CN108545019A - A kind of safety driving assist system and method based on image recognition technology - Google Patents
A kind of safety driving assist system and method based on image recognition technology Download PDFInfo
- Publication number
- CN108545019A CN108545019A CN201810306278.XA CN201810306278A CN108545019A CN 108545019 A CN108545019 A CN 108545019A CN 201810306278 A CN201810306278 A CN 201810306278A CN 108545019 A CN108545019 A CN 108545019A
- Authority
- CN
- China
- Prior art keywords
- image
- driver
- photographic device
- driving
- vehicle
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R11/00—Arrangements for holding or mounting articles, not otherwise provided for
- B60R11/04—Mounting of cameras operative during drive; Arrangement of controls thereof relative to the vehicle
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/56—Context or environment of the image exterior to a vehicle by using sensors mounted on the vehicle
- G06V20/588—Recognition of the road, e.g. of lane markings; Recognition of the vehicle driving pattern in relation to the road
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
- G06V40/16—Human faces, e.g. facial parts, sketches or expressions
- G06V40/168—Feature extraction; Face representation
- G06V40/171—Local features and components; Facial parts ; Occluding parts, e.g. glasses; Geometrical relationships
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60R—VEHICLES, VEHICLE FITTINGS, OR VEHICLE PARTS, NOT OTHERWISE PROVIDED FOR
- B60R2300/00—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle
- B60R2300/10—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of camera system used
- B60R2300/105—Details of viewing arrangements using cameras and displays, specially adapted for use in a vehicle characterised by the type of camera system used using multiple cameras
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Oral & Maxillofacial Surgery (AREA)
- General Health & Medical Sciences (AREA)
- Human Computer Interaction (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a kind of safety driving assist system and method based on image recognition technology, the system and method carries out driving behavior detection and deviation detection using image recognition technology, the system includes processor and two groups of photographic devices being attached thereto respectively, wherein the first photographic device acquires track track information, second photographic device acquires driver's driving behavior information, the processor of the system handles the image information of above-mentioned photographic device acquisition, and judge driver whether dangerous driving and driving vehicle whether run-off-road.
Description
Technical field
The present invention relates to auxiliary driving technology fields, and in particular to pre- dangerous driving prevention and lane departure warning base always
In the safety driving assist system and method for image recognition technology.
Background technology
With universal, more and more people's selection car steering of car steering.But in long range driving procedure, drive
Member is often since fatigue driving leads to traffic accident.With the fast development of vehicle electronics technology, major automobile factory
Quotient and R&D institution are all in the electronic technology of preventing fatigue driving that begins one's study.Traditional prevention technique is that vehicle-mounted computer acquisition is driven
The information such as driving time, the driver behavior of the person of sailing find that continuous driving or steering wheel do not occur driver for a long time for a long time
It when variation, issues warning signal, driver is reminded to take care driving.
The patent of invention of Patent No. ZL201510264368.3 discloses a kind of physiological driver's condition monitoring and reply
System, including intelligent wristwatch, car-mounted terminal and cloud server, the current physiological data of the intelligent wristwatch acquisition driver,
And cloud server is sent it to, the history physiological data of driver is stored in the cloud server, and according to driving
The history physiological data of the current physiological data and driver of member judges the current driving condition of driver, when judging result shows
When driver is currently at the bad steering state caused by attention problem, to car-mounted terminal alert, by vehicle-mounted
Terminal notifying driver pays attention to driving, when judging result shows that driver is currently at the state for being unable to normal driving, actively
Contact medical aid system or traffic police center are handled.This method for acquiring physiological driver's feature using intelligent wristwatch is deposited
In prodigious uncertainty:Such as driver forgets to wear Intelligent bracelet etc..Data processing also needs to, by cloud server, ring
Answer speed slow, processing procedure is cumbersome.
Invention content
Goal of the invention:The present invention provides a kind of safety driving assist system and method based on image recognition technology, should
System and method carries out driving behavior detection and deviation detection using image recognition technology, and whether it can detect driver
Dangerous driving, travel route whether run-off-road, the early warning of response can be sent out in time.
Technical solution:To achieve the above object, the invention discloses a kind of safety driving assist system, which includes place
Reason device and two groups of photographic devices being attached thereto respectively, wherein the first photographic device acquires track track information, second takes the photograph
As device acquisition driver's driving behavior information, the processor of the system handles the image information of above-mentioned photographic device acquisition, and
Judge driver whether dangerous driving and driving vehicle whether run-off-road.
