CN115472028A - Intelligent early warning induction method and system for tunnel emergency stop zone - Google Patents
Intelligent early warning induction method and system for tunnel emergency stop zone Download PDFInfo
- Publication number
- CN115472028A CN115472028A CN202211034857.6A CN202211034857A CN115472028A CN 115472028 A CN115472028 A CN 115472028A CN 202211034857 A CN202211034857 A CN 202211034857A CN 115472028 A CN115472028 A CN 115472028A
- Authority
- CN
- China
- Prior art keywords
- parking
- vehicle
- module
- tunnel
- early warning
- 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
- 230000006698 induction Effects 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 18
- 238000001514 detection method Methods 0.000 claims abstract description 38
- 230000000007 visual effect Effects 0.000 claims abstract description 12
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 10
- 238000007635 classification algorithm Methods 0.000 claims description 7
- 238000004891 communication Methods 0.000 claims description 7
- 230000005764 inhibitory process Effects 0.000 claims description 7
- 230000002708 enhancing effect Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 239000013598 vector Substances 0.000 claims description 6
- 230000003287 optical effect Effects 0.000 claims description 5
- 238000004880 explosion Methods 0.000 claims description 4
- 230000003993 interaction Effects 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000003064 k means clustering Methods 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims description 3
- 238000005070 sampling Methods 0.000 claims description 3
- 238000012706 support-vector machine Methods 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000013139 quantization Methods 0.000 claims description 2
- 230000001939 inductive effect Effects 0.000 claims 2
- 230000009286 beneficial effect Effects 0.000 abstract description 2
- 230000006399 behavior Effects 0.000 description 14
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000000391 smoking effect Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 125000001475 halogen functional group Chemical group 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000009471 action Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000036541 health Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 210000001503 joint Anatomy 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 210000003205 muscle Anatomy 0.000 description 1
- 230000001629 suppression Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
- E21F17/00—Methods or devices for use in mines or tunnels, not covered elsewhere
-
- E—FIXED CONSTRUCTIONS
- E21—EARTH OR ROCK DRILLING; MINING
- E21F—SAFETY DEVICES, TRANSPORT, FILLING-UP, RESCUE, VENTILATION, OR DRAINING IN OR OF MINES OR TUNNELS
- E21F17/00—Methods or devices for use in mines or tunnels, not covered elsewhere
- E21F17/18—Special adaptations of signalling or alarm devices
-
- 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
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
- G08G1/166—Anti-collision systems for active traffic, e.g. moving vehicles, pedestrians, bikes
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04M—TELEPHONIC COMMUNICATION
- H04M11/00—Telephonic communication systems specially adapted for combination with other electrical systems
- H04M11/04—Telephonic communication systems specially adapted for combination with other electrical systems with alarm systems, e.g. fire, police or burglar alarm systems
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N7/00—Television systems
- H04N7/14—Systems for two-way working
- H04N7/141—Systems for two-way working between two video terminals, e.g. videophone
Landscapes
- Engineering & Computer Science (AREA)
- Mining & Mineral Resources (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Signal Processing (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Geochemistry & Mineralogy (AREA)
- Geology (AREA)
- Multimedia (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention discloses an intelligent early warning induction method and system for a tunnel emergency stop zone, which comprises the following steps: the system comprises a computing terminal, a parking belt, a display module, a voice reminding module, a rear-view detection module, an event detection module, a figure identification module, a figure analysis module, an emergency telephone, a visual module, a tunnel broadcast, an indicator lamp, a flashing induction mark and a flashing lamp belt. The invention has the beneficial effects that: the vehicle early warning system can give an early warning to the rear driving vehicle after the vehicle drives into the tunnel parking zone, can display the road condition of the rear in real time, and effectively induces the vehicle to safely drive away from the tunnel parking zone.
Description
Technical Field
The invention relates to the technical field of tunnel safety management, in particular to an intelligent early warning induction method and system for a tunnel emergency stop zone.
Background
The tunnel parking area is a temporary parking area used for emergency parking of vehicles or other reasons when the vehicles are in failure temporarily, and the tunnel parking area always has the situation that the vehicles occupy the parking area illegally, so that great hidden danger is brought to safe passing of the tunnel vehicles. The management and control of the tunnel parking zone need to be strengthened, and the emergency handling capacity of the tunnel is improved.
In the prior art, by detecting that a vehicle enters and reminding relevant departments to carry out illegal parking treatment, a monitor can also know the field situation through a monitoring picture, and carry out remote guidance work or arrange the field treatment of inspection personnel aiming at a fault or accident vehicle. However, the safety driving of the rear vehicle is possibly influenced due to insufficient text or voice warning information, the problem of how to safely drive the vehicle away also exists, a driver can only check the rear road condition through a rearview mirror, the condition of the rear vehicle cannot be accurately mastered due to the viewing angle, and the collision accident with the rear vehicle is still easily caused during driving away.
