CN105991913A - Method for location of feature object based on machine vision - Google Patents
Method for location of feature object based on machine vision Download PDFInfo
- Publication number
- CN105991913A CN105991913A CN201510053491.0A CN201510053491A CN105991913A CN 105991913 A CN105991913 A CN 105991913A CN 201510053491 A CN201510053491 A CN 201510053491A CN 105991913 A CN105991913 A CN 105991913A
- Authority
- CN
- China
- Prior art keywords
- inner circle
- connected domain
- machine vision
- petal
- region
- 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.)
- Granted
Links
Landscapes
- Image Analysis (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention provides a method for location of a feature object based on machine vision. The image of a feature object is provided by a machine vision system, and the method employs the image to locate the feature object, and comprises the following steps: i) employing a blob technology or a template matching technology to identify the area of the feature object from the image; and ii) obtaining the central coordinates of the area. The method can be used for locating the central coordinates of the lens of a camera module group of the central coordinates of a petal groove, and can employ the central coordinates of the lens and the central coordinates of the petal groove to calculate the angle of the petal groove.
Description
Technical field
The present invention relates to field of machine vision, particularly relate to a kind of method based on machine vision location feature object.
Background technology
Along with the fast development of automation equipment, a lot of high-precision automation equipments need to use machine vision
Stand-by motor positions.As a example by the automatic focusing process of cell-phone camera module (CCM), current automatic tune
Burnt machine is when to camera shooting module group focusing, it is necessary first to focusing handwheel inserts the petal groove of CCM camera lens accurately
In, just can carry out follow-up focus operations, but camera lens entirety size only has 6mm to 10mm, and petal
Well width, cannot accurately be determined if depending merely on motor according to different module models only between 1mm to 2mm certainly
Position, it is therefore desirable to utilize machine vision to come stand-by motor location.
Additionally, the petal groove of qualified camera module should be parallel with lens chip, but in actual production peace
In dress, have petal groove unavoidably when do not place parallel with lens chip, be now accomplished by utilizing machine to regard
The angle of feel location petal groove, thus stand-by motor carries out follow-up correction operation.Obtain utilizing machine vision
Behind position in camera view of petal groove and angular deviation, according to the physical coordinates of camera module (namely
Mechanical coordinate) and the ratio of image pixel point coordinates (i.e. pixel coordinate in the camera view of machine vision)
Coefficient, can calculate the angle that motor X, the position of Y-axis needs location and U axle need to rotate.
But, a whole set of Vision Builder for Automated Inspection (such as Keyence, Omron etc.) the most on the market is versatility
Vision, debugs more complicated, and expensive.Accordingly, it would be desirable to develop a kind of relatively simple and knowledge of low cost
Other method.
Summary of the invention
It is an object of the present invention to provide a kind of method based on machine vision location feature object, the method
Feature object coordinate information in pixel coordinate system can be obtained, thus realize the location to feature object.
A kind of method positioning a feature object based on machine vision of offer is provided, should
Method Machine Vision Recognition system based on an automatic focusing machine, can be positioned on a camera module by the method
One feature object, it is thus achieved that this feature object coordinate information in pixel coordinate system.
A kind of method based on machine vision location feature object of offer, the party are provided
Method need not special Vision Builder for Automated Inspection, available general CCD camera, and by independently writing software
System is identified, thus reduces the holistic cost of mechanical recognition system.
Further object is that a kind of method based on machine vision location feature object of offer, in essence
Spend close in the case of, the method is debugged more conveniently and quickly than Keyemce vision, and double mutual when working
Do not disturb.
A kind of method based on machine vision location feature object of offer is provided, passes through
The method can position optical center coordinate and/or the petal groove center seat of a camera module in automatic focusing apparatus
Mark.
The method that further object is that the petal angle of the v-groove providing a kind of location one camera module, should
Method utilizes optical center coordinate and the angle of petal groove center coordinate setting petal groove of camera module.
For reaching object above, the present invention provides a kind of method based on machine vision location feature object, by one
Vision Builder for Automated Inspection provides an image of described feature object, and described method utilizes feature described in described framing
Object, said method comprising the steps of:
I) blob technology or template matching technique is utilized to identify the region of described feature object from described image；
Ii) coordinate of a specified point in described region is obtained.
