CN107301379A - A kind of steering wheel and manipulator's detection method and system based on machine vision - Google Patents
A kind of steering wheel and manipulator's detection method and system based on machine vision Download PDFInfo
- Publication number
- CN107301379A CN107301379A CN201710389936.1A CN201710389936A CN107301379A CN 107301379 A CN107301379 A CN 107301379A CN 201710389936 A CN201710389936 A CN 201710389936A CN 107301379 A CN107301379 A CN 107301379A
- Authority
- CN
- China
- Prior art keywords
- interest
- single area
- manipulator
- steering wheel
- point
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- 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
- G06V20/59—Context or environment of the image inside of a vehicle, e.g. relating to seat occupancy, driver state or inner lighting conditions
- G06V20/597—Recognising the driver's state or behaviour, e.g. attention or drowsiness
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/255—Detecting or recognising potential candidate objects based on visual cues, e.g. shapes
-
- 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
- G06V40/107—Static hand or arm
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Theoretical Computer Science (AREA)
- Human Computer Interaction (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of steering wheel based on machine vision and manipulator's detection method and system, pass through hough ellipses detection outgoing direction disks position, accurately detect that 360 single area-of-interests whether there is hand on steering wheel with reference to half-tone information and Skin Color Information, it whether there is manipulator on omnirange disk so as to judge single-frame images, and combine multi frame detection and multiframe judgement, decide whether send alarm signal, remind driver, reduce the generation of traffic accident;Real-time is high, accuracy rate is high, robustness and stability are high.
Description
Technical field
The present invention relates to technical field of image processing, more particularly, to a kind of steering wheel based on machine vision and manipulator
Detection method and system.
Background technology
The generation of traffic accident has substantial connection with the incorrect driver behavior of driver, and some driving behaviors are easily driven
The person of sailing despises, but easily causes very serious consequence, especially driver's both hands departure direction disk.When running into urgent feelings
Condition, such as burst vehicle flat tire, road conditions change or vehicle occurs suddenly in front, easily cause driver right
Vehicle is fast and effectively controlled, so as to cause extremely serious consequence.
The examined equipment complexity of method and price factor influence based on manipulator on physiological signal measurements steering wheel
While, in addition it is also necessary to driver's wearing device, the driver behavior of driver is had a certain impact, therefore, it is difficult to promote realization.
Based on sensor detected vehicle state so as to detect that the method measurement result of manipulator on steering wheel is easily grasped by driving
Make the presence of the influences of Individual differences, while also there are problems that measurement data is not accurate enough and the limit of hardware cost relatively
System, causes false alarm rate higher, therefore also fail to obtain preferable application effect.
Manipulator's detection algorithm is simple to operate on existing steering wheel, mainly first calibrates the position of steering wheel in image
Put, then carry out the interception of area-of-interest, manipulator's detection is finally carried out on the region of interest.Existing method needs artificial
Manually the position of steering wheel is marked, and inclined rectangle is taken in the interception of area-of-interest, needed when the later stage is detected
Angle is done into normalized, the algorithm process time is increased, and the size of area-of-interest is fixed, and determines whether operation
The proportion threshold value of hand is fixed, it is impossible to adaptive various scene types, if driver's arm hangs down when being placed on leg naturally, can be produced
Flase drop.
The content of the invention
It is an object of the invention to overcome above-mentioned technical deficiency, a kind of steering wheel and manipulator based on machine vision are proposed
Detection method and system, solve above-mentioned technical problem of the prior art.