For example, a "one-way shield tunnel structure with emergency temporary parking hole" disclosed in chinese patent literature has a publication number: CN109989766a, filing date: in 29/04 in 2019, the one-way shield tunnel structure with the emergency temporary parking hole provided by the invention has the advantages that the emergency temporary parking hole or the parking bay is arranged by locally opening the hole at the right side of the shield tunnel in the driving direction, so that the diameter of the main line tunnel shield is kept unchanged, the engineering risk is reduced, and the engineering investment is saved, but the problems that the early warning cannot be carried out on a vehicle driving behind after the vehicle drives into the parking zone of the tunnel, the road condition at the rear cannot be displayed in real time, and the vehicle cannot be effectively induced to safely drive away from the parking zone of the tunnel exist.
Disclosure of Invention
The invention provides an intelligent early warning induction method and system for a tunnel emergency stop zone, aiming at the defects that the prior art can not early warn a vehicle running behind after the vehicle enters the tunnel stop zone, can not display the road condition behind in real time and can not effectively induce the vehicle to safely leave the tunnel stop zone.
The technical scheme of the invention is as follows, and an intelligent early warning induction system for a tunnel emergency stop zone comprises:
the computing terminal is used for controlling data interaction and linkage control;
a parking belt for standardizing a parking position;
the display module is used for playing a rear road condition video in real time and connecting the display module with the computing terminal;
the voice reminding module is used for issuing voice reminding information and connecting the voice reminding information with the computing terminal;
the rear-view detection module is used for shooting real-time road conditions at the rear and connecting the computing terminal with the display module;
the event detection module is used for identifying a parking event and vehicle information of a vehicle and is connected with the computing terminal;
the figure identification module is used for acquiring the behavior and the state of a figure and connecting the figure identification module with a computing terminal;
the figure analysis module is used for analyzing the behavior and the state of a figure and connecting the figure identification module with the computing terminal;
the emergency telephone is used for carrying out telephone communication with the vehicle owner and connecting the vehicle owner with the computing terminal;
the visible module is used for carrying out video communication with the vehicle owner and connecting with the computing terminal;
the tunnel broadcast is used for broadcasting information to the car owner and connecting the calculation terminal;
the indicating lamp is used for displaying the parking state of the parking belt and is connected with the computing terminal;
the explosion and flash induction mark is used for enhancing the identification degree of a parking belt and a traffic lane and is connected with a computing terminal;
and the flashing lamp strip is used for enhancing the identification degree of the parking strip and the traffic lane and connecting the computing terminal with the parking strip.
In the scheme, after the fact that the tunnel parking has the vehicle to drive in is found, the parking event can be reported in time, early warning of the vehicle driving behind is started, the voice reminding module is used for dissuading the vehicle from violating parking, the tunnel management station is connected with the field vehicle through the emergency telephone and the visual module, the site handling and rescue of the fault vehicle are remotely guided, the rear road condition is displayed in real time through the rear-view detection module and the display module, and the vehicle can be effectively induced to safely drive away from the parking area.
Preferably, the display module displays license plate information, persuasion information and countdown,
wherein the countdown includes an persuasion countdown and a violation countdown.
In the scheme, for the vehicles without the white list license plates, the dissuading countdown is started first, and after the countdown is finished, if the vehicles still do not drive away, the illegal stopping countdown is started.
Preferably, the event detection module is a camera; the rearview detection module is a highlight inhibition camera, and the camera shooting direction is aligned to the direction of the vehicle head; the figure recognition module is a high-precision camera, and the camera shooting direction is aligned to the direction of the vehicle head.
In the scheme, the camera of the event detection module identifies the parking event and the vehicle information of the vehicle. When the highlight inhibition camera of the rearview detection module is installed, the direction of a vehicle head needs to be aligned, the imaging effect of the rearview camera is considered, because the light of a headlamp of a vehicle running in a tunnel influences a shot video, the halo is large, the video image is fuzzy, and when the highlight inhibition camera is used, the highlight inhibition camera is started, a highlight point is automatically distinguished, and an area near the highlight point is compensated to obtain a clearer image. The high-precision camera of the person identification module is used for acquiring the behaviors and states of persons, and the high-precision camera can improve the image quality so as to improve the target tracking and classifying precision.
Preferably, the indicator light, the flashing induction mark and the flashing light strip are in two states, and correspond to parked vehicles and non-parked vehicles in the parking zone respectively; the emergency telephone and the visual module are provided with an operation scheme for rapidly contacting the tunnel management station.
In this scheme, the pilot lamp is used for showing parking area parking state, and the pilot lamp state is green lamp, and the parking is in-band not parkked, and the pilot lamp state is the red lamp, and the parking has been in-band parked, reminds the rear vehicle this parking area to be occupied. The explosion and flash induction mark is in a closed state when the parking zone does not park, and is in an open state when the parking zone parks, and the induction mark flickers. When exploding the sudden strain of a muscle lamp area and not parkking in the parking area, be in the off-state, when parking in-band has already parkked, be in the on-state, the lamp area twinkles. The states of the indicator light, the flashing induction mark and the flashing light strip are controlled by the computing terminal. The difference between the parked state and the non-parked state in the parking line is improved, and the condition that a plurality of vehicles enter the parking line is avoided.
Preferably, the computing terminal adopts a GPU + CPU heterogeneous computing architecture.
In the scheme, the GPU + CPU heterogeneous computing architecture is adopted to improve the intellectualization and the computing speed of the system.
Preferably, the intelligent early warning induction method for the emergency stop zone of the tunnel comprises the following steps:
s1: judging whether a vehicle is parked in the parking line or not based on the event detection module, if so, acquiring license plate information by the event detection module;
s2: judging whether the vehicle is a white list vehicle or not based on the license plate information, if so, recording parking records, and if not, performing step S3;
s3: carrying out dissuading countdown, judging whether the vehicle is still parked in the stop line after dissuading countdown is finished, if so, carrying out step S4, and if not, recording the stop record;
s4: judging whether the vehicle is parked in an accident or a fault vehicle, if so, remotely guiding or processing the vehicle on site by a worker, and recording parking records, otherwise, performing the step S5;
s5: and carrying out illegal parking countdown, judging whether the vehicle is still parked in the parking line after the illegal parking countdown is finished, if so, carrying out field processing and recording parking by a worker, and if not, recording the parking.
In the scheme, a white list license plate is provided, special vehicles and social vehicles are effectively distinguished, different processing flows are provided, vehicles violating parking are dissuaded through a voice reminding module and dissuaded from countdown is started, whether the vehicles are in accidents or not is judged if the vehicles are not driven away after dissuading from countdown is finished, the vehicles violating parking countdown is started for the non-accident vehicles, and parking records are recorded when all the vehicles are driven away by workers for field processing.
Preferably, when the vehicles without the white list license plates drive into the parking zone and do not drive away when the vehicle exceeds the persuasion countdown, the voice reminding module plays persuasion voice; when a vehicle stops in a parking area, the display module plays a rear road condition video in real time, and the display module displays an auxiliary line of 50 meters, an auxiliary line of 100 meters and an auxiliary line of 200 meters.
In the scheme, the voice reminding module dissuades the vehicles without the white list license plates, and the display module plays the rear road condition video in real time, so that the safety of the vehicles driving away from the parking belt is improved.
Preferably, steps S3 and S4 further include: the position of the car owner is accurately tracked through a target tracking algorithm, and the state and the behavior of the car owner are identified through a classification algorithm.
In this scheme, cross the accurate car owner position of pursuit of target tracking algorithm, through categorised algorithm discernment car owner state and action, judge whether the car owner has the civilized condition such as smoking, lost article, perhaps have the sick and wounded condition such as driver fatigue, health anomaly, carry out video recording to the civilized condition, assist the sick and wounded condition.
Preferably, the target tracking algorithm comprises the steps of:
s311: acquiring a video acquired by a person identification module;
s312: selecting an initial frame, taking the driving position as an initial tracking frame, and generating a plurality of candidate tracking frames in a next frame;
s313: extracting the characteristics of the candidate tracking frame, and grading the candidate tracking frame;
s314: fusing a plurality of candidate tracking frames with high scores to obtain an optimal candidate tracking frame;
the candidate tracking frame is generated based on a particle filtering or sliding window method, the characteristics of the candidate tracking frame are depth characteristics learned by a large number of training samples, and the optimal candidate tracking frame is obtained through all prediction weighted averages.
Preferably, the classification algorithm comprises the steps of:
s321: acquiring a video sequence, respectively sampling feature points on a plurality of scales of a picture by adopting grid division, and filtering partial feature points;
s322: calculating the movement speed of the characteristic points and tracking the key points;
s323: extracting a direction gradient histogram, an optical flow histogram, a motion boundary histogram and track features along a track;
s324: extracting features from the training set data, and clustering the features by using a K-means clustering algorithm to obtain a feature dictionary;
s325: carrying out quantitative coding on the test data by using the feature dictionary to obtain a vector with a fixed length;
s326: classifying the feature vectors after coding quantization by using a support vector machine;
the partial feature points comprise feature points with less transformation, and the motion speed of the feature points is obtained through optical flow median calculation in the neighborhood of the feature points.
The invention has the beneficial effects that: after the fact that the vehicles drive into the tunnel parking area is found, parking events can be reported in time, early warning of vehicles driving behind is started, the vehicles driving behind are dissuaded from the tunnel management station through the voice reminding module, the tunnel management station is in contact with field vehicles through the emergency telephone and the visual module, the site handling and rescue of fault vehicles are remotely guided, the rear road condition is displayed in real time through the rear-view detection module and the display module, and the vehicles can be effectively induced to safely drive away from the parking area.
Drawings
Fig. 1 is a schematic structural diagram of an intelligent early warning induction system for a tunnel emergency stop zone according to the present invention.
Fig. 2 is a flowchart of an intelligent early warning induction method for a tunnel emergency stop zone according to the present invention.
Fig. 3 is a structural layout diagram of an intelligent early warning guidance system for a tunnel emergency stop zone according to the present invention.
In the figure 1, a computing terminal; 2. a parking belt; 3. a display module; 4. a voice reminding module; 5. a rear view detection module; 6. an event detection module; 7. a person identification module; 8. a person analysis module; 9. an emergency call; 10. a visual module; 11. broadcasting in a tunnel; 12. an indicator light; 13. a flashing induction mark; 14. explode and dodge lamp area.
Detailed Description
The technical scheme of the invention is further specifically described by the following embodiments and the accompanying drawings.
Example (b): as shown in fig. 1, an intelligent early warning induction system for tunnel emergency stop zone includes:
the computing terminal 1 is used for controlling data interaction and linkage control;
a parking band 2 for standardizing a parking position;
the display module 3 is used for playing a rear road condition video in real time and connecting the rear road condition video with the computing terminal 1;
the voice reminding module 4 is used for issuing voice reminding information and connecting the computing terminal 1;
the rear-view detection module 5 is used for shooting rear real-time road conditions and is connected with the computing terminal 1 and the display module 3;
the event detection module 6 is used for identifying a parking event and vehicle information of a vehicle and is connected with the computing terminal 1;
the person identification module 7 is used for acquiring the behavior and the state of a person and connecting the person identification module with the computing terminal 1;
the person analysis module 8 is used for analyzing the behavior and the state of a person and connecting the person identification module 7 with the computing terminal 1;
the emergency telephone 9 is used for carrying out telephone communication with the vehicle owner and connecting the computing terminal 1;
the visual module 10 is used for carrying out video communication with the vehicle owner and connecting the computing terminal 1;
the tunnel broadcast 11 is used for broadcasting information to the car owner and connecting the computing terminal 1;
the indicator lamp 12 is used for displaying the parking state of the parking belt 2 and is connected with the computing terminal 1;
the explosion and flash induction mark 13 is used for enhancing the identification degree of the parking belt 2 and the traffic lane and is connected with the computing terminal 1;
and the flashing lamp strip 14 is used for enhancing the identification degree of the parking strip 2 and the traffic lane and connecting the computing terminal 1 with the parking strip 2.
The state information of the display module 3, the voice reminding module 4, the rear-view detection module 5, the character recognition module 7, the flashing induction mark 13, the flashing lamp strip 14, the event detection module 6, the emergency telephone 9, the visual module 10, the tunnel broadcast 11, the indicator lamp 12 and other equipment is subjected to data interaction and linkage control through the computing terminal 1 and is in butt joint with a tunnel management station platform through a network.
The display module 3 can be an LED full-color induction screen, and a P3 full-color LED screen of 2 meters by 1 meter can be adopted in implementation in consideration of the display effect of the LED full-color induction screen. Display module 3 shows license plate information, dissuades information and countdown, and wherein, the countdown is including dissuading from the countdown and the count down of violating the stop, dissuades from the countdown and the count down of violating the stop can self-defined count down time.
The event detection module 6 is used for identifying parking events and vehicle information of vehicle access, and comprises detection that a plurality of vehicles simultaneously occupy the parking belt 2. The parking event information includes time, place, license plate, photo, video, etc. The event detection module 6 can be a camera, and in consideration of the accuracy rate of identifying the vehicle entering and the vehicle leaving by the camera, an electronic fence technology needs to be adopted to perform region division on the parking belt 2, so that the probability of false alarm and missing alarm is reduced. If a vehicle stays in the current parking zone 2 and a second vehicle enters, the vehicle can be detected and the alarm is reported to the platform. When a plurality of vehicles respectively drive away, the drive-away information of each vehicle can be detected.
The vehicle information comprises license plate information, a white list license plate of a tunnel management station system is matched according to the license plate information, the white list license plate effectively distinguishes special vehicles and social vehicles, and different processing flows are provided. The white list license plate comprises license plates of construction operation vehicles, inspection personnel vehicles and the like, when the vehicles of the white list license plate enter the parking belt 2, only parking records are recorded, and when the vehicles of the non-white list license plate drive into the parking belt 2 and do not drive away when the driving-away time exceeds the illegal parking and discouraging time, the display module 3 and the voice reminding module 4 remind the vehicle owner to drive away from the parking belt 2. The monitoring personnel can report the illegal evidence-taking materials to relevant departments for punishment.
When the event detection module 6 detects that the parking belt 2 is occupied, the linkage display module 3 and the voice reminding module 4 remind a rear vehicle, and the front parking belt 2 has the vehicle to pay attention to driving safety.
The voice reminding module 4 can be used for dissuading from a loudspeaker, when the vehicles of the license plates of the non-white list drive into the parking belt 2 and do not drive away when the dissuading time is exceeded, the voice reminding module 4 plays dissuading voice.
The rearview detection module 5 can be a rearview camera, needs to be aligned to the direction of the vehicle head during installation, takes the imaging effect of the rearview camera into consideration, and because the light of the headlights of the vehicles running in the tunnel can affect the shot video, the halo is large and the video image is fuzzy, a strong light inhibition camera needs to be adopted during implementation, the strong light inhibition camera is started, the strong light spot is automatically distinguished, and the area near the strong light spot is compensated to obtain a clearer image.
When a vehicle stops at the parking belt 2, the display module 3 plays a rear road condition video in real time, and the display module 3 displays a 50-meter auxiliary line, a 100-meter auxiliary line and a 200-meter auxiliary line to help a driver judge the distance of the rear vehicle, so that the driver can drive away according to the conditions of the display module 3 and a rearview mirror, and the probability of accidents caused by collision with the rear vehicle is reduced. The large-scale electronic rearview mirror formed by the rearview detection module 5 and the display module 3 can provide comprehensive and reliable back road condition information for drivers, and ensures effective and safe driving away from the parking belt 2.
The person identification module 7 is used for acquiring the behavior and state of a person, can be a high-precision camera, and needs to be aligned with the direction of the vehicle head during installation. The character analysis module 8 is used for analyzing the behavior and state of characters, the character analysis module 8 adopts a target tracking technology to collect character information, and the target tracking algorithm is as follows: acquiring a video acquired by a person identification module 7; in a certain frame when the car door is not opened, a driving position is taken as an initial tracking frame, a plurality of candidate tracking frames are generated in the next frame, and a particle filtering or sliding window method is adopted; extracting the features of the candidate tracking frames, wherein the features are depth features of a large number of training samples after learning, and scoring the candidate tracking frames; and fusing the candidate tracking frames with high scores to obtain an optimal candidate tracking frame, and performing weighted average on all predictions to obtain the optimal candidate tracking frame. The behavior of the person is analyzed, the behavior is used for accurately identifying the situations of fatigue driving, smoking, lost objects and the like, and the classification algorithm is as follows: acquiring a video sequence, respectively sampling feature points on a plurality of scales of a picture by adopting grid division, and filtering out points with less transformation; calculating the motion speed of the feature points so as to track the key points, wherein the motion speed of the feature points is obtained by calculating the light stream median in the neighborhood of the feature points; extracting a direction gradient histogram, an optical flow histogram, a motion boundary histogram and track features along a track; extracting features from the training set data, and clustering the features by using a K-means clustering algorithm to obtain a feature dictionary; carrying out quantitative coding on the test data by using the feature dictionary words to obtain vectors with fixed length; and classifying the coded and quantized feature vectors by using a support vector machine. The state and the behavior of the vehicle owner are obtained through a classification algorithm, bad behaviors such as smoking, throwing, fighting and the like are prominently recorded, and if the vehicle owner is in states such as pain, lying on the ground and the like, a worker is assigned to support remotely or on site in time.
The indicator light 12 is used for showing 2 parking states in the parking area, and the state of the indicator light 12 is green, and the parking area 2 is not parked, and the state of the indicator light 12 is red, and the parking area 2 is parked, and the rear vehicle is reminded that the parking area 2 is occupied. The state of the indicator light 12 is controlled by the computing terminal 1.
The computing terminal 1 adopts a GPU + CPU heterogeneous computing architecture, and in consideration of high system intelligence and improvement of the computing speed of the system, the computing terminal 1 needs to adopt a GPU + CPU scheme.
When the owner of the vehicle needs to call for help, contact can be made with the tunnel management station through the emergency telephone 9 and the visual module 10. The emergency telephone 9 is attached with a telephone of a tunnel management station, and can contact the tunnel management station by dialing, and also can contact the tunnel management station by pressing the number key 0 of the emergency telephone 9. The visual module 10 is provided with a function button for contacting the tunnel management station, the function button is clicked to apply for contacting the tunnel management station, and the tunnel management station establishes video communication after application.
The guidance sign 13 is in a closed state when the parking belt 2 is not parked, and is in an open state when the parking belt 2 is parked, and the guidance sign flickers. The state of the flash induction mark 13 is controlled by the computing terminal 1.
When not parkking in the area of stopping 2, the lamp area 14 that explodes flashes is in the off-state, when stopping in the area of stopping 2, is in the on-state, and the lamp area glimmers. The state of the flashing light strip 14 is controlled by the computing terminal 1.
When the vehicle resides in the parking belt 2, the event detection module 6 identifies the license plate of the vehicle and displays the license plate on the display module 3, the license plate is compared with the white list license plate, whether the vehicle is the vehicle with the white list license plate is judged, if not, the display module 3 also displays dissuasion information and dissuasion countdown, and if the dissuasion countdown is finished, the vehicle does not drive away, and a parking event is reported to the tunnel management station platform. The staff contact the car owner through the emergency telephone 9 or the tunnel broadcast 11 to confirm the reason of parking, if the car is an accident or a fault car, the staff is remotely guided to assist or send the staff to the site for disposal, if the car is illegally parked, the illegal parking countdown is started, the car which is not driven away after the timeout is reported to the related department, and the staff is informed to the site to persuade.
The computing terminal 1 is connected with a tunnel management station platform and a related department platform. After the vehicle leaves, the computing terminal 1 needs to locally cache information including time, place, license plate, parking duration, snapshot picture, whole parking process video and the like of the parking record and upload the information to the tunnel management station platform for recording. And the violation records are reported to relevant department platforms, so that the relevant departments can conveniently obtain evidence, and the occurrence of illegal emergency lane occupation events is reduced.
As shown in fig. 2, an intelligent early warning induction method for a tunnel emergency stop zone includes the following steps:
s1: judging whether a vehicle is parked in the parking line or not based on the event detection module 6, if so, acquiring license plate information by the event detection module 6;
s2: judging whether the vehicle is a white list vehicle or not based on the license plate information, if so, recording parking records, and if not, performing step S3;
s3: performing a dissuasion countdown, judging whether the vehicle is still parked in the stop line after the dissuasion countdown is finished, if so, performing a step S4, and if not, recording the stop record;
s4: judging whether the vehicle is parked in an accident or a fault vehicle, if so, remotely guiding or processing the vehicle on site by a worker, and recording parking records, otherwise, performing the step S5;
s5: and carrying out illegal parking countdown, judging whether the vehicle is still parked in the parking line after the illegal parking countdown is finished, if so, carrying out field processing and recording parking by a worker, and if not, recording the parking.
In steps S3 and S4, further comprising: detecting the main behavior of the vehicle, accurately tracking the position of the vehicle owner through a target tracking algorithm, and identifying the state and the behavior of the vehicle owner through a classification algorithm.
By utilizing a video AI detection algorithm, each event and vehicle information of vehicle access can be identified, if a vehicle stops in the current parking zone 2 and a second vehicle enters, the second vehicle can also be detected and reported to the platform for warning. When a plurality of vehicles respectively drive away, the drive-away information of each vehicle can be detected. The electronic rearview mirror has a safe driving-away induction function, and can provide comprehensive and reliable rear road condition information for drivers through a large-scale electronic rearview mirror consisting of the rearview detection module 5 and the display module 3, so that effective and safe driving-away from the parking belt 2 is ensured. An automatic strong light suppression algorithm is added for the rearview detection module 5, so that a strong light spot is automatically distinguished, and the area near the strong light spot is compensated to obtain a clearer image. And providing a white list license plate, effectively distinguishing special vehicles and social vehicles, and providing different processing flows. And the process of reporting violation records to relevant department platforms is provided, and violation evidence collection is facilitated.
Claims (10)
1. The utility model provides a tunnel emergency stop takes intelligent early warning induction system which characterized in that includes:
the computing terminal is used for controlling data interaction and linkage control;
a parking belt for standardizing a parking position;
the display module is used for playing a rear road condition video in real time and connecting the display module with the computing terminal;
the voice reminding module is used for issuing voice reminding information and connecting the voice reminding information with the computing terminal;
the rear-view detection module is used for shooting real-time road conditions at the rear and connecting the computing terminal with the display module;
the event detection module is used for identifying a parking event and vehicle information of a vehicle and is connected with the computing terminal;
the figure identification module is used for acquiring the behavior and the state of a figure and connecting the figure identification module with a computing terminal;
the figure analysis module is used for analyzing the behavior and the state of a figure and connecting the figure identification module with the computing terminal;
the emergency telephone is used for carrying out telephone communication with the vehicle owner and connecting the vehicle owner with the computing terminal;
the visible module is used for carrying out video communication with the vehicle owner and connecting with the computing terminal;
the tunnel broadcast is used for broadcasting information to the vehicle owner and connecting the vehicle owner with the computing terminal;
the indicating lamp is used for displaying the parking state of the parking belt and is connected with the computing terminal;
the explosion and flash induction mark is used for enhancing the identification degree of a parking belt and a traffic lane and is connected with a computing terminal;
and the flashing lamp strip is used for enhancing the identification degree of the parking strip and the traffic lane and connecting the computing terminal with the parking strip.
2. The intelligent early warning induction system for the emergency stop zone of the tunnel according to claim 1, wherein the display module displays the license plate information, the persuasion information and the countdown,
wherein the countdown includes an persuasion countdown and a violation countdown.
3. The intelligent early warning induction system for the tunnel emergency stop zone according to claim 1, wherein the event detection module is a camera; the rearview detection module is a highlight inhibition camera, and the camera shooting direction is aligned to the direction of the vehicle head; the figure recognition module is a high-precision camera, and the camera shooting direction is aligned to the direction of the vehicle head.
4. The intelligent early warning and inducing system for the tunnel emergency stop zone according to claim 1, wherein the indicator light, the flashing inducing sign and the flashing light zone have two states, which respectively correspond to parked vehicles and non-parked vehicles in the stop zone; the emergency telephone and the visual module are provided with an operation scheme for rapidly contacting the tunnel management station.
5. The intelligent early warning induction system for the tunnel emergency stop zone according to claim 1, wherein the computing terminal adopts a GPU + CPU heterogeneous computing architecture.
6. An intelligent early warning induction method for a tunnel emergency stop zone, which is suitable for the intelligent early warning induction system for the tunnel emergency stop zone according to any one of claims 1 to 5, and is characterized by comprising the following steps:
s1: judging whether a vehicle is parked in the parking line or not based on the event detection module, if so, acquiring license plate information by the event detection module;
s2: judging whether the vehicle is a white list vehicle or not based on the license plate information, if so, recording parking records, and if not, performing step S3;
s3: performing a dissuasion countdown, judging whether the vehicle is still parked in the stop line after the dissuasion countdown is finished, if so, performing a step S4, and if not, recording the stop record;
s4: judging whether the vehicle is parked in an accident or a fault vehicle, if so, remotely guiding or processing the vehicle on site by a worker, and recording parking records, otherwise, performing the step S5;
s5: and carrying out illegal parking countdown, judging whether the vehicle is still parked in the parking line after the illegal parking countdown is finished, if so, carrying out field processing and recording parking by a worker, and if not, recording the parking.
7. The intelligent early warning induction method for the tunnel emergency stop zone according to claim 6, wherein when the vehicles with the non-white list license plate drive into the stop zone and do not drive away when exceeding the persuasion countdown, the voice reminding module plays persuasion voice; when a vehicle stops in a parking area, the display module plays a rear road condition video in real time, and the display module displays an auxiliary line of 50 meters, an auxiliary line of 100 meters and an auxiliary line of 200 meters.
8. The intelligent early warning induction method for the tunnel emergency stop zone according to claim 6, wherein the steps S3 and S4 further comprise: the position of the car owner is accurately tracked through a target tracking algorithm, and the state and the behavior of the car owner are identified through a classification algorithm.
9. The intelligent early warning induction method for the tunnel emergency stop zone according to claim 6 or 8, wherein the target tracking algorithm comprises the following steps:
s311: acquiring a video acquired by a person identification module;
s312: selecting an initial frame, taking a driving position as an initial tracking frame, and generating a plurality of candidate tracking frames in a next frame;
s313: extracting the characteristics of the candidate tracking frame, and grading the candidate tracking frame;
s314: fusing a plurality of candidate tracking frames with high scores to obtain an optimal candidate tracking frame;
the candidate tracking frame is generated based on a particle filtering or sliding window method, the characteristics of the candidate tracking frame are depth characteristics learned by a large number of training samples, and the optimal candidate tracking frame is obtained through all prediction weighted averages.
10. The intelligent early warning induction method for the tunnel emergency stop zone according to claim 6 or 8, wherein the classification algorithm comprises the following steps:
s321: acquiring a video sequence, respectively sampling feature points on a plurality of scales of a picture by adopting grid division, and filtering partial feature points;
s322: calculating the movement speed of the characteristic points and tracking the key points;
s323: extracting a direction gradient histogram, an optical flow histogram, a motion boundary histogram and track features along a track;
s324: extracting features from the training set data, and clustering the features by using a K-means clustering algorithm to obtain a feature dictionary;
s325: carrying out quantitative coding on the test data by using the feature dictionary to obtain a vector with a fixed length;
s326: classifying the feature vectors after coding quantization by using a support vector machine;
the partial feature points comprise feature points with less transformation, and the motion speed of the feature points is obtained through optical flow median calculation in the neighborhood of the feature points.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211034857.6A CN115472028A (en) | 2022-08-26 | 2022-08-26 | Intelligent early warning induction method and system for tunnel emergency stop zone |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211034857.6A CN115472028A (en) | 2022-08-26 | 2022-08-26 | Intelligent early warning induction method and system for tunnel emergency stop zone |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115472028A true CN115472028A (en) | 2022-12-13 |
Family
ID=84371044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211034857.6A Pending CN115472028A (en) | 2022-08-26 | 2022-08-26 | Intelligent early warning induction method and system for tunnel emergency stop zone |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115472028A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117636687A (en) * | 2024-01-25 | 2024-03-01 | 江西方兴科技股份有限公司 | Early warning method and system for tunnel emergency stop zone |
Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679125A (en) * | 2012-09-24 | 2014-03-26 | 致伸科技股份有限公司 | Human face tracking method |
| CN106585624A (en) * | 2016-12-07 | 2017-04-26 | 深圳市元征科技股份有限公司 | Driver state monitoring method and device |
| CN106652477A (en) * | 2017-03-06 | 2017-05-10 | 北京惠泽智业科技有限公司 | Violation recognition method and device |
| CN209412722U (en) * | 2018-12-27 | 2019-09-20 | 杭州桐森信息技术有限公司 | It stops in emergency in a kind of tunnel band induction warning system |
| CN110516705A (en) * | 2019-07-19 | 2019-11-29 | 平安科技（深圳）有限公司 | Method for tracking target, device and computer readable storage medium based on deep learning |
| CN210777189U (en) * | 2019-12-27 | 2020-06-16 | 广东华路交通科技有限公司 | Highway tunnel emergency stop takes early warning system |
| CN112660138A (en) * | 2019-09-30 | 2021-04-16 | 比亚迪股份有限公司 | Method and device for detecting health of driver based on steering wheel, steering wheel and vehicle |
| CN213211343U (en) * | 2020-09-30 | 2021-05-14 | 陕西高速交通工贸有限公司 | Highway tunnel emergency stop area alarm system |
| CN215450426U (en) * | 2021-03-01 | 2022-01-07 | 广州中国科学院计算机网络信息中心 | Detection device for emergency lane of expressway |
| CN216311075U (en) * | 2021-09-15 | 2022-04-15 | 浙江台州甬台温高速公路有限公司 | Automatic warning device for emergency stop zone of highway tunnel |
| CN114613172A (en) * | 2022-01-25 | 2022-06-10 | 浙江数智交院科技股份有限公司 | Intelligent sensing and early warning system for tunnel emergency stop zone |
-
2022
- 2022-08-26 CN CN202211034857.6A patent/CN115472028A/en active Pending
Patent Citations (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103679125A (en) * | 2012-09-24 | 2014-03-26 | 致伸科技股份有限公司 | Human face tracking method |
| CN106585624A (en) * | 2016-12-07 | 2017-04-26 | 深圳市元征科技股份有限公司 | Driver state monitoring method and device |
| CN106652477A (en) * | 2017-03-06 | 2017-05-10 | 北京惠泽智业科技有限公司 | Violation recognition method and device |
| CN209412722U (en) * | 2018-12-27 | 2019-09-20 | 杭州桐森信息技术有限公司 | It stops in emergency in a kind of tunnel band induction warning system |
| CN110516705A (en) * | 2019-07-19 | 2019-11-29 | 平安科技（深圳）有限公司 | Method for tracking target, device and computer readable storage medium based on deep learning |
| CN112660138A (en) * | 2019-09-30 | 2021-04-16 | 比亚迪股份有限公司 | Method and device for detecting health of driver based on steering wheel, steering wheel and vehicle |
| CN210777189U (en) * | 2019-12-27 | 2020-06-16 | 广东华路交通科技有限公司 | Highway tunnel emergency stop takes early warning system |
| CN213211343U (en) * | 2020-09-30 | 2021-05-14 | 陕西高速交通工贸有限公司 | Highway tunnel emergency stop area alarm system |
| CN215450426U (en) * | 2021-03-01 | 2022-01-07 | 广州中国科学院计算机网络信息中心 | Detection device for emergency lane of expressway |
| CN216311075U (en) * | 2021-09-15 | 2022-04-15 | 浙江台州甬台温高速公路有限公司 | Automatic warning device for emergency stop zone of highway tunnel |
| CN114613172A (en) * | 2022-01-25 | 2022-06-10 | 浙江数智交院科技股份有限公司 | Intelligent sensing and early warning system for tunnel emergency stop zone |
Non-Patent Citations (2)
| Title |
|---|
| YAQILYU: "单目标跟踪综述", 《知乎，HTTPS://ZHUANLAN.ZHIHU.COM/P/503735985》 * |
| 郭玉杰: "基于视频的目标感知与行为识别", 《万方数据库》, pages 1 - 71 * |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117636687A (en) * | 2024-01-25 | 2024-03-01 | 江西方兴科技股份有限公司 | Early warning method and system for tunnel emergency stop zone |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN101313345B (en) | System and method for detecting driving against road traffic regulation | |
| CN107481521B (en) | The traffic events intelligence means of proof and system based on car networking | |
| CN107067730B (en) | Network appointment vehicle-man-vehicle inconsistency monitoring method based on bayonet equipment | |
| CN109146914B (en) | Drunk driving behavior early warning method for expressway based on video analysis | |
| US20120148092A1 (en) | Automatic traffic violation detection system and method of the same | |
| CN110047272B (en) | A kind of intelligent transportation pedestrian behavior monitoring and alarming system based on big data | |
| CN209388480U (en) | Freeway tunnel safe operation monitors system | |
| CN111353369B (en) | Application method and system of urban road side parking high-level video in auxiliary criminal investigation | |
| CN110288738B (en) | Intelligent comprehensive management and control system and method for bicycle lane | |
| CN112749630A (en) | Intelligent video monitoring method and system for road conditions | |
| CN104134350A (en) | Intelligent dome camera system for traffic violation behavior recognition | |
| CN115472028A (en) | Intelligent early warning induction method and system for tunnel emergency stop zone | |
| KR102500975B1 (en) | Apparatus for learning deep learning model and method thereof | |
| CN113689700A (en) | Method and device for supervising driving of construction sidewalk in mountainous area | |
| CN209486891U (en) | Shared bicycle, which is disobeyed, stops detection system | |
| CN111062238B (en) | Escalator flow monitoring method and system based on human skeleton information and multi-target tracking | |
| CN117414558B (en) | Fire-fighting equipment quality monitoring system based on Internet of things | |
| CN110809137A (en) | Campus safety trampling prevention monitoring system and method | |
| CN111901562B (en) | Campus-oriented audio and video intelligent monitoring system | |
| CN112532928A (en) | Bus-mounted system based on 5G and face recognition and use method | |
| CN111784859A (en) | Special embedded lane control system for expressway | |
| CN116363903A (en) | Anti-collision early warning system for underground passage or tunnel structure | |
| CN115662154A (en) | Intelligent signal lamp system based on video monitoring | |
| CN114821875A (en) | Security check, ticket check and epidemic prevention integrated intelligent device and use method thereof | |
| CN211554957U (en) | Be used for swimming pool personnel that fall into water to discriminate positioner and 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 |