Preferably, when the shape of described feature object is close to time circular, in step) in utilize blob technology
Identify that the region of described feature object comprises the following steps:
A) from described image, find out the inner circle that the round degree of approximation meets with described feature object with area connect
Territory, is described region；
B) described inner circle connected domain is extracted.
Preferably, in step a), process by described image being done binary-state threshold, it is thus achieved that several described
Connected domain, finds out the described inner circle company that the round degree of approximation and area meet with described camera lens from several described connected domains
Logical territory.
Preferably, in step b), the step of described inner circle connected domain is extracted particularly as follows: described inner circle is connected
Territory is done dilation erosion and is processed, from described image by described inner circle connected area segmentation out.
Preferably, step) in the center that described specified point is described region, namely when described feature object
Close to time circular, described specified point is the center of circle.Step) further include steps of
C) edge of described inner circle connected domain is extracted；
D) edge of described inner circle connected domain is done round matching, obtain inner circle central coordinate of circle.
Preferably, in step c), described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains
Edge to described inner circle connected domain.
Preferably, in step d), method of least square is used to do round matching.
When step) in when can not use region described in blob technology identification, in step) in utilize template
The region of feature object described in match cognization, now, step) comprise the following steps:
A) one template of described feature object is provided；
B) described image is done template matching based on edge, obtain the spy consistent with described shape of template
Levy region, the most described region.
Preferably, step A) comprise the following steps:
A1) register a pictures as reference base picture, described reference base picture is done medium filtering；
A2) region of described feature object is split from described benchmark image, as described feature
The template of object.
Preferably, in step) in, described specified point is the center of described characteristic area.
The present invention also provides for a kind of method of camera lens positioning a camera module based on machine vision, a machine regard
Vision system provides an image of described camera module, and described method utilizes camera lens described in described framing, described
Method comprises the following steps:
H) from described image, the inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens；
I) described inner circle connected domain is extracted；
J) central coordinate of circle (x of described inner circle connected domain is obtained0,y0)。
The present invention also provides for a kind of method of petal groove positioning a camera module based on machine vision, by a machine
Visual system provides an image of described camera module, and described method utilizes petal groove described in described framing,
Said method comprising the steps of:
H) template of one petal groove is provided；
I) described image is done template matching based on edge, obtain matching with described petal slot template
One characteristic area；
J) centre coordinate (x of described characteristic area is obtained1,y1)。
The present invention also provides for a kind of method of petal angle of the v-groove positioning a camera module based on machine vision, by one
Vision Builder for Automated Inspection provides an image of described camera module, and described method utilizes petal described in described framing
The angle of groove, said method comprising the steps of:
X) from described image, obtain the centre coordinate (x of the camera lens of described camera module0,y0)；
Y) from described image, obtain the centre coordinate (x of the petal groove of described camera module1,y1)；
Z) angle of described petal groove is calculated.
Preferably, formula is utilized: calculate the angle of described petal groove.
Accompanying drawing explanation
Fig. 1 is the flow chart of the method based on machine vision location feature object of the present invention.
Fig. 2 is the flow process of the method for the petal angle of the v-groove positioning a camera module based on machine vision of the present invention
Figure.
Fig. 3 is method identification one camera module based on machine vision location feature object utilizing the present invention
The schematic diagram of the petal angle of the v-groove.
Detailed description of the invention
Hereinafter describe and be used for disclosing the present invention so that those skilled in the art are capable of the present invention.In below describing
Preferred embodiment be only used as citing, it may occur to persons skilled in the art that other obvious modification.With
The ultimate principle of the present invention defined in lower description can apply to other embodiments, deformation program, improvement side
Case, equivalent and the other technologies scheme without departing from the spirit and scope of the present invention.
The present invention relates to the Machine Vision Recognition system of an automatic focusing machine, described Machine Vision Recognition system is used for
A feature object of a camera module is positioned when automatic focusing.Described Machine Vision Recognition system provides one
CCD camera and a computer system.Described CCD camera is for obtaining the image of described feature object, institute
The pixel stating CCD camera selects according to the required precision of described Machine Vision Recognition system.Preferably,
Described CCD camera can be the CCD camera of black and white.Described computer system is used for processing described CCD phase
The described image that machine obtains, i.e. identifies described feature object from described image, and calculates described feature object
Coordinate and/or angle.Present invention generally provides one and utilize image described in described computer system processor, and obtain
The coordinate of described feature object in described image and/or the method for angle.