To reach above-mentioned technical purpose, technical scheme provides a kind of steering wheel and operation based on machine vision
Hand detection method, including:
S1, one frame direction disk image of collection, detect that steering wheel image obtains steering wheel institute using hough Ellipses Detections
Ellipse, and determine 360 points on ellipse to be oval in the way of determining a point on ellipse at interval of 1 °
Line test point;
S2, the boundary rectangle from steering wheel image where interception steering wheel are used as the first picture, the whole of the first picture
Scope is area-of-interest, and area-of-interest is pre-processed;
S3, one single area-of-interest of acquisition, the single area-of-interest are first to choose oval on the first picture
Any one point in line test point, then centered on the point of selection, interception size is size, and angle is used as list for 0 rectangle
One area-of-interest, size size changes according to area-of-interest size adaptation;
S4, the binary image for obtaining the single area-of-interest obtained in S3, the binary image is represented in two dimension
Projected in coordinate system and to X-direction, the gray value for obtaining each row of the binary image is 255 pixel number A, and finding should
The specific continuous several columns of binary image and the length L for calculating continuous several columns, the binary image are specifically continuous
Several columns meet wherein each column pixel number A be all higher than first threshold T1;
S5, the single area-of-interest progress Face Detection obtained to S3, obtain the binaryzation of the single area-of-interest
Hand skin color gray value is 255 in image, binary image, and identification ellipse test point falls the point in the single area-of-interest
Collection, the gray value for the point that test point is concentrated is 255 number N;
S6, work as L>Second Threshold T2 and N>During the 3rd threshold value T3, judge that the single area-of-interest that S3 is obtained has operation
Hand, otherwise, judges that manipulator is not present in the single area-of-interest that S3 is obtained;
S7, circulation perform S3-S6 steps, and the central point using the S3 single area-of-interests obtained is with 1 ° as starting point
Step-length according to clockwise direction obtain single area-of-interest one by one, judge 360 centered on ellipse test point it is single
Area-of-interest whether there is manipulator;
Whether the single area-of-interest that S8, judgement have manipulator is continuous, and calculating continuously has the single of manipulator
The number B of area-of-interest, works as B>During the 4th threshold value T4, judge there is manipulator in the frame direction disk image gathered in S1,
Otherwise, judge manipulator is not present in the frame direction disk image gathered in S1;
S9, circulation perform step S1-S8, the steering wheel image of collection setting frame number, the steering wheel figure of synthetic setting frame number
Manipulator's judged result of picture decides whether to remind driver.
The present invention also provides a kind of steering wheel based on machine vision and manipulator's detecting system, including:
Ellipses detection module:A frame direction disk image is gathered, is obtained using hough Ellipses Detections detection steering wheel image
Take the ellipse where steering wheel, and determined in the way of a point on ellipse is determined at interval of 1 ° on ellipse 360
Individual point is ellipse test point;
Area-of-interest acquisition module:Boundary rectangle where steering wheel is intercepted from steering wheel image is used as the first figure
Piece, the four corner of the first picture is area-of-interest, and area-of-interest is pre-processed;
Single area-of-interest acquisition module:A single area-of-interest is obtained, the single area-of-interest is first
Any one point in ellipse test point is chosen on the first picture, then centered on the point of selection, interception size is size,
The rectangle that angle is 0 changes as single area-of-interest, size size according to area-of-interest size adaptation;
First detection module:Obtain the binaryzation of the single area-of-interest obtained in single area-of-interest acquisition module
Image, the binary image is represented to project in two-dimensional coordinate system and to X-direction, each row of the binary image are obtained
Gray value is 255 pixel number A, finds the specific continuous several columns of the binary image and calculates continuous several columns
Length L, the pixel number A that the specific continuous several columns of the binary image meet wherein each column is all higher than first threshold T1;
Second detection module:Colour of skin inspection is carried out to the single area-of-interest that single area-of-interest acquisition module is obtained
Survey, it is 255 to obtain hand skin color gray value in the binary image of the single area-of-interest, binary image, and identification is oval
Line test point falls the point set in the single area-of-interest, and the gray value for the point that test point is concentrated is 255 number N;
Single area-of-interest judge module:Work as L>Second Threshold T2 and N>During the 3rd threshold value T3, judge single interested
There is manipulator in the single area-of-interest that region acquisition module is obtained, otherwise, judge that single area-of-interest acquisition module is obtained
Manipulator is not present in the single area-of-interest taken;
First circulation module:Circulation performs single area-of-interest acquisition module to single area-of-interest judge module
Operation, the central point of the single area-of-interest obtained using single area-of-interest acquisition module is starting point, with 1 ° for step-length
Single area-of-interest is obtained one by one according to clockwise direction, judges that 360 single senses centered on ellipse test point are emerging
Interesting region whether there is manipulator;
Single frames judge module:Judge whether the single area-of-interest that there is manipulator is continuous, calculates and continuously there is behaviour
Make the number B of the single area-of-interest of hand, work as B>During the 4th threshold value T4, the frame direction gathered in ellipses detection module is judged
There is manipulator in disk image, otherwise, judge manipulator is not present in the frame direction disk image gathered in ellipses detection module;
Multiframe judge module:Circulation performs ellipses detection module to the operation of single frames judge module, collection setting frame number
Steering wheel image, manipulator's judged result of the steering wheel image of synthetic setting frame number decides whether to remind driver.