Further, the first photographic device includes at least one camera, and the second photographic device includes at least one camera shooting
Head.
Further, processor is connected with alarming device, and alarming device includes loud speaker, and processor detects driver
Other dangerous driving behaviors such as fatigue driving, driving vehicle run-off-road send out warning instruction to processor immediately, remind and drive
Member takes care driving:Reduce the corresponding operatings such as speed, adjustment direction disk.
Further, the first photographic device includes monocular cam or binocular camera, and above-mentioned camera is mounted on vehicle
Inside rear-view mirror neighbouring position or upper windscreen position acquire the lane line information of vehicle body both sides and vehicle body both sides, judge
Driving vehicle whether run-off-road.More accurate distance measurement function can be realized when using binocular camera.
Further, the safe driving assistant system includes:Vehicle panoramic image collection module is configured as obtaining vehicle
Panoramic picture, wherein vehicle panoramic image collection module include the first photographic device;Target gray image collection module, by with
It is set to and obtains target gray image corresponding with the vehicle panoramic image;Candidate feature point is configured as identification module
Candidate feature point pair is identified from the target gray image;Track characteristic point determining module, is configured as according to the time
Characteristic point pair is selected, determines track characteristic point;Lane line deviation detection module is configured as according to the track feature point extraction vehicle
Diatom, and generation deviation is determined whether according to the position relationship between vehicle and the lane line.
Further, the target gray image collection module includes:Transform subblock is configured as the vehicle is complete
Scape image is converted to corresponding initial gray image;Enhancing processing submodule, is configured as making the initial gray image
For image to be reinforced, and enhancing processing is carried out to the image to be reinforced, the image obtained after processing is the target gray figure
Picture;Or the target gray image collection module includes:Transform subblock is configured as converting the vehicle panoramic image
For corresponding initial gray image;Image segmentation submodule is configured as the initial gray image being divided into multiple ashes
Spend subgraph;Enhancing processing submodule, is configured to using each grayscale sub-image as image to be reinforced, and to institute
It states image to be reinforced and carries out enhancing processing, the image obtained after processing is target gray subgraph;Image mosaic submodule, by with
It is set to and the target gray subgraph is spliced, obtain the target gray image.
Further, the installation of the second photographic device in the car neighbouring position above instrument board, room mirror neighbouring position,
One or a combination set of A column neighbouring positions close to driver side, it is ensured that the second photographic device can collect the people of driver
Face image information.Second photographic device includes at least one camera, can increase the camera shooting of corresponding number according to actual needs
Head.Preferably, when the second photographic device includes a camera, which is mounted on instrument board top position, and acquisition drives
Member face direct picture.Preferably, when the second photographic device includes two cameras, camera is separately mounted to close to driver
The A columns neighbouring position and room mirror neighbouring position of side.
Further, when the second photographic device acquisition facial image, while human eye and mouth characteristic being identified, by adopting
Collect the closure information of human eye and mouth, and then can more accurately judge whether driver fatigue state occurs.
Further, camera of second photographic device preferably with infrared function.Infrared photography also can be clearly at night
Acquire the face image of driver.
Further, processor is connected with positioning device, and positioning device includes satellite positioning device and micro- inertial positioning
Device.The satellite positioning device supports GPS/GLONASS/ Big Dipper station-keeping modes.Micro- inertial positioning device includes gyroscope, adds
Speedometer, microprocessor carry out Camera calibration by measuring vehicle three-axis attitude angle (or angular speed) and acceleration, can
To make up the deficiency of satellite positioning component.
Further, processor is connected with communication device, once emergency occurs, communication device is sent out to outside automatically
Send distress signals.When not fainting in steering position or suddenly when cam device collects driver, is injured, immediately to outgoing
Go out emergency instruction, and sends scene photograph and vehicle geographical location information simultaneously.The scene photograph and geographical location that system is sent
Information can submit necessary information reference to rescue centre, to make corresponding urgent measure.
Further, processor is connected by vehicle-mounted CAN bus with vehicle-mounted computer, carries out related data backup, related
Warning information is shown by vehicle-mounted computer display, while also can externally be sent by the wireless communication device of vehicle-mounted computer
Related data information.
Further, system further includes the supply unit being connected with system, and supply unit gives the system power supply, supply unit
Include the solar panel on vehicle body, avoids using Vehicular accumulator cell power supply.