When utilizing described computer system to position described feature object, need to set up one based on machine vision
Pixel coordinate system (XP,YP).The coordinate of the described feature object obtained is that the pixel in pixel coordinate system is sat
Mark.Want to obtain the mechanical coordinate of described feature object, by vision proportionality coefficient, pixel coordinate can be converted to
Mechanical coordinate.It addition, the central shaft that the angle of described feature object refers to described feature object is sat with described pixel
Angle between the coordinate axes of mark system.
Therefore, described in the present invention based on machine vision position described feature object be feature object described in specific bit
Position in pixel coordinate system.
When positioning the described feature object in described image, the shape according to described feature object is needed to select to close
Suitable method positions.If the shape of described feature object is close to time circular, mainly utilize blob technology
Position.If the shape of described feature object can not pass through blob technological orientation, then use template matching
Method position.
The present invention provides a kind of method positioning a feature object based on machine vision, a Vision Builder for Automated Inspection carry
For an image of described feature object, described method utilizes feature object described in described framing, described method
Comprise the following steps:
I) blob technology or template matching technique is utilized to identify the region of described feature object from described image；
Ii) coordinate of a specified point in described region is obtained.
It is noted that the described specified point in described region is a bit that position relationship determines with described region,
Rather than a random point in described region.Such as, when described region is circular, described specified point can be
The center of circle in described region.
Preferably, when the shape of described feature object is close to time circular, in step) in utilize blob technology
Identify the region of described feature object.
The method in the region of feature object described in blob technology identification is utilized to further include steps of
A) from described image, find out the inner circle that the round degree of approximation meets with described feature object with area connect
Territory, is described region；
B) described inner circle connected domain is extracted.
Preferably, in step a), process by described image being done binary-state threshold, it is thus achieved that several described
Connected domain, finds out the described inner circle company that the round degree of approximation and area meet with described camera lens from several described connected domains
Logical territory.
Preferably, in step b), the step in described region is extracted particularly as follows: do swollen to described inner circle connected domain
Swollen corrosion treatmentCorrosion Science, from described image by described inner circle connected area segmentation out.
Preferably, step) in the center that described specified point is described region, namely when described feature object
Close to time circular, described specified point is the center of circle.When utilizing the region of feature object described in blob technology identification,
Step) comprise the following steps:
C) edge of described inner circle connected domain is extracted；
D) edge of described inner circle connected domain is done round matching, obtain inner circle central coordinate of circle.
Preferably, in step c), described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains
Edge to described inner circle connected domain.
Preferably, in step d), method of least square is used to do round matching.
When described feature object cannot with blob technology identification time, in step) in utilize the side of template matching
The region of feature object described in method identification.
The method in the region of feature object described in template matching identification is utilized to further include steps of
A) one template of described feature object is provided；
B) described image is done template matching based on edge, obtain the spy consistent with described shape of template
Levy region, the most described region.
Preferably, step A) comprise the following steps:
A1) register a pictures as reference base picture, described reference base picture is done medium filtering；
A2) region of described feature object is split from described benchmark image, as described feature
The template of object.
Preferably, when in step) in use template matching method identification described in region time, in step)
In, described specified point is the center of described characteristic area.
The described method that the present invention provides can be known when utilizing an automatic focusing machine to focus a camera module
The position of the petal groove of the most described camera module upper surface and angle.Due to the mistake in the actual production course of processing
Difference, position and the angle of the petal groove of different camera modules are not necessarily identical, it is therefore desirable to utilize this
The described petal groove of each camera module is positioned by bright described method.
It is noted that described petal groove is several grooves at edge, described camera lens upper surface, automatic focusing
Each pawl of the focusing handwheel of machine is corresponding with a described petal groove respectively, and each pawl of described focusing handwheel inserts
In each described petal groove, the movement of described camera lens can be controlled, thus realize the adjustment to described lens focus.
The shape of the most each described petal groove is identical and is mutually formed equidistantly in the edge of described camera lens,
Namely angle between each described petal groove is identical.It is difficult to ensure that each described shooting mould during installing
The position of the described petal groove of group is all consistent, therefore when described focusing handwheel is properly inserted in described petal groove,
Need to accurately identify the angle of described petal groove, owing to the angle between each described petal groove is identical, as long as therefore
Identify that the angle of one of them described petal groove i.e. may determine that the angle of each described petal groove, it is achieved thereby that right
The location of each described petal groove.