Compared with prior art, beneficial effects of the present invention include:The present invention takes hough ellipses detection travel direction disks
Positioning, ellipse where outgoing direction disk can be automatically positioned by algorithm, automaticity is high;Take the list that the anglec of rotation is 0
One area-of-interest, method is simple, and real-time is high;In the prior art only by judging the binary image of single area-of-interest
Determine that the region whether there is similar hand, hung down naturally and close to steering wheel if there is driver's arm, from the angle of imaging
Degree sees similar hand on the steering wheel, easily causes the situation of flase drop, and the present invention adds a constraints, is existed using ellipse
Steering wheel center (direction disk shape can regard two ellipses as, and ellipse one is scheduled in the middle of two ellipses), even if there is arm certainly
Be not in still hand on situation about so hanging down, steering wheel place ellipse, it is ellipse only when hand is really covered on the steering wheel
Round wires can be just capped, and improve the accuracy rate of algorithm;The method that multiple threshold values of the prior art use fixed threshold, no
Can adaptive various scenes and different size of image, the present invention according to the size of area-of-interest it is adaptive should determine that it is single
Region of interest domain sizes, while according to single size adaptive threshold value T1, T2, T3 interested, improving the robustness of algorithm;
The present invention judges to decide whether to remind driver, raising Stability and veracity and algorithm speed using multiframe.
Brief description of the drawings
Fig. 1 is a kind of steering wheel and manipulator's detection method flow chart based on machine vision that the present invention is provided;
Fig. 2 is a kind of steering wheel and manipulator's detecting system structured flowchart based on machine vision that the present invention is provided.
In accompanying drawing:1st, based on machine vision steering wheel and manipulator's detecting system, 11, ellipses detection module, 12, feel emerging
Interesting region acquisition module, 13, single area-of-interest acquisition module, 14, first detection module, the 15, second detection module, 16,
Single area-of-interest judge module, 17, first circulation module, 18, single frames judge module, 19, multiframe judge module.
Embodiment
In order to make the purpose , technical scheme and advantage of the present invention be clearer, it is right below in conjunction with drawings and Examples
The present invention is further elaborated.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
It is not used in the restriction present invention.
The invention provides a kind of steering wheel based on machine vision and manipulator's detection method, including:
S1, one frame direction disk image of collection, detect that steering wheel image obtains steering wheel institute using hough Ellipses Detections
Ellipse, and determine 360 points on ellipse to be oval in the way of determining a point on ellipse at interval of 1 °
Line test point;
S2, the boundary rectangle from steering wheel image where interception steering wheel are used as the first picture, the whole of the first picture
Scope is area-of-interest, and area-of-interest is pre-processed;
S3, one single area-of-interest of acquisition, the single area-of-interest are first to choose oval on the first picture
Any one point in line test point, then centered on the point of selection, interception size is size, and angle is used as list for 0 rectangle
One area-of-interest, size size changes according to area-of-interest size adaptation;
S4, the binary image for obtaining the single area-of-interest obtained in S3, the binary image is represented in two dimension
Projected in coordinate system and to X-direction, the gray value for obtaining each row of the binary image is 255 pixel number A, and finding should
The specific continuous several columns of binary image and the length L for calculating continuous several columns, the binary image are specifically continuous
Several columns meet wherein each column pixel number A be all higher than first threshold T1, T1 according to size adaptive changes;
S5, the single area-of-interest progress Face Detection obtained to S3, obtain the binaryzation of the single area-of-interest
Hand skin color gray value is 255 in image, binary image, and identification ellipse test point falls the point in the single area-of-interest
Collection, the gray value for the point that test point is concentrated is 255 number N;
S6, work as L>Second Threshold T2 and N>During the 3rd threshold value T3, judge that the single area-of-interest that S3 is obtained has operation
Hand, otherwise, judges that manipulator is not present in the single area-of-interest that S3 is obtained, T2, T3 are according to size adaptive changes;
S7, circulation perform S3-S6 steps, and the central point using the S3 single area-of-interests obtained is with 1 ° as starting point
Step-length according to clockwise direction obtain single area-of-interest one by one, judge 360 centered on ellipse test point it is single
Area-of-interest whether there is manipulator;
Whether the single area-of-interest that S8, judgement have manipulator is continuous, and calculating continuously has the single of manipulator
The number B of area-of-interest, works as B>During the 4th threshold value T4, judge there is manipulator in the frame direction disk image gathered in S1,
Otherwise, judge manipulator is not present in the frame direction disk image gathered in S1;
S9, circulation perform step S1-S8, the steering wheel image of collection setting frame number, the steering wheel figure of synthetic setting frame number
Manipulator's judged result of picture decides whether to remind driver.