The invention also discloses a kind of safety assistant driving method, this method uses two groups of photographic devices, the first camera shooting dress
Collection vehicle road ahead image is set, and sends an image to processor;Second photographic device acquires driver's driving behavior letter
Breath, and send an image to processor;Processor handles above-mentioned image information respectively, judge fleet vehicles whether run-off-road,
Judge whether driver dangerous driving behavior occurs;If the processor detects that fleet vehicles run-off-road, driver are endangered
Dangerous driving behavior, processor send warning instruction to alarming device, driver safety are reminded to drive immediately.
Further, a kind of method of the safety assistant driving method includes road in front of the first photographic device collection vehicle
Road image is converted into road gray level image by road image, the processor, and edge detection, inspection are carried out to the road gray-scale map
Road traffic marking is surveyed, the road traffic marking is marked, and marks boundary at left and right sides of vehicle body, detection in real time is appointed
The distance change on side vehicle body boundary and its homonymy road traffic marking.
Further, another method of the safety assistant driving method includes that the first photographic device obtains vehicle panoramic
Image;Obtain target gray image corresponding with the vehicle panoramic image;Time is identified from the target gray image
Characteristic point pair is selected, according to the candidate feature point pair, determines track characteristic point;According to the track feature point extraction lane line,
And determined whether that deviation occurs according to the position relationship between vehicle and the lane line.
Wherein, the step of acquisition target gray image corresponding with the vehicle panoramic image includes:It will be described
Vehicle panoramic image is converted to corresponding initial gray image;Using the initial gray image as image to be reinforced, and it is right
The image to be reinforced carries out enhancing processing, and the image obtained after processing is the target gray image;Or it is described acquisition with
The step of vehicle panoramic image corresponding target gray image includes;The vehicle panoramic image is converted to corresponding
Initial gray image;The initial gray image is divided into multiple grayscale sub-images;It respectively will each gray scale subgraph
Enhancing processing is carried out as being used as image to be reinforced, and to the image to be reinforced, the image obtained after processing is target gray
Image;The target gray subgraph is spliced, the target gray image is obtained.
Further, which further includes that the second photographic device acquires driver's driving behavior information,
The processor identifies driver's face characteristic image, and human eye feature part and people are then identified in face characteristic image
Mouth characteristic, judges whether human eye and mouth are closed respectively.
Further, during identifying driver's driving behavior, pending driver's driving behavior image is put into people
Artificial neural networks are trained, and neural metwork training step includes convolution, activation, pond step.
Further, collected driver's driving behavior picture is converted, the picture converted is denoted as img, figure
Target identification position is denoted as label label in piece img, and artificial nerve network model is set to Y (img)=label, the damage in model
Function is lost to be denoted as surelyWherein m is sample number, and i takes 1 to arrive
m;K is number of tags, and l takes 1 to arrive k;E is natural constant.
Advantageous effect:This system can detect deviation and the driving behavior state of driver simultaneously, be travelled by judging
Vehicle whether run-off-road, driver with the presence or absence of dangerous driving behavior come remind driver carry out safe driving.The present invention examines
When surveying driver's driving behavior, multiple driving behavior information such as human eye state and mouth are detected, promptly and accurately judge driver
Driving behavior.The present invention installs positioning device additional, and when emergency occurs for driver, system sends out distress signal in time
While send out vehicle specific location and driver's presence states.
Description of the drawings
Fig. 1 is the system construction drawing of the present invention.
Fig. 2 is the schematic diagram that the first photographic device detects deviation.
Fig. 3 a are the photo schematic diagram that the second photographic device acquires at the A columns close to driver side.
Fig. 3 b are the photo schematic diagram that the second photographic device acquires above instrument board.
Fig. 3 c are the photo schematic diagram that the second photographic device acquires at rearview mirror.
Fig. 4 is driver's picture that camera acquires above instrument board and identifies face partial schematic diagram..
Fig. 5 is to identify human eye feature part and mouth characteristic schematic diagram respectively on facial image..
Fig. 6 a are that photographic device identifies that driving demand power disperses schematic diagram.
Fig. 6 b are that photographic device identifies schematic diagram of making a phone call in driving procedure.
Fig. 6 c are that photographic device identifies schematic diagram of smoking in driving procedure.
Specific implementation mode
The present invention is further described with reference to the accompanying drawings and embodiments.