As shown in Fig. 3 A, 3B, for the schematic diagram of an embodiment of the camera lens 11 of camera module 10, in figure
Showing that the outward flange of described camera lens 11 forms several groove 12, described groove 12 is described petal groove 12.
The angle of each petal groove 12 shown in the angle of each described petal groove 12 shown in Fig. 3 A and Fig. 3 B is not
With.When determining the angle of described petal groove 12, with the central coordinate of circle of described camera lens 11 as reference point, by institute
State the angle of line in the center of petal groove 12 and the center of circle of described camera lens 11 as the angle of described petal groove.
Therefore identify that the angle of described petal groove 12 translates into and identify the center of circle of described camera lens 11 and described petal groove
The center of 12.
In a preferred embodiment, the described feature object of the described method identification of the present invention is a camera module
Camera lens, owing to the shape of described camera lens is close to circular, therefore identify that described camera lens uses blob technology.
The present invention provides a kind of method of camera lens positioning a camera module based on machine vision, by a machine vision
System provides an image of described camera module, and described method utilizes camera lens described in described framing, described side
Method comprises the following steps:
H) from described image, the inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens；
I) described inner circle connected domain is extracted；
J) central coordinate of circle (x of described inner circle connected domain is obtained0,y0)。
Preferably, what step h) was concrete is: process by described image does binary-state threshold, it is thus achieved that several
Described connected domain, from several described connected domains, find out that the round degree of approximation and area and described camera lens meets described in
Circle connected domain.
Preferably, step i) extracts described inner circle connected domain to comprise the following steps:
I1) described inner circle connected domain being done dilation erosion process, coefficient is several pixels；
I2) by expanded that region is subtracted each other with the region corroded, then a width of 5 of lines can be obtained
The round region of pixel；
I3) then described round region is split from described image, be described inner circle connected domain.
Preferably, step j) comprises the following steps:
J 1) extract inner circle connected domain edge；
J2) edge of described inner circle connected domain is done round matching, obtain central coordinate of circle (x0,y0)。
Preferably, in step j 1) in, described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains
Edge to described inner circle connected domain.
Preferably, in step j2) in, use method of least square to do round matching.
Preferably, in step j2) in, the described central coordinate of circle of sub-pix point precision can be obtained.
In a further advantageous embodiment, described feature object is the petal groove of described camera module, due to described
The shape of petal groove is special, needs the method using template matching to be identified.
The present invention provides a kind of method of petal groove positioning a camera module based on machine vision, a machine regard
Vision system provides an image of described camera module, and described method utilizes petal groove described in described framing, institute
The method of stating comprises the following steps:
H) template of one petal groove is provided；
I) described image is done template matching based on edge, obtain matching with described petal slot template
One characteristic area；
J) centre coordinate (x of described characteristic area is obtained1,y1)。
Preferably, step H) comprise the following steps:
H1) register a pictures as reference base picture, described reference base picture is done medium filtering；
H2) region of described petal groove is split from described benchmark image, as described petal groove
Template.
It is noted that a pictures of the described reference base picture preferable described camera module that is image quality,
With image Segmentation Technology from described reference base picture by petal groove region segmentation out, as template.Utilizing machine
During the petal groove of the camera module on device visual identity production line, just mate on the basis of described template,
Petal groove region can be just identified as with the region that described module mates
Preferably, described characteristic area is a rectangular area, the length and width of described characteristic area and described petal channel mould
The length and width of plate are consistent.
For described camera module, described petal groove center is described petal groove with the line of described optical center
Central shaft, hence with the coordinate of coordinate and the described petal groove center of described optical center, institute can be calculated
State the angle between central shaft and the coordinate axes of petal groove, namely the angle of described petal groove
The present invention also provides for a kind of method of petal angle of the v-groove positioning a camera module based on machine vision, by one
Vision Builder for Automated Inspection provides an image of described camera module, and described method utilizes petal described in described framing
The angle of groove, said method comprising the steps of:
X) from described image, obtain the centre coordinate (x of the camera lens of described camera module0,y0)；
Y) from described image, obtain the centre coordinate (x of the petal groove of described camera module1,y1)；
Z) angle of described petal groove is calculated.
It is noted that step X) and step Y) order can change.