In steering wheel and manipulator's detection method of the present invention based on machine vision, step S2:
The pretreatment operation of gray processing and rim detection is carried out to area-of-interest.
In steering wheel and manipulator's detection method of the present invention based on machine vision, step S4:
It is to the method that the single area-of-interest progress binarization operation obtained in S3 obtains binary image:
Face Detection is carried out to single area-of-interest using Skin Color Information empirical value, binary image, binaryzation is obtained
Skin pixel point gray value is 255 in image, and non-skin pixel point gray value is 0 or calculates being averaged for single area-of-interest
Gray scale, carries out binarization of gray value to single area-of-interest as threshold value using average gray, obtains binary image.
In steering wheel and manipulator's detection method of the present invention based on machine vision, step S8:
The central point of single area-of-interest is ellipse test point, if the central point of some single area-of-interests is
Continuous point, then judge that each single area-of-interest is continuous on ellipse.
Synthetic setting frame number in steering wheel and manipulator's detection method of the present invention based on machine vision, step S9
Steering wheel image manipulator's judged result decide whether remind driver the step of be:
The frame number C that manipulator is not present in the steering wheel image of setting frame number is obtained, if C>5th threshold value T5, then remind
Driver takes care driving.
The present invention also provides a kind of steering wheel based on machine vision and manipulator's detecting system 1, including:
Ellipses detection module 11:A frame direction disk image is gathered, steering wheel image is detected using hough Ellipses Detections
The ellipse where steering wheel is obtained, and is determined in the way of a point on ellipse is determined at interval of 1 ° on ellipse
360 points are ellipse test point;
Area-of-interest acquisition module 12:Boundary rectangle where steering wheel is intercepted from steering wheel image is used as the first figure
Piece, the four corner of the first picture is area-of-interest, and area-of-interest is pre-processed;
Single area-of-interest acquisition module 13:A single area-of-interest is obtained, the single area-of-interest is
Any one point in ellipse test point is first chosen on the first picture, then centered on the point of selection, interception size is
Size, the rectangle that angle is 0 changes as single area-of-interest, size size according to area-of-interest size adaptation;
First detection module 14:Obtain the two-value of the single area-of-interest obtained in single area-of-interest acquisition module
Change image, the binary image is represented to project in two-dimensional coordinate system and to X-direction, each row of the binary image are obtained
Gray value be 255 pixel number A, find the specific continuous several columns of the binary image and calculate continuous several columns
Length L, the pixel number A that the specific continuous several columns of the binary image meet wherein each column is all higher than first threshold
T1, T1 are according to size adaptive changes;
Second detection module 15:Colour of skin inspection is carried out to the single area-of-interest that single area-of-interest acquisition module is obtained
Survey, it is 255 to obtain hand skin color gray value in the binary image of the single area-of-interest, binary image, and identification is oval
Line test point falls the point set in the single area-of-interest, and the gray value for the point that test point is concentrated is 255 number N;
Single area-of-interest judge module 16:Work as L>Second Threshold T2 and N>During the 3rd threshold value T3, judge that single sense is emerging
There is manipulator in the single area-of-interest that interesting region acquisition module is obtained, otherwise, judge single area-of-interest acquisition module
Manipulator is not present in the single area-of-interest obtained, and T2, T3 are according to size adaptive changes;
First circulation module 17:Circulation performs single area-of-interest acquisition module to single area-of-interest judge module
Operation, using single area-of-interest acquisition module obtain single area-of-interest central point as starting point, with 1 ° for step
It is long to obtain single area-of-interest one by one according to clockwise direction, judge 360 single senses centered on ellipse test point
Interest region whether there is manipulator;
Single frames judge module 18:Judge whether the single area-of-interest that there is manipulator is continuous, calculates and continuously exists
The number B of the single area-of-interest of manipulator, works as B>During the 4th threshold value T4, the frame side gathered in ellipses detection module is judged
There is manipulator into disk image, otherwise, judge operation is not present in the frame direction disk image gathered in ellipses detection module
Hand;
Multiframe judge module 19:Circulation performs ellipses detection module to the operation of single frames judge module, collection setting frame number
Steering wheel image, manipulator's judged result of the steering wheel image of synthetic setting frame number decides whether to remind driver.