The invention discloses a kind of safety driving assist system, which carries out driving behavior inspection using image recognition technology
It surveys and deviation detects.System construction drawing is as shown in Figure 1, the system includes processor and two groups of camera shootings being attached thereto respectively
Device, the first photographic device acquire track track information, and the second photographic device acquires driver's driving behavior information, this is
The processor of system handles the image information of above-mentioned photographic device acquisition, and judge driver whether dangerous driving and traveling
Vehicle whether run-off-road.
Further, the first photographic device includes at least one camera, and the second photographic device includes at least one camera shooting
Head.
Further, the first photographic device includes monocular cam or binocular camera, and above-mentioned camera is mounted on vehicle
Inside rear-view mirror neighbouring position or upper windscreen position, collection vehicle road ahead image, as shown in Figure 2.Identify vehicle
The corresponding lane line information in body both sides and vehicle body both sides, judge driving vehicle whether run-off-road.When using binocular camera energy
Enough realize more accurate distance measurement function.
Further, the installation of the second photographic device in the car neighbouring position above instrument board, room mirror neighbouring position,
One or a combination set of A column neighbouring positions close to driver side, it is ensured that the second photographic device can collect the people of driver
Face image information.Second photographic device includes at least one camera, can increase the camera shooting of corresponding number according to actual needs
Head.
Preferably, when the second photographic device includes a camera, which is mounted on instrument board top position, acquisition
Driver face direct picture.
Preferably, when the second photographic device includes two cameras, camera is separately mounted to the A close to driver side
Column neighbouring position and room mirror neighbouring position.
Further, when the second photographic device acquisition facial image, while human eye and mouth characteristic image information are captured, led to
The closure information of acquisition human eye and mouth is crossed, and then to judge whether driver fatigue state occurs.
Further, which is connected with alarming device, and alarming device includes loud speaker, and processor detects driving
Member's fatigue driving or driving vehicle run-off-road send out warning instruction to processor immediately, remind driver to take care and drive
It sails:Reduce the corresponding operatings such as speed, adjustment direction disk.
Further, which is connected with positioning device, and positioning device includes satellite positioning device and inertial positioning
Device.
Further, which is connected with communication device, once emergency occurs, communication device is automatically to outside
Send distress signals.When cam device collects driver not in steering position, or when fainting suddenly, is injured, immediately
Emergency instruction is sent out, and sends scene photograph and vehicle position information simultaneously.
Further, which is connected by vehicle-mounted CAN bus with vehicle-mounted computer, carries out related data backup, phase
It closes warning information to be shown by vehicle-mounted computer display, driver safety is reminded to drive.The system can also pass through vehicle simultaneously
The wireless communication device for carrying computer, externally sends related data information.
Safety driving assist system according to the present invention, one embodiment include:Startup safety driving assist system module,
First camera module, lane detection module, obtains Image Edge-Detection module, Hough transformation vehicle at image pre-processing module
Road fitting module, lane identification module, lane information module, track parameter acquisition module, deviation judgment module, above-mentioned mould
Block is sequentially connected.By means of above-mentioned technical proposal, deviation detection is carried out by starting module, when the automobile detected according to
The analysis of data judged whether on track traveling generate deviation the phenomenon that, and detect automobile deviate when to driver into
Row is reminded, and is started alarming device, is prevented automobile from being travelled in violation of rules and regulations on road, reduce the generation of accident, improve the safety of driving,
Ensure the safety between vehicle.
The invention also discloses a kind of safety assistant driving method, this method uses two groups of photographic devices, the first camera shooting dress
Collection vehicle road ahead image is set, and sends an image to processor;Second photographic device acquires driver's driving behavior letter
Breath, and send an image to processor;Processor handles above-mentioned image information respectively, judge fleet vehicles whether run-off-road,
Judge whether driver dangerous driving behavior occurs;If the processor detects that fleet vehicles run-off-road, driver are endangered
Dangerous driving behavior, processor send warning instruction to alarming device, driver safety are reminded to drive immediately.
A kind of embodiment of the safety assistant driving method includes:First photographic device collection vehicle road ahead image,
Road image is converted into road gray level image by the processor, is carried out edge detection to the road gray-scale map, is detected road
The road traffic marking is marked in traffic marking, and marks boundary at left and right sides of vehicle body, detects either side vehicle in real time
The distance change on boundary and its homonymy road traffic marking at one's side.
Another embodiment of the safety assistant driving method includes that the first photographic device obtains vehicle panoramic image;It obtains
Take target gray image corresponding with the vehicle panoramic image;Candidate feature point is identified from the target gray image
It is right, according to the candidate feature point pair, determine track characteristic point;According to the track feature point extraction lane line, and according to vehicle
Position relationship between the lane line determines whether that deviation occurs.