Preferably, step X) comprise the following steps:
X1) from described image, the inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens；
X2) described inner circle connected domain is extracted；
X3) central coordinate of circle (x of described inner circle connected domain is obtained0,y0)。
Preferably, in step X1) in, process by described image being done binary-state threshold, it is thus achieved that Shuo Gesuo
State connected domain, from several described connected domains, find out the described inner circle that the round degree of approximation and area meet with described camera lens
Connected domain.
Preferably, step X2) in extract described inner circle connected domain comprise the following steps:
X21) described inner circle connected domain being done dilation erosion process, coefficient is several pixels；
X22) by expanded that region is subtracted each other with the region corroded, then lines can be obtained a width of
The round region of 5 pixels；
X23) described round region is split from described image, be described inner circle connected domain.
Preferably, step X3) comprise the following steps:
X31) edge of described inner circle connected domain is extracted；
X32) edge of described inner circle connected domain is done round matching, obtain central coordinate of circle (x0,y0)。
Preferably, in step X31) in, described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus
Obtain the edge of described inner circle connected domain.
Preferably, in step X32) in, use method of least square to do round matching.
Preferably, in step X32) in, the described central coordinate of circle of sub-pix point precision can be obtained.
Preferably, step Y) comprise the following steps:
Y1) template of one petal groove is provided；
Y2) described image is done template matching based on edge, obtain and the template phase of described petal groove
One characteristic area of coupling；
Y3) centre coordinate (x of described characteristic area is obtained1,y1)。
Preferably, step Y1) comprise the following steps:
Y11) register a pictures as reference base picture, described reference base picture is done medium filtering；
Y12) region of described petal groove is split from described benchmark image, as described petal
The template of groove.
Preferably, step Z) in, utilize formula: calculate the angle of described petal groove.
It should be understood by those skilled in the art that the embodiments of the invention shown in foregoing description and accompanying drawing are only used as
Illustrate and be not limiting as the present invention.The purpose of the present invention is completely and be effectively realized.The function of the present invention and
Structural principle is shown the most in an embodiment and illustrates, without departing under described principle, and embodiments of the present invention
Can there be any deformation or amendment.
Claims (29)
1. a method based on machine vision location feature object, is provided described spy by a Vision Builder for Automated Inspection
Levying an image of object, described method utilizes feature object described in described framing, it is characterised in that described
Method comprises the following steps:
I) blob technology or template matching technique is utilized to identify the region of described feature object from described image；
Ii) coordinate of a specified point in described region is obtained.
2. method based on machine vision location feature object as claimed in claim 1, in step) institute
State the center that specified point is described region.
3. method based on machine vision location feature object as claimed in claim 2, in step) in
The region of feature object described in blob technology identification is utilized to comprise the following steps:
A) from described image, find out the inner circle that the round degree of approximation meets with described feature object with area connect
Territory, described inner circle connected domain is described region；
B) described inner circle connected domain is extracted.
4. method based on machine vision location feature object as claimed in claim 3, in step a),
Process by described image being done binary-state threshold, it is thus achieved that several described connected domains, from several described connected domains
Find out the described inner circle connected domain that the round degree of approximation and area meet with described camera lens.
5. the method based on machine vision location feature object as described in claim 3 or 4, in step b)
The step of middle extraction described inner circle connected domain particularly as follows: do dilation erosion and process, from institute to described inner circle connected domain
State in image by described inner circle connected area segmentation out.
6. method based on machine vision location feature object as claimed in claim 3, step) include
Following steps:
C) edge of described inner circle connected domain is extracted；
D) edge of described inner circle connected domain is done round matching, obtain inner circle central coordinate of circle.
7. method based on machine vision location feature object as claimed in claim 5, step) include
Following steps:
C) edge of described inner circle connected domain is extracted；
D) edge of described inner circle connected domain is done round matching, obtain inner circle central coordinate of circle.
Method based on machine vision location feature object the most as claimed in claims 6 or 7, in step c)
In, described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains the edge of described inner circle connected domain.
Method based on machine vision location feature object the most as claimed in claims 6 or 7, in step d)
In, use method of least square to do round matching.
10. method based on machine vision location feature object as claimed in claim 1 or 2, in step
The region of feature object described in template matching identification is utilized to comprise the following steps in):
A) one template of described feature object is provided；
B) described image is done template matching based on edge, obtain the spy consistent with described shape of template
Levy region, the most described region.
11. methods based on machine vision location feature object as claimed in claim 10, step A) bag
Include following steps:
A1) register a pictures as reference base picture, described reference base picture is done medium filtering；
A2) region of described feature object is split from described benchmark image, as described feature
The template of object.