Steering wheel and manipulator's detecting system 1 of the present invention based on machine vision, area-of-interest acquisition module
In 12:
The pretreatment operation of gray processing and rim detection is carried out to area-of-interest.
In steering wheel and manipulator's detecting system 1 of the present invention based on machine vision, first detection module 14:
Face Detection is carried out to single area-of-interest using Skin Color Information empirical value, binary image, binaryzation is obtained
Skin pixel point gray value is 255 in image, and non-skin pixel point gray value is 0 or calculates being averaged for single area-of-interest
Gray scale, carries out binarization of gray value to single area-of-interest as threshold value using average gray, obtains binary image.
In steering wheel and manipulator's detecting system 1 of the present invention based on machine vision, single frames judge module 18:
The central point of single area-of-interest is ellipse test point, if the central point of some single area-of-interests is
Continuous point, then judge that each single area-of-interest is continuous on ellipse.
In steering wheel and manipulator's detecting system 1 of the present invention based on machine vision, multiframe judge module 19:
The frame number C that manipulator is not present in the steering wheel image of setting frame number is obtained, if C>5th threshold value T5, then remind
Driver takes care driving.
Compared with prior art, beneficial effects of the present invention include:The present invention takes hough ellipses detection travel direction disks
Positioning, ellipse where outgoing direction disk can be automatically positioned by algorithm, it is not necessary to manually mark the position of steering wheel;It is existing
With the presence of the technology area-of-interest anglec of rotationTherefore the normalization of progress angle is needed, and angle normalization is time-consuming, and
EMS memory occupation is big, and the present invention takes the single area-of-interest that the anglec of rotation is 0, and method is simple, and real-time is high;In the prior art
Only by judging that the binary image of single area-of-interest determines that the region whether there is similar hand, if there is driver's hand
Arm hangs down naturally and close to steering wheel, similar hand on the steering wheel, easily causes the situation of flase drop from the point of view of imaging, and
The present invention one constraints of addition, using ellipse, in steering wheel center, (direction disk shape can regard two ellipses, ellipse as
One is scheduled in the middle of two ellipses), even if there is a situation where that arm hangs down naturally, be not in still on ellipse where steering wheel
Hand, only when hand is really covered on the steering wheel, ellipse can be just capped, and improve the accuracy rate of algorithm;Prior art
In multiple threshold values method for using fixed threshold, it is impossible to adaptive various scenes and different size of image, root of the present invention
Single region of interest domain sizes are should determine that according to the size of area-of-interest is adaptive, while adaptive according to single size interested
Threshold value T1, T2, T3 are should determine that, the robustness of algorithm is improved;The present invention is judged to decide whether to remind driver using multiframe, improved
Stability and veracity and algorithm speed.
The embodiment of present invention described above, is not intended to limit the scope of the present invention..Any basis
Various other corresponding changes and deformation that the technical concept of the present invention is made, should be included in the guarantor of the claims in the present invention
In the range of shield.