Wherein, the step of acquisition target gray image corresponding with the vehicle panoramic image includes:It will be described
Vehicle panoramic image is converted to corresponding initial gray image;Using the initial gray image as image to be reinforced, and it is right
The image to be reinforced carries out enhancing processing, and the image obtained after processing is the target gray image;Or it is described acquisition with
The step of vehicle panoramic image corresponding target gray image includes;The vehicle panoramic image is converted to corresponding
Initial gray image;The initial gray image is divided into multiple grayscale sub-images;It respectively will each gray scale subgraph
Enhancing processing is carried out as being used as image to be reinforced, and to the image to be reinforced, the image obtained after processing is target gray
Image;The target gray subgraph is spliced, the target gray image is obtained.
Further, which further includes that the second photographic device acquires driver's driving behavior information,
The processor identifies driver's face characteristic image, and human eye feature part and people are then identified in face characteristic image
Mouth characteristic, judges whether human eye and mouth are closed respectively.
Further, during identifying driver's driving behavior, pending driver's driving behavior image is put into people
Artificial neural networks are trained, and neural metwork training step includes convolution, activation, pond step.
Further, collected driver's driving behavior picture is converted, the picture converted is denoted as img, figure
Target identification position is denoted as label label in piece img, and artificial nerve network model is set to Y (img)=label, the damage in model
Function is lost to be denoted as surelyWherein m is sample number, and i takes 1
To m;K is number of tags, and l takes 1 to arrive k;E is natural constant.
Driver is judged whether for fatigue driving to identify whether human eye is closed below, it is specific to introduce how to handle face
Image information.The photographic device position of acquisition face video fixed first, there are three optional for the installation site of photographic device:One
It is proximate near the A columns of driver side, second is that near room mirror, third, near above instrument board, collected figure
As shown in Fig. 3 a- Fig. 3 c.Camera obtains the image at driver position with the frame per second of 5-30 frames/second in real time.In conjunction with priori,
A part (driver head region) for our interception images is used as area-of-interest, and fixation and recognition is carried out to driver's face.
By picture from RGB color { R:[0,255],G:[0,255],B:[0,255] } it converts to RGB color
{R:[-1.0,1.0],G:[-1.0,1.0],B:[- 1.0,1.0] }, the picture that note has been converted is img, the following institute of conversion formula
Show:
Face location label information in image being arranged as to, { whether the region is background, bounding box centre coordinate, bounding box
Length, the width of bounding box }, be denoted as label label, boundary box label introduces face surrounding enviroment information.
The picture converted and face location label are put into artificial neural network to be trained, neural network design knot
Structure is as follows.
Note neural network model is Y (img)=label.
Wherein, for judging that the position is that two labels of background or face (k=2) use loss function
M is sample number, for judging the loss function of the position face bounding box for loss=| | Y (img)-label |
|2。
Picture is obtained into image img after processing, image img, which is put into neural network model Y, can obtain face location
Label information label can be intercepted in facial image such as Fig. 4 according to label shown in white edge.
The content of eye portion instruction in face picture is denoted as MSG by us, by the face picture intercepted from RGB color
Space { R:[0,255],G:[0,255],B:[0,255] } it converts to RGB color { R:[-1.0,1.0],G:[-1.0,
1.0],B:[- 1.0,1.0] }, the picture converted is remembered into MSG_img, and conversion formula is as follows:
Position of human eye label information in image being arranged as to, { whether the region is background or certain human eye, bounding box center
Coordinate, the length of bounding box, the width of bounding box }, it is denoted as MSG_label, boundary box label introduces human eye surrounding enviroment information.
The picture converted and position of human eye label are put into artificial neural network to be trained, neural network design knot
Structure is as follows.
Note neural network model is MSG_Y (img)=label.
Wherein, for judging that the position is that two labels of background or human eye (k=2) use loss function
M is sample number, for judging that the loss function of the position human eye bounding box is
Loss=| | MSG_Y (img)-label | |2
Human eye picture is obtained into image MSG_img after processing, MSG_img is put into neural network model MSG_Y
Position of human eye label information MSG_label can be obtained, eye image can be intercepted according to MSG_label, as shown in white edge in Fig. 5, into
And judge whether its human eye is closed.