The method of 12. 1 kinds of camera lenses based on machine vision positioning shooting module, is carried by a Vision Builder for Automated Inspection
For an image of described camera module, described method utilizes camera lens described in described framing, it is characterised in that
Said method comprising the steps of:
H) from described image, the inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens；
I) described inner circle connected domain is extracted；
J) central coordinate of circle (x of described inner circle connected domain is obtained0,y0)。
The method of 13. camera lenses based on machine vision positioning shooting module as claimed in claim 12, wherein
What step h) was concrete is: process by described image does binary-state threshold, it is thus achieved that several described connected domains,
The described inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens from several described connected domains.
The method of 14. camera lenses based on machine vision positioning shooting module as described in claim 12 or 13,
Step i) wherein extracts described inner circle connected domain comprise the following steps:
I1) described inner circle connected domain being done dilation erosion process, coefficient is several pixels；
I2) by expanded that region is subtracted each other with the region corroded, then a width of 5 of lines can be obtained
The round region of pixel；
I3) then described round region is split from described image, be described inner circle connected domain.
The method of 15. camera lenses based on machine vision positioning shooting module as claimed in claim 14, wherein
Step j) comprises the following steps:
J1) edge of described inner circle connected domain is extracted；
J2) edge of described inner circle connected domain is done round matching, obtain central coordinate of circle (x0,y0)。
The method of 16. camera lenses based on machine vision positioning shooting module as claimed in claim 15, in step
Rapid j1) in, described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains described inner circle connected domain
Edge.
The method of 17. camera lenses based on machine vision positioning shooting module as claimed in claim 15, in step
Rapid j2) in, use method of least square to do round matching.
The method of 18. 1 kinds of petal grooves based on machine vision positioning shooting module, is carried by a Vision Builder for Automated Inspection
For an image of described camera module, described method utilizes petal groove described in described framing, it is characterised in that
Said method comprising the steps of:
H) template of one petal groove is provided；
I) described image doing template matching based on edge, obtain matching with described petal slot template one is special
Levy region；
J) centre coordinate (x of described characteristic area is obtained1,y1)。
The method of 19. petal grooves based on machine vision positioning shooting module as claimed in claim 18, its
Middle step H) comprise the following steps:
H1) register a pictures as reference base picture, described reference base picture is done medium filtering；
H2) region of described petal groove is split from described benchmark image, as described petal groove
Template.
The method of 20. 1 kinds of petal angle of the v-grooves based on machine vision positioning shooting module, by a machine vision system
System provides an image of described camera module, and described method utilizes the angle of petal groove described in described framing,
It is characterized in that, said method comprising the steps of:
X) from described image, obtain the centre coordinate (x of the camera lens of described camera module0,y0)；
Y) from described image, obtain the centre coordinate (x of the petal groove of described camera module1,y1)；
Z) angle of described petal groove is calculated.
21. the method for the petal angle of the v-groove based on machine vision positioning shooting module as claimed in claim 20,
Utilize formula: calculate the angle of described petal groove.
The 22. petal angle of the v-grooves based on machine vision positioning shooting module as described in claim 20 or 21
Method, wherein step X) comprise the following steps:
X1) from described image, the inner circle connected domain that the round degree of approximation and area meet is found out with described camera lens；
X2) described inner circle connected domain is extracted；
X3) central coordinate of circle (x of described inner circle connected domain is obtained0,y0)。
The method of the 23. petal angle of the v-grooves based on machine vision positioning shooting module as claimed in claim 22,
In step X1) in, process by described image being done binary-state threshold, it is thus achieved that several described connected domains, from
Several described connected domains are found out the described inner circle connected domain that the round degree of approximation and area meet with described camera lens.
The 24. petal angle of the v-grooves based on machine vision positioning shooting module as described in claim 22 or 23
Method, wherein step X2) in extract described inner circle connected domain comprise the following steps:
X21) described inner circle connected domain being done dilation erosion process, coefficient is several pixels；
X22) by expanded that region is subtracted each other with the region corroded, then lines a width of 5 can be obtained
The round region of individual pixel；
X23) described round region is split from described image, be described inner circle connected domain.