Claims (10)
1. a kind of steering wheel and manipulator's detection method based on machine vision, it is characterised in that including step:
S1, one frame direction disk image of collection, using where hough Ellipses Detections detection steering wheel image acquisition steering wheel
Ellipse, and determine that 360 points on ellipse are examined as ellipse in the way of a point on ellipse is determined at interval of 1 °
Measuring point;
S2, the boundary rectangle from steering wheel image where interception steering wheel are used as the first picture, the whole of first picture
Scope is area-of-interest, and area-of-interest is pre-processed;
S3, one single area-of-interest of acquisition, the single area-of-interest are first to choose described on first picture
Any one point in ellipse test point, then centered on the point of selection, interception size is size, and angle is made for 0 rectangle
For single area-of-interest, size size changes according to area-of-interest size adaptation;
S4, the binary image for obtaining the single area-of-interest obtained in S3, the binary image is represented in two-dimensional coordinate
Projected in system and to X-direction, the gray value for obtaining each row of the binary image is 255 pixel number A, finds the two-value
Change the specific continuous several columns of image and calculate the length L of continuous several columns, if the binary image is specific continuous
The pixel number A that dry row meet wherein each column is all higher than first threshold T1, T1 according to size adaptive changes;
S5, the single area-of-interest progress Face Detection obtained to S3, obtain the binary image of the single area-of-interest,
Hand skin color gray value is 255 in binary image, recognizes that the ellipse test point falls the point in the single area-of-interest
Collection, the gray value for the point that test point is concentrated is 255 number N;
S6, work as L>Second Threshold T2 and N>During the 3rd threshold value T3, judge that the single area-of-interest that S3 is obtained has manipulator, it is no
Then, judge that manipulator is not present in the single area-of-interest that S3 is obtained, T2, T3 are according to size adaptive changes;
S7, circulation perform S3-S6 steps, and the central point using the S3 single area-of-interests obtained is starting point, with 1 ° for step-length
According to clockwise direction obtain single area-of-interest one by one, judge 360 centered on the ellipse test point it is single
Area-of-interest whether there is manipulator;
Whether the single area-of-interest that S8, judgement have manipulator is continuous, calculates the continuous single sense that there is manipulator emerging
The number B in interesting region, works as B>During the 4th threshold value T4, judge there is manipulator in the frame direction disk image gathered in S1, otherwise,
Judge manipulator is not present in the frame direction disk image gathered in S1;
S9, circulation perform step S1-S8, and collection sets the steering wheel image of frame number, the steering wheel image of synthetic setting frame number
Manipulator's judged result decides whether to remind driver.
2. steering wheel and manipulator's detection method as claimed in claim 1 based on machine vision, it is characterised in that step S2
In:
The pretreatment operation of gray processing and rim detection is carried out to area-of-interest.
3. steering wheel and manipulator's detection method as claimed in claim 1 based on machine vision, it is characterised in that step S4
In:
It is to the method that the single area-of-interest progress binarization operation obtained in S3 obtains binary image:
Face Detection is carried out to single area-of-interest using Skin Color Information empirical value, binary image, binary image is obtained
Middle skin pixel point gray value is 255, and non-skin pixel point gray value is 0 or calculates the average ash of single area-of-interest
Degree, carries out binarization of gray value to single area-of-interest as threshold value using average gray, obtains binary image.
4. steering wheel and manipulator's detection method as claimed in claim 1 based on machine vision, it is characterised in that step S8
In:
The central point of the single area-of-interest is the ellipse test point, if the center of some single area-of-interests
Point is the continuous point on ellipse, then judges that each single area-of-interest is continuous.
5. steering wheel and manipulator's detection method as claimed in claim 1 based on machine vision, it is characterised in that step S9
Manipulator's judged result of the steering wheel image of middle synthetic setting frame number decide whether remind driver the step of be:
The frame number C that manipulator is not present in the steering wheel image of setting frame number is obtained, if C>5th threshold value T5, then remind and drive
Member takes care driving.
