of S；Wherein, T0Indicate the minimum time section that signal lamp state persistently samples, unit is the second.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811334431.6A CN109740412A (en) | 2018-11-09 | 2018-11-09 | A kind of signal lamp failure detection method based on computer vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811334431.6A CN109740412A (en) | 2018-11-09 | 2018-11-09 | A kind of signal lamp failure detection method based on computer vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109740412A true CN109740412A (en) | 2019-05-10 |
Family
ID=66355557
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811334431.6A Pending CN109740412A (en) | 2018-11-09 | 2018-11-09 | A kind of signal lamp failure detection method based on computer vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109740412A (en) |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533940A (en) * | 2019-08-15 | 2019-12-03 | 北京百度网讯科技有限公司 | Method, apparatus, equipment and the computer storage medium of abnormal traffic signal lamp identification |
| CN110749601A (en) * | 2019-10-24 | 2020-02-04 | 中国民用航空总局第二研究所 | Airport runway lamp detection system and method based on images |
| 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 |
| CN111681442A (en) * | 2020-06-09 | 2020-09-18 | 贾若然 | Signal lamp fault detection device based on image classification algorithm |
| 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 |
| WO2021129611A1 (en) * | 2019-12-24 | 2021-07-01 | 上海高德威智能交通***有限公司 | Monitoring scenario detection method and apparatus, and electronic device |
| CN113129591A (en) * | 2021-04-13 | 2021-07-16 | 江苏智通交通科技有限公司 | Traffic signal lamp fault detection method based on deep learning target detection |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202720785U (en) * | 2012-04-12 | 2013-02-06 | 中国计量学院 | Fault detection apparatus for LED traffic lights |
| CN105913041A (en) * | 2016-04-27 | 2016-08-31 | 浙江工业大学 | Pre-marked signal lights based identification method |
| CN106530772A (en) * | 2016-10-14 | 2017-03-22 | 深圳尚桥信息技术有限公司 | Intelligent traffic signal lamp, control system of the same and emergency control method of the same |
| CN108320556A (en) * | 2018-03-30 | 2018-07-24 | 无锡智高点技术研发有限公司 | A kind of traffic signal light fault detection method and system |
| US20180281802A1 (en) * | 2017-03-31 | 2018-10-04 | Subaru Corporation | Traveling control system for vehicle |
-
2018
- 2018-11-09 CN CN201811334431.6A patent/CN109740412A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN202720785U (en) * | 2012-04-12 | 2013-02-06 | 中国计量学院 | Fault detection apparatus for LED traffic lights |
| CN105913041A (en) * | 2016-04-27 | 2016-08-31 | 浙江工业大学 | Pre-marked signal lights based identification method |
| CN106530772A (en) * | 2016-10-14 | 2017-03-22 | 深圳尚桥信息技术有限公司 | Intelligent traffic signal lamp, control system of the same and emergency control method of the same |
| US20180281802A1 (en) * | 2017-03-31 | 2018-10-04 | Subaru Corporation | Traveling control system for vehicle |
| CN108320556A (en) * | 2018-03-30 | 2018-07-24 | 无锡智高点技术研发有限公司 | A kind of traffic signal light fault detection method and system |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110533940A (en) * | 2019-08-15 | 2019-12-03 | 北京百度网讯科技有限公司 | Method, apparatus, equipment and the computer storage medium of abnormal traffic signal lamp identification |
| CN110533940B (en) * | 2019-08-15 | 2022-06-28 | 北京百度网讯科技有限公司 | Method, device and equipment for identifying abnormal traffic signal lamp in automatic driving |
| CN110749601A (en) * | 2019-10-24 | 2020-02-04 | 中国民用航空总局第二研究所 | Airport runway lamp detection system and method based on images |
| 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 |
| WO2021129611A1 (en) * | 2019-12-24 | 2021-07-01 | 上海高德威智能交通***有限公司 | Monitoring scenario detection method and apparatus, and electronic device |
| CN111681442A (en) * | 2020-06-09 | 2020-09-18 | 贾若然 | Signal lamp fault detection device based on image classification algorithm |
| 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 |
| CN113129591A (en) * | 2021-04-13 | 2021-07-16 | 江苏智通交通科技有限公司 | Traffic signal lamp fault detection method based on deep learning target detection |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109740412A (en) | A kind of signal lamp failure detection method based on computer vision | |
| CN110197589B (en) | Deep learning-based red light violation detection method | |
| CN110136447B (en) | Method for detecting lane change of driving and identifying illegal lane change | |
| CN105744232B (en) | A kind of method of the transmission line of electricity video external force damage prevention of Behavior-based control analytical technology | |
| CN104751634B (en) | The integrated application method of freeway tunnel driving image acquisition information | |
| CN105913041B (en) | It is a kind of based on the signal lamp recognition methods demarcated in advance | |
| TWI430212B (en) | Abnormal behavior detection system and method using automatic classification of multiple features | |
| CN105185118B (en) | A kind of vehicles peccancy detection means and its method based on video detection | |
| CN104809887B (en) | A kind of retrograde detection method of vehicle on expressway and autoalarm | |
| CN104537360B (en) | Vehicle does not give way peccancy detection method and its detecting system | |
| CN108537154A (en) | Transmission line of electricity Bird's Nest recognition methods based on HOG features and machine learning | |
| CN104992449A (en) | Information identification and surface defect on-line detection method based on machine visual sense | |
| CN109817013A (en) | Parking stall state identification method and device based on video flowing | |
| CN106096504A (en) | A kind of model recognizing method based on unmanned aerial vehicle onboard platform | |
| CN111325048B (en) | Personnel gathering detection method and device | |
| CN103065519A (en) | Detecting system of ramp fixed point stopping and starting and detecting method of ramp fixed point stopping and starting | |
| CN105844245A (en) | Fake face detecting method and system for realizing same | |
| CN111626275B (en) | Abnormal parking detection method based on intelligent video analysis | |
| CN110127479A (en) | A kind of elevator door switch method for detecting abnormality based on video analysis | |
| CN107909601A (en) | A kind of shipping anti-collision early warning video detection system and detection method suitable for navigation mark | |
| CN103500330A (en) | Semi-supervised human detection method based on multi-sensor and multi-feature fusion | |
| CN108447305A (en) | A kind of new friendship rule are lower to give precedence to pedestrian's method for early warning and early warning system | |
| CN112149595A (en) | Method for detecting lane line and vehicle violation by using unmanned aerial vehicle | |
| CN105957341A (en) | Wide area traffic jam detection method based on unmanned plane airborne platform | |
| CN201037962Y (en) | Electronic pile examining system |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 323000 block B, building 2, No. 268, Shiniu Road, nanmingshan street, Liandu District, Lishui City, Zhejiang Province Applicant after: ZHEJIANG HAOTENG ELECTRON TECHNOLOGY Co.,Ltd. Address before: 323000 Room 201, building 12, Tianning incubation base, Lvgu Information Industrial Park, Lishui City, Zhejiang Province Applicant before: ZHEJIANG HAOTENG ELECTRON TECHNOLOGY Co.,Ltd. |
|
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190510 |