The method of the 25. petal angle of the v-grooves based on machine vision positioning shooting module as claimed in claim 24,
Wherein step X3) comprise the following steps:
X31) edge of described inner circle connected domain is extracted；
X32) edge of described inner circle connected domain is done round matching, obtain central coordinate of circle (x0,y0)。
The method of the 26. petal angle of the v-grooves based on machine vision positioning shooting module as claimed in claim 25,
In step X31) in, described inner circle connected domain is done the Threshold segmentation of sub-pix point, thus obtains described inner circle
The edge of connected domain.
The side of the 27. petal angle of the v-grooves positioning a camera module based on machine vision as claimed in claim 26
Method, in step X32) in, use method of least square to do round matching.
28. as arbitrary in claim 20,21 or 27 as described in petal based on machine vision positioning shooting module
The method of the angle of the v-groove, wherein step Y) comprise the following steps:
Y1) template of one petal groove is provided；
Y2) described image is done template matching based on edge, obtain matching with the template of described petal groove
A characteristic area；
Y3) centre coordinate (x of described characteristic area is obtained1,y1)。
The method of the 29. petal angle of the v-grooves based on machine vision positioning shooting module as claimed in claim 28,
Wherein step Y1) comprise the following steps:
Y11) register a pictures as reference base picture, described reference base picture is done medium filtering；
Y12) region of described petal groove is split from described benchmark image, as described petal groove
Template.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510053491.0A CN105991913B (en) | 2015-02-03 | 2015-02-03 | Method for positioning petal slot angle of camera module based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510053491.0A CN105991913B (en) | 2015-02-03 | 2015-02-03 | Method for positioning petal slot angle of camera module based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105991913A true CN105991913A (en) | 2016-10-05 |
| CN105991913B CN105991913B (en) | 2019-12-31 |
Family
ID=57037272
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510053491.0A Active CN105991913B (en) | 2015-02-03 | 2015-02-03 | Method for positioning petal slot angle of camera module based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105991913B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107271446A (en) * | 2017-07-26 | 2017-10-20 | 江南大学 | A kind of visible detection method of engine timing gear room tree lace |
| CN107818582A (en) * | 2017-11-22 | 2018-03-20 | 航天科工智能机器人有限责任公司 | The detecting system and method for full-automatic identification carrier band reel angle and bar code information |
| CN107860773A (en) * | 2017-11-06 | 2018-03-30 | 凌云光技术集团有限责任公司 | Automatic optical detecting system and its bearing calibration for PCB |
| CN108109173A (en) * | 2016-11-25 | 2018-06-01 | 宁波舜宇光电信息有限公司 | Vision positioning method, camera system and automation equipment |
| CN108548654A (en) * | 2018-02-07 | 2018-09-18 | 深圳市锐光科技有限公司 | Camera defect inspection method based on machine vision |
| CN111275756A (en) * | 2020-02-20 | 2020-06-12 | 珠海格力智能装备有限公司 | Spool positioning method and apparatus |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102721364A (en) * | 2011-03-30 | 2012-10-10 | 比亚迪股份有限公司 | Positioning method and positioning device for workpiece |
| CN102799865A (en) * | 2012-07-03 | 2012-11-28 | 天津大学 | Image boundary polar-coordinated discrete sequence based angle identification method |
| US20130121598A1 (en) * | 2011-11-15 | 2013-05-16 | Futurewei Technologies, Inc. | System and Method for Randomized Point Set Geometry Verification for Image Identification |
| CN103729632A (en) * | 2014-01-22 | 2014-04-16 | 哈尔滨工业大学 | Round Mark point positioning method based on connected region filtering |
| CN103792760A (en) * | 2014-01-14 | 2014-05-14 | 宁波舜宇光电信息有限公司 | Localization computing and position correction method for automatic focusing executive mechanism |
-
2015
- 2015-02-03 CN CN201510053491.0A patent/CN105991913B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102721364A (en) * | 2011-03-30 | 2012-10-10 | 比亚迪股份有限公司 | Positioning method and positioning device for workpiece |
| US20130121598A1 (en) * | 2011-11-15 | 2013-05-16 | Futurewei Technologies, Inc. | System and Method for Randomized Point Set Geometry Verification for Image Identification |
| CN102799865A (en) * | 2012-07-03 | 2012-11-28 | 天津大学 | Image boundary polar-coordinated discrete sequence based angle identification method |