6. a kind of steering wheel and manipulator's detecting system based on machine vision, it is characterised in that including:
Ellipses detection module:A frame direction disk image is gathered, steering wheel image acquisition side is detected using hough Ellipses Detections
360 points on ellipse are determined to the ellipse where disk, and in the way of a point on ellipse is determined at interval of 1 °
For the ellipse test point;
Area-of-interest acquisition module:Boundary rectangle where steering wheel is intercepted from steering wheel image is used as first figure
Piece, the four corner of first picture is area-of-interest, and area-of-interest is pre-processed;
Single area-of-interest acquisition module:A single area-of-interest is obtained, the single area-of-interest is first in institute
Any one point chosen on the first picture in the ellipse test point is stated, then centered on the point of selection, interception size is
Size, the rectangle that angle is 0 changes as single area-of-interest, size size according to area-of-interest size adaptation;
First detection module:Obtain the binary picture of the single area-of-interest obtained in single area-of-interest acquisition module
Picture, the binary image is represented to project in two-dimensional coordinate system and to X-direction, the ash of each row of the binary image is obtained
Angle value is 255 pixel number A, finds the specific continuous several columns of the binary image and calculates the length of continuous several columns
L is spent, the pixel number A that the specific continuous several columns of the binary image meet wherein each column is all higher than first threshold T1, T1
According to size adaptive changes;
Second detection module:Face Detection is carried out to the single area-of-interest that single area-of-interest acquisition module is obtained, obtained
It is 255 to take hand skin color gray value in the binary image of the single area-of-interest, binary image, recognizes the ellipse
Test point falls the point set in the single area-of-interest, and the gray value for the point that test point is concentrated is 255 number N;
Single area-of-interest judge module:Work as L>Second Threshold T2 and N>During the 3rd threshold value T3, single area-of-interest is judged
There is manipulator in the single area-of-interest that acquisition module is obtained, otherwise, judge what single area-of-interest acquisition module was obtained
Manipulator is not present in single area-of-interest, and T2, T3 are according to size adaptive changes;
First circulation module:Circulation performs single area-of-interest acquisition module to the behaviour of single area-of-interest judge module
Make, the central point of the single area-of-interest obtained using single area-of-interest acquisition module is pressed as starting point with 1 ° for step-length
Single area-of-interest is obtained one by one according to clockwise direction, judges 360 single senses centered on the ellipse test point
Interest region whether there is manipulator;
Single frames judge module:Judge whether the single area-of-interest that there is manipulator is continuous, and calculating continuously has manipulator
Single area-of-interest number B, work as B>During the 4th threshold value T4, the frame direction disk figure gathered in ellipses detection module is judged
There is manipulator as in, otherwise, judge manipulator is not present in the frame direction disk image gathered in ellipses detection module;
Multiframe judge module:Circulation performs ellipses detection module to the operation of single frames judge module, the direction of collection setting frame number
Disk image, manipulator's judged result of the steering wheel image of synthetic setting frame number decides whether to remind driver.
7. steering wheel and manipulator's detecting system as claimed in claim 6 based on machine vision, it is characterised in that interested
In the acquisition module of region:
The pretreatment operation of gray processing and rim detection is carried out to the area-of-interest.
8. steering wheel and manipulator's detecting system as claimed in claim 6 based on machine vision, it is characterised in that the first inspection
Survey in module:
Face Detection is carried out to single area-of-interest using Skin Color Information empirical value, binary image, binary image is obtained
Middle skin pixel point gray value is 255, and non-skin pixel point gray value is 0 or calculates the average ash of single area-of-interest
Degree, carries out binarization of gray value to single area-of-interest as threshold value using average gray, obtains binary image.
9. steering wheel and manipulator's detecting system as claimed in claim 6 based on machine vision, it is characterised in that single frames is sentenced
In disconnected module:
The central point of the single area-of-interest is the ellipse test point, if the center of some single area-of-interests
Point is the continuous point on ellipse, then judges that each single area-of-interest is continuous.
10. steering wheel and manipulator's detecting system as claimed in claim 6 based on machine vision, it is characterised in that multiframe
In judge module:
The frame number C that manipulator is not present in the steering wheel image of setting frame number is obtained, if C>5th threshold value T5, then remind and drive
Member takes care driving.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710389936.1A CN107301379A (en) | 2017-05-27 | 2017-05-27 | A kind of steering wheel and manipulator's detection method and system based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710389936.1A CN107301379A (en) | 2017-05-27 | 2017-05-27 | A kind of steering wheel and manipulator's detection method and system based on machine vision |
Publications (1)
Publication Number | Publication Date |
---|---|
CN107301379A true CN107301379A (en) | 2017-10-27 |
Family
ID=60137201
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710389936.1A Pending CN107301379A (en) | 2017-05-27 | 2017-05-27 | A kind of steering wheel and manipulator's detection method and system based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107301379A (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108082080A (en) * | 2017-12-06 | 2018-05-29 | 东风商用车有限公司 | Vehicle control system and control method thereof |
CN111797670A (en) * | 2019-04-04 | 2020-10-20 | Aptiv技术有限公司 | Method and device for determining whether a hand cooperates with a manual steering element of a vehicle |
CN113076874A (en) * | 2021-04-01 | 2021-07-06 | 安徽嘻哈网络技术有限公司 | Steering wheel angle detection system |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102263937A (en) * | 2011-07-26 | 2011-11-30 | 华南理工大学 | Driver's driving behavior monitoring device and monitoring method based on video detection |
CN104276080A (en) * | 2014-10-16 | 2015-01-14 | 北京航空航天大学 | Bus driver hand-off-steering-wheel detection warning system and warning method |
CN104899548A (en) * | 2015-04-09 | 2015-09-09 | 北京工商大学 | Video detection method for number of operation hands on steering wheel |
-
2017
- 2017-05-27 CN CN201710389936.1A patent/CN107301379A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN102263937A (en) * | 2011-07-26 | 2011-11-30 | 华南理工大学 | Driver's driving behavior monitoring device and monitoring method based on video detection |
CN104276080A (en) * | 2014-10-16 | 2015-01-14 | 北京航空航天大学 | Bus driver hand-off-steering-wheel detection warning system and warning method |
CN104899548A (en) * | 2015-04-09 | 2015-09-09 | 北京工商大学 | Video detection method for number of operation hands on steering wheel |
Non-Patent Citations (3)
Title |
---|
ANDREY RESHETNEV等: "Automobile Driving Interface Based on Hand Trajectory Recognition", 《2006 INTERNATIONAL FORUM ON STRATEGIC TECHNOLOGY》 * |
李良福等: "一种基于随机Hough变换的椭圆检测算法研究", 《模式识别与人工智能》 * |
郭克友等: "一种方向盘上操作手数目检测算法", 《计算机工程与应用》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108082080A (en) * | 2017-12-06 | 2018-05-29 | 东风商用车有限公司 | Vehicle control system and control method thereof |
CN111797670A (en) * | 2019-04-04 | 2020-10-20 | Aptiv技术有限公司 | Method and device for determining whether a hand cooperates with a manual steering element of a vehicle |
CN113076874A (en) * | 2021-04-01 | 2021-07-06 | 安徽嘻哈网络技术有限公司 | Steering wheel angle detection system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US9070023B2 (en) | System and method of alerting a driver that visual perception of pedestrian may be difficult | |
US8983123B2 (en) | Target-object distance measuring device and vehicle mounted with the device | |
CN104408460B (en) | A kind of lane detection and tracking detection method | |
CN104129389B (en) | Method for effectively judging and recognizing vehicle travelling conditions and device thereof | |
JP3987048B2 (en) | Vehicle periphery monitoring device | |
CN106598221A (en) | Eye key point detection-based 3D sight line direction estimation method | |
CN107292251B (en) | Driver fatigue detection method and system based on human eye state | |
CN110987189B (en) | Method, system and device for detecting temperature of target object | |
US9158738B2 (en) | Apparatus for monitoring vicinity of a vehicle | |
CN106846734A (en) | A kind of fatigue driving detection device and method | |
CN106802134A (en) | A kind of line-structured light machine vision tire wear measurement apparatus | |
CN107301379A (en) | A kind of steering wheel and manipulator's detection method and system based on machine vision | |
CN104123532B (en) | Target object detection and target object quantity confirming method and device | |
CN109886205B (en) | Real-time safety belt monitoring method and system | |
CN109145696B (en) | Old people falling detection method and system based on deep learning | |
CN112381870B (en) | Binocular vision-based ship identification and navigational speed measurement system and method | |
CN111079589B (en) | Automatic height detection method based on depth camera shooting and height threshold value pixel calibration | |
CN103308523B (en) | Method for detecting multi-scale bottleneck defects, and device for achieving method | |
US9298995B2 (en) | Red-eye determination device | |
JPWO2013008305A1 (en) | 瞼 Detection device | |
US8174578B2 (en) | Vehicle periphery monitoring device | |
CN104535006B (en) | A kind of bottle cap gap width evaluation method of utilization transmission-type illumination imaging systems | |
CN111067530A (en) | Subway passenger height automatic detection method and system based on depth camera shooting | |
WO2013008304A1 (en) | Eyelid detection device | |
KR20150010248A (en) | Method and apparatus for surveillance by using 3-dimension image data |
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 |
Application publication date: 20171027 |
|
RJ01 | Rejection of invention patent application after publication |