Mouth image is similarly intercepted by above-mentioned similar method, as shown in black surround in Fig. 5, and then whether judges its mouth
It is closed.
Using above-mentioned image processing method, and then it can determine whether out whether driver attention occurs and concentrate, make a phone call, take out
Cigarette, the dangerous driving behaviors such as drink, referring to Fig. 6 a- Fig. 6 c.
Although embodiment of the present invention is described above in association with attached drawing, the invention is not limited in above-mentioned
Specific embodiments and applications field, above-mentioned specific embodiment are only schematical, directiveness, rather than restricted
's.Those skilled in the art under the enlightenment of this specification, in the range for not departing from the claims in the present invention and being protected
In the case of, a variety of forms can also be made, these belong to the row of protection of the invention.
Claims (10)
1. a kind of safety driving assist system, which is characterized in that the system includes that processor and be attached thereto respectively two groups are taken the photograph
As device and alarming device, wherein the first photographic device acquires track track information, the acquisition of the second photographic device drives
The processor of member's driving behavior information, the system handles the image information of above-mentioned photographic device acquisition, and whether judges driver
Dangerous driving, driving vehicle whether run-off-road.
2. safety driving assist system according to claim 1, which is characterized in that the first photographic device includes at least one
Camera, the second photographic device include at least one camera.
3. safety driving assist system according to claim 2, which is characterized in that the second photographic device includes a camera shooting
When head, which is mounted on instrument board top position, acquisition driver face direct picture;Second photographic device includes two
When camera, described two cameras are separately mounted near the A columns neighbouring position of driver side and room mirror
Position.
4. according to the safety driving assist system described in one of claim 1-3, which is characterized in that the second photographic device acquires people
When face image, while identifying human eye and mouth characteristic.
5. according to the safety driving assist system described in one of claim 1-3, which is characterized in that safe driving auxiliary system
System includes:Vehicle panoramic image collection module is configured as obtaining vehicle panoramic image, wherein vehicle panoramic image collection module
Including the first photographic device;Target gray image collection module is configured as obtaining corresponding with the vehicle panoramic image
Target gray image;Candidate feature point is configured as identifying candidate feature from the target gray image to identification module
Point pair;Track characteristic point determining module is configured as, according to the candidate feature point pair, determining track characteristic point;Lane line is inclined
From detection module, it is configured as according to the track feature point extraction lane line, and according to the position between vehicle and the lane line
The relationship of setting determines whether that deviation occurs.
6. a kind of safety assistant driving method, which is characterized in that the method uses two groups of photographic devices, the first photographic device to adopt
Collect vehicle front road image, and sends an image to processor;Second photographic device acquires driver's driving behavior information, and
Send an image to processor;Processor handles above-mentioned image information respectively, judge fleet vehicles whether run-off-road, judge to drive
Whether the person of sailing there is dangerous driving behavior;If the processor detects that there is dangerous driving in fleet vehicles run-off-road, driver
Behavior, processor send warning instruction to alarming device, driver safety are reminded to drive immediately.
7. safety assistant driving method according to claim 6, which is characterized in that in front of the first photographic device collection vehicle
Road image is converted into road gray level image by road image, the processor, and edge detection is carried out to the road gray-scale map,
Road traffic marking is detected, the road traffic marking is marked, and marks boundary at left and right sides of vehicle body, is detected in real time
The distance change on either side vehicle body boundary and its homonymy road traffic marking.
8. safety assistant driving method according to claim 6, which is characterized in that the second photographic device acquisition driver drives
Behavioural information is sailed, the processor identifies driver's face characteristic image, then identifies human eye in face characteristic image
Characteristic and mouth characteristic, judge whether human eye and mouth are closed respectively.
9. safety assistant driving method according to claim 6, which is characterized in that in identification driver's driving behavior process
In, pending driver's driving behavior image is put into artificial neural network and is trained, neural metwork training step includes volume
Product, activation, pond step.