| CN103792760A (en) * | 2014-01-14 | 2014-05-14 | 宁波舜宇光电信息有限公司 | Localization computing and position correction method for automatic focusing executive mechanism |
| CN103729632A (en) * | 2014-01-22 | 2014-04-16 | 哈尔滨工业大学 | Round Mark point positioning method based on connected region filtering |
Non-Patent Citations (1)
| Title |
|---|
| 李婷等: "《基于机器视觉的圆定位技术研究》", 《计算机工程与应用》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108109173A (en) * | 2016-11-25 | 2018-06-01 | 宁波舜宇光电信息有限公司 | Vision positioning method, camera system and automation equipment |
| CN108109173B (en) * | 2016-11-25 | 2022-06-28 | 宁波舜宇光电信息有限公司 | Visual positioning method, camera system and automation equipment |
| CN107271446A (en) * | 2017-07-26 | 2017-10-20 | 江南大学 | A kind of visible detection method of engine timing gear room tree lace |
| CN107271446B (en) * | 2017-07-26 | 2020-01-10 | 江南大学 | Visual detection method for rubber wire of timing gear chamber of engine |
| CN107860773A (en) * | 2017-11-06 | 2018-03-30 | 凌云光技术集团有限责任公司 | Automatic optical detecting system and its bearing calibration for PCB |
| CN107860773B (en) * | 2017-11-06 | 2021-08-03 | 凌云光技术股份有限公司 | Automatic optical detection system for PCB and correction method thereof |
| CN107818582A (en) * | 2017-11-22 | 2018-03-20 | 航天科工智能机器人有限责任公司 | The detecting system and method for full-automatic identification carrier band reel angle and bar code information |
| CN107818582B (en) * | 2017-11-22 | 2020-02-28 | 航天科工智能机器人有限责任公司 | Detection system and method for full-automatically identifying carrier tape reel angle and bar code information |
| CN108548654A (en) * | 2018-02-07 | 2018-09-18 | 深圳市锐光科技有限公司 | Camera defect inspection method based on machine vision |
| CN111275756A (en) * | 2020-02-20 | 2020-06-12 | 珠海格力智能装备有限公司 | Spool positioning method and apparatus |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105991913B (en) | 2019-12-31 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105991913A (en) | Method for location of feature object based on machine vision | |
| CN107424160B (en) | System and method for finding image central line by visual system | |
| JP6507730B2 (en) | Coordinate transformation parameter determination device, coordinate transformation parameter determination method, and computer program for coordinate transformation parameter determination | |
| CN105701492B (en) | A kind of machine vision recognition system and its implementation | |
| WO2017092631A1 (en) | Image distortion correction method for fisheye image, and calibration method for fisheye camera | |
| KR102649038B1 (en) | System and method for finding lines in an image with a vision system | |
| JP6899189B2 (en) | Systems and methods for efficiently scoring probes in images with a vision system | |
| US9491452B2 (en) | Camera calibration | |
| WO2014045508A1 (en) | Inspection device, inspection method, and inspection program | |
| JP2011043969A (en) | Method for extracting image feature point | |
| Daftry et al. | Flexible and User-Centric Camera Calibration using Planar Fiducial Markers. | |
| Li et al. | A practical comparison between Zhang's and Tsai's calibration approaches | |
| CN113538583A (en) | Method for accurately positioning position of workpiece on machine tool and vision system | |
| Ding et al. | A robust detection method of control points for calibration and measurement with defocused images | |
| Redzuwan et al. | Affine versus projective transformation for SIFT and RANSAC image matching methods | |
| US10252417B2 (en) | Information processing apparatus, method of controlling information processing apparatus, and storage medium | |
| CN110519508B (en) | Method for identifying characteristic object in automatic aligning process of camera module | |
| CN106488204B (en) | Have the depth camera of self-aligning and self-aligning method | |
| CN104424601B (en) | Centering assembly method and device for special-shaped body assembly parts | |
| CN112509058B (en) | External parameter calculating method, device, electronic equipment and storage medium | |
| KR101923581B1 (en) | Normal vector extraction apparatus and method thereof based on stereo vision for hull underwater inspection using underwater robot | |
| KR101559941B1 (en) | A system for detecting fault of print pattern | |
| CN113407030B (en) | Visual positioning method, related device, equipment and storage medium | |
| CN115239816A (en) | Camera calibration method, system, electronic device and storage medium | |
| US9721371B2 (en) | Systems and methods for stitching metallographic and stereoscopic images |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |