CN113808425A - Traffic light passing early warning device and method for intelligent driving - Google Patents
Traffic light passing early warning device and method for intelligent driving Download PDFInfo
- Publication number
- CN113808425A CN113808425A CN202111121827.4A CN202111121827A CN113808425A CN 113808425 A CN113808425 A CN 113808425A CN 202111121827 A CN202111121827 A CN 202111121827A CN 113808425 A CN113808425 A CN 113808425A
- Authority
- CN
- China
- Prior art keywords
- traffic light
- early warning
- distance
- vehicle
- intelligent driving
- 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
Images
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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60Q—ARRANGEMENT OF SIGNALLING OR LIGHTING DEVICES, THE MOUNTING OR SUPPORTING THEREOF OR CIRCUITS THEREFOR, FOR VEHICLES IN GENERAL
- B60Q9/00—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling
- B60Q9/008—Arrangement or adaptation of signal devices not provided for in one of main groups B60Q1/00 - B60Q7/00, e.g. haptic signalling for anti-collision purposes
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R1/00—Optical viewing arrangements; Real-time viewing arrangements for drivers or passengers using optical image capturing systems, e.g. cameras or video systems specially adapted for use in or on vehicles
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/09—Arrangements for giving variable traffic instructions
- G08G1/0962—Arrangements for giving variable traffic instructions having an indicator mounted inside the vehicle, e.g. giving voice messages
- G08G1/0968—Systems involving transmission of navigation instructions to the vehicle
- G08G1/0969—Systems involving transmission of navigation instructions to the vehicle having a display in the form of a map
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/16—Anti-collision systems
Abstract
The invention discloses a traffic light passing early warning device which can realize the detection of traffic lights of a current lane and the anti-collision early warning of pedestrians and vehicles. The comprehensive reminding and anti-collision early warning of traffic light voice and head-up display images of the driving route direction are realized. The main effects of the invention are as follows: the machine vision technology is combined with the automobile technology, the head-up display technology and the solar charging technology are utilized, the comprehensive reminding of the voice and the image of the traffic light on the current route is realized, and the device can be widely applied to various motor vehicles, the utilization rate of the device is improved, and the comfort level of a user is improved.
Description
Technical Field
The invention relates to the technical field of intelligent auxiliary driving, in particular to a traffic light passing early warning device and method.
Background
Traffic lights are arranged at the traffic intersections to relieve traffic pressure and dredge traffic safely and orderly. Drivers often break rules and regulations due to fatigue or carelessness to break the red light or cause traffic accidents to cause loss of lives and properties, so that traffic light detection and early warning become an important technology in advanced assistant driving and unmanned driving.
At present, the traffic light can not be detected by the conventional automobile data recorder, the state of the traffic light can be detected by the conventional traffic light detection device, the prompt information is redundant, the reminding effect is limited, the condition of the traffic light of the driving route of the driver can not be effectively reported in a targeted manner, the current driving route of the traffic light can not be judged by effectively combining the road information, the collision early warning of the traffic light can not be realized, and the practicability is poor and the robustness is general.
Disclosure of Invention
In order to solve the above problems, the invention provides a traffic light passing early warning device, which can realize the traffic light detection of the current lane and the anti-collision early warning of pedestrians and vehicles. The anti-collision early warning system has the advantages that the machine vision technology is combined with the sensor information of the automobile, the head-up display technology and the solar charging technology are utilized, the anti-collision early warning is effectively reminded of voice and images of traffic lights in the driving route direction, meanwhile, accidents caused by head lowering are avoided, and the user experience degree and the comfort degree are improved.
In order to achieve the purpose and achieve the technical effects, the invention is realized by the following technical scheme: the device comprises a map display screen, a head-up display projection screen, a camera, a solar cell panel and an image processor, wherein the map display screen and the camera are arranged in the opposite direction of a base, the head-up display projection screen is close to a front windshield of an automobile, images can be reflected to a cab through a front window glass, the display screen faces people in the automobile, and the solar cell panel is arranged on the same side of the camera.
As an improvement of the invention, one side of the device is provided with a USB charging interface.
As an improvement of the invention, the power supply is a lithium battery, and USB charging or solar dual charging is adopted.
As an improvement of the invention, the light-blocking frame (4) is arranged at the periphery of the head-up display projection screen (3).
As an improvement of the invention, a shutdown or restart button and a voice broadcast volume adjusting button are arranged on two sides of a map display screen, a shutdown (restart) button is pressed for 3s for shutdown, a shutdown (restart) button is pressed for 6s-10s for restart, and the volume adjusting button is rotated to adjust the broadcast volume.
As an improvement of the invention, the head-up display projection screen projects the vehicle speed, the oil consumption, the map, the traffic light condition and the early warning condition to the front windshield, the camera records the driving image before flameout, and the power supply of the whole device is turned off after flameout.
As an improvement of the present invention, a method for performing early warning using a traffic light co-traveling early warning device for intelligent driving, the method comprising the steps of:
1) the method comprises the following steps of feeding back position information and lane information of a traffic light through common map navigation software, when the map navigation software positions that the distance between an automobile and the traffic light is smaller than a set threshold (a factory sets a reference threshold when the threshold leaves a factory, a user can modify the distance by himself, the positioning distance of the map navigation software is the distance between the traffic light and the front calculated on the navigation software, and the prior art is adopted), detecting the state of the traffic light by a camera, detecting left or right steering light signals of the automobile and lane information of the map navigation software, and judging the left turn, the right turn or the straight running of the automobile (when the automobile is turned, signals can be read by a vehicle machine after the left or right steering light of the automobile is turned on, and the method belongs to the prior art), and after the left turn, the right turn or the straight running lane information of the map navigation software appears, judging the straight running, the left turn or the right turn direction of the automobile by combining the steering light signals, and the specific steps are as follows: if the map navigation software has left turn, right turn and straight movement three direction information, the left turn light of the automobile is turned on, then the automobile is judged to turn left at the moment, and similarly, the right turn light of the automobile is turned on, then the automobile is judged to turn right at the moment, if no turn light information exists, then the automobile is straight movement, after the automobile driving direction is judged, when the vertical distance between the automobile and the traffic light is lower than a set distance, the corresponding traffic light state in the automobile driving direction is broadcasted by voice, and the color of the traffic light is projected to remind people in the automobile on the front windshield;
2) when the vertical distance between the automobile and the traffic light is smaller than a second set distance, detecting whether the automobile speed is greater than 0, and if the automobile speed is greater than 0, broadcasting the color of the projected traffic light by voice to remind personnel in the automobile to run the red light;
3) the camera identifies and tracks the target, the target is detected through a deep learning target detection algorithm, the target is tracked through a tracking algorithm, data are processed through an image processor, and a detection model is trained in the cloud or the local.
4) When the values of the first distance and the second distance are set to be reduced in sequence, when the fact that the target enters a safety zone in the front of the vehicle (namely the target enters a zone in the advancing direction of the vehicle of the user) set in front of the vehicle is detected, whether the speed of the vehicle is larger than 0 is detected, if the speed is larger than 0, a voice broadcast reminding and a projection danger symbol are used for reducing the speed of pedestrians or vehicles for people in the vehicle, the occurrence of collision events is prevented, the speeds of the pedestrians, the bicycles, the electric bicycles and the vehicle are different, and the safety zones of different targets are set according to different users of the detected targets.
Monocular distance measurement is adopted based on the camera distance measurement in the steps, and the distance from the camera to a known object or target is calculated by using similar triangles.
Wherein f is the focal length of the camera, p is the pixel width of the target (traffic light) to be measured, w is the actual width of the target (traffic light) to be measured, d is the distance d from the target (traffic light) to be measured to the camera, and f, p and w are known quantities, so that d can be calculated.
As an improvement of the invention, the first distance is a distance for starting to detect the traffic light and is represented by a vertical distance between the vehicle and the traffic light, the second distance is a distance for judging the running of the traffic light, and the behavior that the vehicle starts running the red light is determined when the second distance is smaller than the first distance.
As an improvement of the invention, the safety area in front of the vehicle in the 4) is an area formed by two safety boundary lines which are parallel to the advancing direction and two safety boundary lines which are perpendicular to the advancing direction in front of the vehicle.
As an improvement of the present invention, in the step 4), a deep learning target detection algorithm is adopted to detect a target, a tracking algorithm is adopted to track the target, data is processed by an image processor, and a detection model is trained in a cloud or a local area.
Based on the scheme, the camera identifies and tracks various vehicles such as pedestrians, bicycles, electric bicycles, automobiles and the like, a deep learning target detection algorithm yolov5 is adopted to detect various vehicles such as pedestrians, bicycles, electric bicycles, automobiles and the like and traffic lights, a tracking algorithm Deepsort is adopted to track targets, an image processor (adopting raspberry group) is adopted to process data, a detection model is trained at the cloud or locally, pictures of various vehicles such as pedestrians, bicycles, electric bicycles, automobiles and the like to be detected and traffic lights are taken as training samples to be sent to a server through a vehicle machine, the server carries out deep learning training after receiving the pictures or locally takes the samples as model input to carry out deep learning training, as the pictures collected and fed back by a user are more and more, the more the pictures contain various conditions, the better detection effect of the deep learning model is, and after the trained detection model is obtained, and the image processor (adopting raspberry pi) calls the trained model for detection, the trained model cloud is pushed to the device, and the continuously updated detection model can be directly used after the detection device downloads the model.
Compared with the prior art, the invention has the following beneficial effects: the invention realizes the traffic light detection of the current lane and the anti-collision early warning of pedestrians and vehicles; the traffic light on the current driving lane is broadcasted through the machine vision technology after the automobile turns left, turns right or runs straight is judged through the automobile sensor information by adopting the machine vision technology and combining the automobile sensor information, and the traffic light voice and image comprehensive reminding and anti-collision early warning in the driving route direction are realized by utilizing the head-up display technology and the solar charging technology, so that accidents caused by head lowering are avoided, and the user experience and comfort are improved; and the device only receives the pushing of the trained model and does not train data, overcomes the defect of insufficient deep learning calculation capacity of the embedded equipment training, can be widely applied to various motor vehicles, helps the vehicles to realize safe driving, and has a wide market.
Drawings
Fig. 1 is a first schematic structural diagram of a traffic light traffic early warning device according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a traffic light traffic early warning device according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram three of a traffic light traffic early warning device according to an embodiment of the present invention;
fig. 4 is a flowchart of a traffic light passing early warning device according to an embodiment of the present invention.
Reference numbers in the figures: volume adjustment button 1, camera 2, new line display projection screen 3, frame 4 that is in the light, solar cell panel 5, shutdown (restart) button 6, map display screen 7, USB interface 8 that charges.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The traffic light passing early warning device for intelligent driving comprises a map display screen 7, a head-up display projection screen 3, a camera 2, a solar cell panel 5 and an image processor, wherein the map display screen 7 and the camera 2 are arranged in opposite directions of a base, the head-up display projection screen is close to a front windshield of an automobile and can reflect images to a cab through a front window glass, the display screen faces to people in the automobile, and the solar cell panel 5 is arranged on the same side of the camera 2.
One specific application of this embodiment is that a USB charging interface 8 is provided beside the map display screen 7.
The image processor used is a raspberry pi.
The power supply is a lithium battery and adopts USB charging or solar energy double charging.
The head-up display projection screen 3 is close to the front windshield of the automobile, and the map display screen 7 faces the people in the automobile.
A shutdown (restart) button 6 and a voice broadcast volume adjusting button 1 are arranged on two sides of the map display screen 7, the shutdown (restart) button 3s is pressed for a long time for shutdown, the shutdown (restart) button 6s-10s is pressed for a long time for restart, and the volume adjusting button 1 is rotated to adjust the broadcast volume.
Map navigation software feeds back traffic light position information and lane information, when the distance between a vehicle and a traffic light is positioned by the map navigation software and is smaller than a set threshold value, the camera 2 detects the state of the traffic light, left (right) steering light signals of the vehicle are detected to be matched with the lane information of the map navigation software, a user is judged to turn left, turn right or go straight, after the pixel diameter of the traffic light is detected, the camera 2 measures the distance, and when the vertical distance between the vehicle and the traffic light is smaller than the set distance by more than one, the state of the traffic light in the driving direction of the vehicle is broadcasted by voice and the color of the traffic light is projected to remind people in the vehicle in front windshield.
Wherein f is the focal length of the camera, p is the pixel width of the target (traffic light) to be measured, w is the actual width of the target (traffic light) to be measured, d is the distance d from the target (traffic light) to be measured to the camera 2, and f, p and w are known quantities, so that d can be calculated.
After the vertical distance between the vehicle and the traffic light is less than the second set distance, whether the vehicle speed is greater than 0 or not is detected, and if the vehicle speed is greater than 0, the color of the projected traffic light is broadcasted through voice to remind people in the vehicle to run the red light.
The method comprises the steps that a camera identifies pedestrians, bicycles, electric bicycles and automobiles and tracks the pedestrians, the bicycles, the electric bicycles and the automobiles, a set distance I value and a set distance II value are sequentially reduced, the set distance I value and the set distance II value are sequentially reduced, a left safety boundary line and a right safety boundary line which are parallel to an advancing direction and a front safety boundary line which is perpendicular to the advancing direction form a safety zone, when a target is detected to enter the safety zone set in front of the automobile (namely the target enters a user automobile advancing direction zone), whether the speed of the automobile is greater than 0 is detected, if the speed of the automobile is greater than 0, voice broadcasting reminding and projection danger symbol in the automobile are carried out, the pedestrians or the automobile decelerates to prevent collision accidents, the pedestrians, the bicycles, the electric bicycles and the automobiles are different in speed, and the safety zones with different targets are automatically set according to different users detecting the target.
The head-up display projection screen 3 projects the vehicle speed, the oil consumption, the map, the traffic light conditions and the early warning conditions to the front windshield, the camera 2 records the driving image before flameout, and the power supply of the whole device is turned off after flameout.
The preferred embodiments of the invention disclosed above are intended to be illustrative only. The preferred embodiments are not intended to be exhaustive or to limit the invention to the precise embodiments disclosed. Obviously, many modifications and variations are possible in light of the above teaching. The embodiments were chosen and described in order to best explain the principles of the invention and the practical application, to thereby enable others skilled in the art to best utilize the invention. The invention is limited only by the claims and their full scope and equivalents.
Claims (10)
1. The utility model provides a traffic lights early warning device that passes for intelligent driving, the device includes map display screen (7), new line display screen (3), camera (2), solar cell panel (5) and image processor, characterized in that, the opposite direction of base sets up map display screen (7) and camera (2), new line display screen (3) are close to car front windshield, accessible front window glass reflection image to driver's cabin, display screen faces the interior personnel of car, the homonymy of camera (2) sets up solar cell panel (5).
2. The traffic light passing early warning device for intelligent driving according to claim 1, wherein the light blocking frame (4) is arranged at the periphery of the head-up display projection screen (3).
3. The traffic light passing early warning device for intelligent driving according to claim 1, wherein: one side of the device is provided with a USB charging interface (8).
4. The traffic light passing early warning device for intelligent driving according to claim 1, wherein: the power supply is a lithium battery and adopts USB charging or solar energy double charging.
5. The traffic light passing early warning device for intelligent driving according to claim 1, wherein: the both sides of the map display screen (7) are provided with a shutdown/restart button (6) and a volume adjusting button (1).
6. The traffic light passing early warning device for intelligent driving according to claim 1, wherein: the head-up display projection screen (3) projects the vehicle speed, the oil consumption, the map, the traffic light condition and the early warning condition to the front windshield.
7. A method for warning by using the traffic light passing warning device for intelligent driving of any one of claims 1 to 6, wherein the method comprises the following steps:
1) after the driving direction of the vehicle is judged by combining with the signal of the automobile steering lamp, the distance measurement is carried out through the camera (2), and when the vertical distance from the vehicle to the traffic light is shorter than a set distance, the state of the traffic light of the lane corresponding to the vehicle at the moment is broadcasted and projected;
2) when the vertical distance between the automobile and the traffic light is less than a second set distance, detecting whether the automobile speed is greater than 0, and prompting the traffic light state and warning information if the automobile speed is greater than 0;
3) identifying and tracking a target by a camera, detecting the target by a deep learning target detection algorithm, tracking the target by adopting a tracking algorithm, processing data by an image processor, and training a detection model at the cloud or locally;
4) and when the values of the first distance and the second distance are set to be reduced in sequence, detecting whether the vehicle speed is greater than 0 or not when the target enters a vehicle front safety zone set in the front of the vehicle, and if the vehicle speed is greater than 0, prompting the current driving state and warning information.
8. The traffic light passing early warning method for intelligent driving according to claim 7, wherein the distance I is a distance for starting traffic light detection and is represented by a vertical distance between a vehicle and a traffic light, the distance II is a distance for judging that the vehicle runs through the traffic light, and the distance II is smaller than the distance II, namely the vehicle starts running the red light.
9. The traffic light traffic early warning method for intelligent driving according to claim 8, wherein the safety zone in front of the vehicle in 4) is a zone formed by a left safety line and a right safety line parallel to the advancing direction and a front safety line perpendicular to the advancing direction.
10. The traffic light traffic early warning method for intelligent driving according to claim 9, wherein the 4) adopts a deep learning target detection algorithm to detect the target, adopts a tracking algorithm to track the target, processes data through an image processor, and trains a detection model at the cloud or locally.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111121827.4A CN113808425A (en) | 2021-09-24 | 2021-09-24 | Traffic light passing early warning device and method for intelligent driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111121827.4A CN113808425A (en) | 2021-09-24 | 2021-09-24 | Traffic light passing early warning device and method for intelligent driving |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113808425A true CN113808425A (en) | 2021-12-17 |
Family
ID=78896717
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111121827.4A Pending CN113808425A (en) | 2021-09-24 | 2021-09-24 | Traffic light passing early warning device and method for intelligent driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113808425A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114694415A (en) * | 2022-03-17 | 2022-07-01 | 安庆师范大学 | Anti-collision early warning method and system for sharp turn curve |
-
2021
- 2021-09-24 CN CN202111121827.4A patent/CN113808425A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114694415A (en) * | 2022-03-17 | 2022-07-01 | 安庆师范大学 | Anti-collision early warning method and system for sharp turn curve |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110517521B (en) | Lane departure early warning method based on road-vehicle fusion perception | |
| Dagan et al. | Forward collision warning with a single camera | |
| CN102910126B (en) | Method and system for safe lane change of auxiliary vehicle | |
| WO2020125178A1 (en) | Vehicle driving prompting method and apparatus | |
| CN109733283B (en) | AR-based shielded barrier recognition early warning system and recognition early warning method | |
| CN105620489A (en) | Driving assistance system and real-time warning and prompting method for vehicle | |
| CN108615383B (en) | Automobile traffic intersection auxiliary passing system based on workshop communication and control method thereof | |
| CN103723073A (en) | Dangerous overtaking early warning device based on multi-sensor data fusion and early warning method thereof | |
| CN106114502A (en) | A kind of intelligent automobile aid system | |
| CN107229906A (en) | A kind of automobile overtaking's method for early warning based on units of variance model algorithm | |
| JP7156011B2 (en) | Information presentation device | |
| CN104135654A (en) | Intelligent 3D (three-dimensional) reverse image method and system thereof | |
| CN112606831A (en) | Anti-collision warning information external interaction method and system for passenger car | |
| CN105818742A (en) | Vehicle early-warning method and system | |
| CN114872713A (en) | Device and method for monitoring abnormal driving state of driver | |
| CN103192758A (en) | Front lamp following turning control method based on machine vision | |
| CN102923000A (en) | Automobile active backing speed limiting control method based on binocular vision target detection | |
| CN103770703A (en) | Driving image identification device and method thereof | |
| CN112319366A (en) | Parking monitoring and early warning method and system | |
| CN113808425A (en) | Traffic light passing early warning device and method for intelligent driving | |
| CN114394108A (en) | Method for reminding driver to start vehicle | |
| CN115384540A (en) | Vehicle-mounted safe driving detection device | |
| KR20170055334A (en) | Appratus and method for assisting a driver based on difficulty level of parking | |
| CN216927819U (en) | Traffic light passing early warning device for intelligent driving | |
| CN111489560A (en) | Detection method and control method for detecting front vehicle starting based on shallow convolutional neural network |
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 |