10. safety assistant driving method according to claim 9, which is characterized in that drive collected driver and go
It is converted for picture, the picture converted is denoted as img, and target identification position is denoted as label label in picture img, artificial god
It is set to Y (img)=label through network model, the loss function in model is denoted as surelyWherein m is sample number, and i takes 1 to arrive m;K is number of tags,
L takes 1 to arrive k;E is natural constant.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810306278.XA CN108545019A (en) | 2018-04-08 | 2018-04-08 | A kind of safety driving assist system and method based on image recognition technology |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810306278.XA CN108545019A (en) | 2018-04-08 | 2018-04-08 | A kind of safety driving assist system and method based on image recognition technology |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108545019A true CN108545019A (en) | 2018-09-18 |
Family
ID=63514091
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810306278.XA Pending CN108545019A (en) | 2018-04-08 | 2018-04-08 | A kind of safety driving assist system and method based on image recognition technology |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108545019A (en) |
Cited By (13)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109583393A (en) * | 2018-12-05 | 2019-04-05 | 宽凳(北京)科技有限公司 | A kind of lane line endpoints recognition methods and device, equipment, medium |
CN109866686A (en) * | 2019-04-04 | 2019-06-11 | 王天使 | The intelligent active safety DAS (Driver Assistant System) and method analyzed in real time based on video |
CN109934175A (en) * | 2019-03-15 | 2019-06-25 | 百度在线网络技术(北京)有限公司 | A kind of automobile navigation method, device, electronic equipment and storage medium |
CN110059666A (en) * | 2019-04-29 | 2019-07-26 | 北京市商汤科技开发有限公司 | A kind of attention detection method and device |
CN110341713A (en) * | 2019-07-12 | 2019-10-18 | 东南(福建)汽车工业有限公司 | A kind of driver's holding steering wheel monitoring system and method based on camera |
CN110733418A (en) * | 2019-10-31 | 2020-01-31 | 杭州鸿泉物联网技术股份有限公司 | TBOX-based auxiliary driving method and device |
CN110956072A (en) * | 2019-07-31 | 2020-04-03 | 多伦科技股份有限公司 | Driving skill training method based on big data analysis |
CN111428600A (en) * | 2020-03-17 | 2020-07-17 | 北京都是科技有限公司 | Smoking detection method, system and device and thermal infrared image processor |
CN111814766A (en) * | 2020-09-01 | 2020-10-23 | 中国人民解放军国防科技大学 | Vehicle behavior early warning method and device, computer equipment and storage medium |
CN112818726A (en) * | 2019-11-15 | 2021-05-18 | 杭州海康威视数字技术股份有限公司 | Vehicle violation early warning method, device, system and storage medium |
CN112950809A (en) * | 2021-01-15 | 2021-06-11 | 上海宏英智能科技股份有限公司 | Safe driving auxiliary monitoring device and working method thereof |
CN114550415A (en) * | 2022-01-28 | 2022-05-27 | 交通运输部公路科学研究所 | Vehicle-road-cooperation-based large-scale vehicle lane-level accurate control method and system |
CN114701440A (en) * | 2022-04-11 | 2022-07-05 | 北京经纬恒润科技股份有限公司 | Auxiliary driving system based on HUD |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103310202A (en) * | 2013-06-27 | 2013-09-18 | 西安电子科技大学 | System and method for guaranteeing driving safety |
US20150307108A1 (en) * | 2014-04-28 | 2015-10-29 | Open Tv Inc. | System and method for detecting potential accident situations with a car |
CN107292214A (en) * | 2016-03-31 | 2017-10-24 | 比亚迪股份有限公司 | Deviation detection method, device and vehicle |
-
2018
- 2018-04-08 CN CN201810306278.XA patent/CN108545019A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103310202A (en) * | 2013-06-27 | 2013-09-18 | 西安电子科技大学 | System and method for guaranteeing driving safety |
US20150307108A1 (en) * | 2014-04-28 | 2015-10-29 | Open Tv Inc. | System and method for detecting potential accident situations with a car |
CN107292214A (en) * | 2016-03-31 | 2017-10-24 | 比亚迪股份有限公司 | Deviation detection method, device and vehicle |
Cited By (18)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109583393A (en) * | 2018-12-05 | 2019-04-05 | 宽凳(北京)科技有限公司 | A kind of lane line endpoints recognition methods and device, equipment, medium |
CN109583393B (en) * | 2018-12-05 | 2023-08-11 | 宽凳(北京)科技有限公司 | Lane line end point identification method and device, equipment and medium |
CN109934175B (en) * | 2019-03-15 | 2021-07-06 | 百度在线网络技术(北京)有限公司 | Vehicle navigation method and device, electronic equipment and storage medium |
CN109934175A (en) * | 2019-03-15 | 2019-06-25 | 百度在线网络技术(北京)有限公司 | A kind of automobile navigation method, device, electronic equipment and storage medium |
CN109866686A (en) * | 2019-04-04 | 2019-06-11 | 王天使 | The intelligent active safety DAS (Driver Assistant System) and method analyzed in real time based on video |
CN110059666A (en) * | 2019-04-29 | 2019-07-26 | 北京市商汤科技开发有限公司 | A kind of attention detection method and device |
CN110059666B (en) * | 2019-04-29 | 2022-04-01 | 北京市商汤科技开发有限公司 | Attention detection method and device |
CN110341713A (en) * | 2019-07-12 | 2019-10-18 | 东南(福建)汽车工业有限公司 | A kind of driver's holding steering wheel monitoring system and method based on camera |
CN110956072A (en) * | 2019-07-31 | 2020-04-03 | 多伦科技股份有限公司 | Driving skill training method based on big data analysis |
CN110956072B (en) * | 2019-07-31 | 2023-06-02 | 多伦科技股份有限公司 | Driving skill training method based on big data analysis |
CN110733418A (en) * | 2019-10-31 | 2020-01-31 | 杭州鸿泉物联网技术股份有限公司 | TBOX-based auxiliary driving method and device |
CN112818726A (en) * | 2019-11-15 | 2021-05-18 | 杭州海康威视数字技术股份有限公司 | Vehicle violation early warning method, device, system and storage medium |
CN111428600A (en) * | 2020-03-17 | 2020-07-17 | 北京都是科技有限公司 | Smoking detection method, system and device and thermal infrared image processor |
CN111814766A (en) * | 2020-09-01 | 2020-10-23 | 中国人民解放军国防科技大学 | Vehicle behavior early warning method and device, computer equipment and storage medium |
CN112950809A (en) * | 2021-01-15 | 2021-06-11 | 上海宏英智能科技股份有限公司 | Safe driving auxiliary monitoring device and working method thereof |
CN114550415A (en) * | 2022-01-28 | 2022-05-27 | 交通运输部公路科学研究所 | Vehicle-road-cooperation-based large-scale vehicle lane-level accurate control method and system |
CN114550415B (en) * | 2022-01-28 | 2022-09-06 | 交通运输部公路科学研究所 | Large vehicle lane-level accurate management and control method and system based on vehicle-road cooperation |
CN114701440A (en) * | 2022-04-11 | 2022-07-05 | 北京经纬恒润科技股份有限公司 | Auxiliary driving system based on HUD |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108545019A (en) | A kind of safety driving assist system and method based on image recognition technology | |
CN108860154B (en) | Driver monitoring device and driver monitoring method | |
EP3008708B1 (en) | Vision augmented navigation | |
JP4600357B2 (en) | Positioning device | |
WO2017170148A1 (en) | Flight device, electronic device and program | |
CN106598039B (en) | A kind of Intelligent Mobile Robot barrier-avoiding method based on laser radar | |
CN106064587B (en) | Multi-lane vehicle distance recognition method and device based on license plate recognition | |
CN104786933A (en) | Panoramic image driving auxiliary device and panoramic image driving auxiliary method | |
CN107054218A (en) | identification information display device and method | |
CN110544368B (en) | Fatigue driving augmented reality early warning device and early warning method | |
US20190147270A1 (en) | Information processing apparatus, driver monitoring system, information processing method and computer-readable storage medium | |
CN107408338A (en) | Driver assistance system | |
CN111294564A (en) | Information display method and wearable device | |
KR101281629B1 (en) | Driving guidance system using sensors | |
TW201927610A (en) | Safety confirmation evaluating device, on-vehicle device, safety confirmation evaluation system having the two, safety confirmation evaluation method, and safety confirmation evaluation program | |
JP2005202787A (en) | Display device for vehicle | |
JP2019159926A (en) | Driver replacement promotion device, drive evaluation system having the same thereon, and computer program | |
CN116129641B (en) | Vehicle security situation calculation method and system based on multi-terminal collaborative identification | |
JP7114953B2 (en) | In-vehicle device, driving evaluation device, driving evaluation system provided with these, data transmission method, and data transmission program | |
JP3865856B2 (en) | Vehicle driving support device | |
WO2021112030A1 (en) | Display control system, display control method, and program | |
WO2021149594A1 (en) | Information provision device, information provision method, information provision program, and recording medium | |
KR101872477B1 (en) | Vehicle | |
JP7070827B2 (en) | Driving evaluation device, in-vehicle device, driving evaluation system equipped with these, driving evaluation method, and driving evaluation program | |
JP7060841B2 (en) | Operation evaluation device, operation evaluation method, and operation evaluation program |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180918 |