CN109215016A - A kind of recognition positioning method of coding maker - Google Patents
A kind of recognition positioning method of coding maker Download PDFInfo
- Publication number
- CN109215016A CN109215016A CN201810875464.5A CN201810875464A CN109215016A CN 109215016 A CN109215016 A CN 109215016A CN 201810875464 A CN201810875464 A CN 201810875464A CN 109215016 A CN109215016 A CN 109215016A
- Authority
- CN
- China
- Prior art keywords
- point
- coding maker
- coding
- image
- positioning method
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T9/00—Image coding
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of recognition positioning method of coding maker, include the following steps: to obtain the image with coding maker；The image of acquisition is pre-processed, binary image is obtained；Image Segmentation Methods Based on Features is carried out to binary image, isolates border circular areas, feature templates matching is carried out to Uncircular field, the shape for determining whether to have default feature；Obtain the skeleton of the characteristic area where the shape with default feature；According to the coordinate of skeleton and the design parameter of default feature, affine transformation matrix is solved；Then inverse affine transformation is carried out, the image coordinate of dot is subjected to inverse transformation, is transformed under coding maker design coordinate system；Dot corresponding with the matched all mark points of design template on image is found out, and coordinate matching decoding is carried out to other dots, obtains encoded point；The position according to corresponding to encoded point obtains coding maker and uniquely encodes.The recognition positioning method improves feature recognition efficiency and coding calculation accuracy for dynamic measurement request.
Description
Technical field
The present invention relates to close-shot photography measure technique fields, in particular to a kind of identification positioning side of coding maker
Method.
Background technique
In close-range photogrammetry application field, the enlargement of measurement object structure size, curved surface complication, test object dynamic
Change has become development trend.Due to camera resolution ratio limitation, visual field block, move cause the factors such as target offset influence, to big
When coarse scale structures in field range carry out dynamic measurement, it is special that body structure surface often lacks the high texture of clear and legible knowledge and magnanimity
Sign is not easy to directly quickly and accurately extract the characteristic information of body structure surface, is not able to satisfy test request.Currently, generalling use
The mode that cooperation mark is manually laid in measured structure generates target point, for identification, tracks, and therefore, has unique encodings
The design and application of the coding maker of value are by extensive research and concern.
Common coding maker mainly has three classes: point profile, ring-like and without obvious geometrical characteristic the one kind of concentric circles
Coding maker.Point-distributed coded marks are to design coordinate system by introducing, and constitute number according to the distribution put in the plane
Coding, belongs to absolute coding, can generate greater number of coding, but decodes recognizer complicated and time consumption, for example, in point minute
In cloth type coding maker technology, application No. is the knowledges of 201010201990.7 Chinese invention patent application point-distributed coded marks
Other method, the encoded point position close-packed arrays of design, the encoded point to guarantee accurately to decode identification, in same coding maker
Cannot be adjacent two-by-two, in this way in the limited design coding space of points, the permutation and combination quantity of encoded point by biggish limitation,
It is difficult to generate the unique encodings of abundant quantity.Meanwhile the coding maker needs at least six dot, in coding maker identification, needs
6 or more dots are extracted, each circle, which is done, to be determined, therefore, it is determined that whether there is mark point, anchor point, encoded point, thus really
Recognize with the presence or absence of a complete mark, calculation amount is very big, and decoding difficulty is big.
The ring-like coding maker design structure of concentric circles is integrally in circular ring shape, by centralized positioning mark and surrounds surrounding band
Shape or dotted bits of coded composition.The type coding maker positioning accuracy is high, but group/cording quantity is limited, belongs to relative coding,
Need to determine start code position, the distortion of the fuzzy and annular code area of image easily leads to identification mistake in dynamic measurement.Same
In heart circular ring type coding maker technology, for example, the Chinese invention patent application of Publication No. CN105303224A discloses one kind
Capacity is big, coded target and its coding/decoding method of strong robustness, and the coding/decoding method is the following steps are included: S1: to the figure of acquisition
As carrying out binary conversion treatment, coded target is extracted；S2: according to the shape and area of the white connected region in coded target
Extract circular index point；S3: it is larger to other circular index points and another area to calculate separately the biggish circular index point of area
Circular index point distance, be judged to witness marker to all equal biggish circular index point of area of each circular index point distance
Point, another biggish circular index point of area are judged to beginning flag point, remaining circular index point is all determined as encoded point；S4: point
The angle that the two neighboring encoded point and position mark point that Ji Suan not light from beginning flag are formed, is denoted as α；α subtracts 1 divided by m
0 number of two encoded point midfeathers is obtained, so that the corresponding unique encodings of coded target are obtained, wherein m=360 °/
K, k are the number of bits of coded.The coding/decoding method improves the accuracy of identification to a certain degree, but it is mainly suitable for annular coding
Mode, it is not strong for profile coding mode applicability, meanwhile, carrying out dynamic measurement and coarse scale structures stitching measure
When, feature recognition efficiency and coding calculation accuracy are to be improved.
Since profile and the ring-like coding maker of concentric circles are generally using homochromy circle marker, in practical application
In the process, since tilt angle of photograph is often bigger, there is also certain distortion for camera itself, can make coding maker orbicular spot position
Relationship between setting is not sufficiently stable.For the identification for solving the problems, such as coding maker, application No. is 201310000907.3 middle promulgateds by the State Council
Bright patent application discloses the coding/decoding method of a kind of profile coloud coding mark, and the coding/decoding method is first by graph
The identification of the color and shape of each figure, primarily determines the index point of coloud coding mark, identifies positioned at coding maker as in
The index point at center, and other 5 index points are found out centered on it, it is sat according to setting of 4 template points in coding maker
Mark and its true coordinate in photographs establish affine Transform Model, find out corresponding affine transformation parameter, and calculate
The true coordinate of encoded point simplifies decoding step to obtain the coding of the coding maker, improves to coloud coding mark
The accuracy of identification, but this method improves hardware requirement, when carrying out dynamic measurement and coarse scale structures stitching measure, still
The right existing characteristics recognition efficiency problem not high with coding calculation accuracy.
Coding maker of the third class without obvious characteristic is usually that some special shapes or their combination are constituted, the method
Group/cording quantity it is limited and characteristics of image identification step is cumbersome.
Therefore, it measures for dynamic and, to the actual requirement of coding maker, how to be improved in coarse scale structures stitching measure
The technical issues of feature recognition efficiency is with encoding calculation accuracy, becoming those skilled in the art's urgent need to resolve.
Summary of the invention
In consideration of it, the invention proposes a kind of recognition positioning method of coding maker, for dynamic measurement and large scale knot
To the actual requirement of coding maker in structure stitching measure, feature recognition efficiency and coding calculation accuracy are improved, at least one.
A kind of recognition positioning method of coding maker proposed by the present invention, the coding maker include setting positioned at coding maker
Mark point, anchor point and the encoded point under coordinate system are counted, the anchor point has the default feature of difference dot, the mark point
It is dot with encoded point, the recognition positioning method includes the following steps:
S20: the image with coding maker is obtained；
S30: pre-processing the image of acquisition, and obtains the binary image comprising edge feature；
S40: carrying out Image Segmentation Methods Based on Features to binary image, isolate border circular areas, carries out feature templates to Uncircular field
Matching, the shape for determining whether to have default feature；If nothing, return step S30；
S50: the skeleton of characteristic area where there is default character shape is obtained；
S60: the design coordinate under coordinate system is designed in coding maker according to the coordinate of skeleton and default feature, is solved
Image coordinate system designs the affine transformation matrix of coordinate system to coding maker；
S70: searching for dot in the default certain contiguous range of feature, identifies dot and solves its centre coordinate, carries out
The image coordinate of dot is carried out inverse transformation by inverse affine transformation, is transformed under coding maker design coordinate system；
S80: dot corresponding with the matched all mark points of design template on image is found out, and other dots are sat
Matching and decoding is marked, the encoded radio of encoded point is obtained；
S90: the position according to corresponding to encoded point obtains coding maker and uniquely encodes.
The recognition positioning method of above-mentioned coding maker, it is preferred that step S50 includes:
S51: the single pixel skeleton of characteristic area is obtained；
S52: seeking the normal orientation of each point on single pixel skeleton, and calculates the light that the edge of feature is preset in normal orientation
Center obtains the sub-pix skeleton at edge.
The recognition positioning method of above-mentioned coding maker, it is preferred that in step s 51,
The single pixel skeleton of characteristic area is obtained using K3M sequential iteration method.
The recognition positioning method of above-mentioned coding maker, it is preferred that in step s 30, pretreatment include filtering, denoising,
One of enhancing is a variety of.
The recognition positioning method of above-mentioned coding maker, it is preferred that in step s 40, isolated by form parameter method
Border circular areas.
The recognition positioning method of above-mentioned coding maker, it is preferred that in step S80, find out on image with design template
The corresponding dot of matched all mark points, and judge that affine matrix solves correctness, if incorrect, return step S30；So
Afterwards, coordinate matching decoding is carried out to other dots by look-up table, obtains the encoded radio of encoded point.
The recognition positioning method of above-mentioned coding maker, it is preferred that in step S90, further include the seat of record location point
The respective value of mark and coding.
The recognition positioning method of above-mentioned coding maker, it is preferred that further include step S10: in quilt before step S20
Object exterior pasting coding maker is surveyed, and carries out Image Acquisition.
The recognition positioning method of above-mentioned coding maker, it is preferred that the mark point includes the first mark point, the second label
Point, third mark point, the anchor point are the crosspoint of a plurality of line segment, and the encoded point includes n point, and n is more than or equal to 1
Integer, first mark point, the second mark point, third mark point, crosspoint are located at square four apex angle
On, the first mark point, crosspoint are located on a diagonal line of square, and the origin for defining coding maker design coordinate system is to hand over
Crunode, and X-axis, the Y-axis of coding maker design coordinate system are parallel to the adjacent both sides of rectangle, encoded point is interspersed in coding maker
It designs in coordinate system.
The recognition positioning method of above-mentioned coding maker, it is preferred that the default feature includes cross.
A kind of recognition positioning method of coding maker proposed by the present invention is surveyed for dynamic measurement and coarse scale structures splicing
Feature recognition efficiency and coding calculation accuracy are improved, together by step S10-S90 to the actual requirement of coding maker in amount
When, in the preferred scheme, a kind of sub-pix skeletal extraction side is proposed based on edge-smoothing and Curvature varying minimum thought
Method can carry out accurately indicating positioning.Meanwhile the decoding of coding maker is carried out based on affine transformation theory and look-up table matching,
Mark point methods coding is simple, readily identified, and feature location precision is high, decoding algorithm fast and stable.
Detailed description of the invention
Attached drawing described herein is used merely to provide and of the invention is explained further, and constitutes a part being applied,
And it does not constitute a limitation of the invention.In the accompanying drawings:
Fig. 1 is the flow chart of the recognition positioning method for the coding maker that the specific embodiment of the invention proposes；
Fig. 2 is a kind of design diagram for coding maker that the specific embodiment of the invention proposes；
Fig. 3 is the design scheme of coding maker 3 encoded points of selection in Fig. 2.
Specific embodiment
To better understand the objects, features and advantages of the present invention, with reference to the accompanying drawing and specific real
Applying mode, the present invention is further described in detail.It should be noted that in the absence of conflict, the implementation of the application
Feature in example and embodiment can be combined with each other.
It should be noted that " first " that occurs in the present invention, " second ", " third " are only used for separator point, no elder generation
Point afterwards, less as the restriction of mark point itself；Meanwhile the "and/or" in the present invention, referring to can both be provided simultaneously with, can also
With selection one of them, such as option A and/or option b, including option A, option b, option A and option b this three kinds of situations.
As shown in Figure 1-3, a kind of recognition positioning method of coding maker proposed by the present invention, coding maker includes being located to compile
Code mark designs mark point, anchor point and the encoded point under coordinate system, and anchor point has the default feature of difference dot, such as more
The crosspoint of line segment, such as the midpoint of square, the midpoint of diamond shape etc., as long as having typically default feature, mark
Note point and encoded point are dot, and recognition positioning method includes the following steps:
S20: the image with coding maker is obtained；
In the step, the image of acquisition can be existing image, be also possible to the image regained, such as: in step
Before rapid S20, further includes step S10: in testee surface mount coding maker, and Image Acquisition is carried out, to be had
There is the image of coding maker.
S30: pre-processing the image of acquisition, and obtains the binary image comprising edge feature；
In the step, pretreatment is primarily to improve the quality of image, specifically, pretreatment may include filtering, go
It one of makes an uproar, enhance or is a variety of, denoising can be accomplished in several ways, such as Gaussian smoothing filter, to be put down
Sliding image, naturally it is also possible to denoising is filtered using adaptive wiener filter method, filtering and enhancing can also adopt
It is carried out with existing mode.Meanwhile obtain binary image can there are many mode, as Two-peak method, P parametric method, iterative method and
OTSU method etc. can specifically refer to the prior art.
S40: carrying out Image Segmentation Methods Based on Features to binary image, isolate border circular areas, carries out feature templates to Uncircular field
Matching, the shape for determining whether to have default feature；If nothing, return step S30；
Image Segmentation Methods Based on Features can use existing mode, such as Otsu dual-threshold voltage, it is preferred that can pass through form parameter method
Border circular areas is isolated, to obtain Uncircular field, is preset consequently facilitating analysing whether to exist in Uncircular field to have
The shape of feature, to carry out precisely quickly positioning.Feature templates are the template with default feature, by that will split
Uncircular field carry out feature templates matching, it is possible to determine that out with the presence or absence of the shape of default feature, simultaneously because default spy
Sign is different from mark point (dot), and quickly searching for coding maker may be implemented using the feature templates matching with default feature
Rope, and high accuracy positioning may be implemented in the coordinate solution of default feature.
Specifically, form parameter method principle is as follows: form parameter includes area, perimeter, circularity, rectangular degree and curvature
Deng.It can according to need selection multiple parameters in different Object identifyings to be screened.For example area A is used simultaneouslys, edge picture
Plain perimeter Ls, circularity ρcIt is ideal circular mark when value meets the region of threshold value simultaneously, remaining is for decision condition
The mould of default feature (such as cross form) further can be used to realize characteristic target and background separation in non-circular marked region
Plate scans for matching to Uncircular field.Formula is as follows:
S50: the skeleton of characteristic area where there is default character shape is obtained；
In the step, can using existing way obtain characteristic area skeleton, skeleton can for pixel framework or its
Its type skeleton, acquisition methods can be Medial-Axis Transformation (Medial Axis Transform, MAT), iteratively substituted algorithm etc.
Deng, it specifically can be with reference to the prior art, pixel framework also may include diversified forms, such as single pixel skeleton, sub-pix skeleton
Include: Deng, it is preferred that step S50
S51: the single pixel skeleton of characteristic area is obtained；
S52: seeking the normal orientation of each point on single pixel skeleton, and calculates the light that the edge of feature is preset in normal orientation
Center obtains the sub-pix skeleton at edge.
Likewise it is preferred that, in step s 51,
The single pixel skeleton of characteristic area is obtained using K3M sequential iteration method.
Obtaining sub-pix skeleton specifically can be as follows:
1) for the profile point p in pixel frameworki(x, y) (gray value that f (x, y) is pixel), is asked by sobel operator
Take gradient vector g [fx,fy]；
2) remember point piNormal direction angle at (x, y) isNote normal direction is (nx,
ny)；
3) after obtaining normal direction, with pi(x, y) is basic point, and the contiguous range of a pixel is chosen along normal direction, right
It carries out 10 times of amplification interpolation, and a series of p' can be obtainedi(x', y'), wherein x'=x+lnx, y'=y+lny, l ∈ (-
0.5,0.5)；
4) in unknown point p'iLeft neighborhood take known three skeletal point pi-3,pi-2,pi-1, asked pair per three adjacent points
It answers curvature to obtain t1 and t2, is based on the smooth thought of linear edge, so that Curvature varying | t2-t1 | < ε, then p'i(x', y') is new
The skeletal point P foundi；
5) all P are connectediSub-pix skeleton can be obtained.
For example, reading line segment in order for the case where default feature of anchor point is the crosspoint of a plurality of line segment
The extreme coordinates of skeleton are averaged four extreme coordinates value summations, can be obtained the accurate positioning coordinate of anchor point.
S60: the design coordinate under coordinate system is designed in coding maker according to the coordinate of pixel framework and default feature, is asked
Solve the affine transformation matrix of image coordinate system to coding maker design coordinate system；
Image coordinate system is the coordinate system where image shoot or acquisition, and affine transformation basic principle is: coding mark
Will can generate rotation, translation, scaling, distortion equiaffine transformation by photographic imagery in photo.Two-dimensional surface it is affine
Transform definition can be as follows with expression formula
X '=Ax+b, x ∈ R2 (3)
Here b=(b1, b2)T∈R2For translation matrix andIt is a nonsingular real matrix, rotation is only
The special case of affine transformation.
Affine transformation matrix A's is defined as follows:
Wherein, k is used to indicate that the scaling of image, θ are used to indicate the rotation angle of image, and a and b are respectively intended to indicate
Inclination of the image on x, the direction y, according to the various affine transformations between our available images of different parameter combinations.It takes the photograph
In shadow measurement, it is the plane domain of a very little relative to whole object that coding maker, which is pasted on testee, meets two
Dimensional plane affine transformation, therefore the transformation matrix that image coordinate system under any position designs coordinate system to coding maker can be sought
Parameter.
S70: dot is searched in the default certain contiguous range of feature, identifies dot simultaneously using the methods of ellipse fitting method
Its centre coordinate is solved, inverse affine transformation is carried out, the image coordinate of dot is subjected to inverse transformation, coding maker design is transformed into and sits
Under mark system；
By the step, the coordinate of mark point and encoded point can be transformed under coding maker design coordinate system, be convenient for
The identification of next step positions.
S80: dot corresponding with the matched all mark points of design template on image is found out, and other dots are sat
Matching and decoding is marked, the encoded radio of encoded point is obtained；
Design template is to have the template of entire coding maker (comprising mark point, anchor point and encoded point and each
The positional relationship of point), specifically, in the step, dot corresponding with the matched all mark points of design template on image is found out,
And judge that affine matrix solves correctness, if incorrect, return step S30；Then, other dots are carried out by look-up table
Coordinate matching decoding, obtains encoded point.
S90: the position according to corresponding to encoded point obtains coding maker and uniquely encodes, which can be encoded point
Serial number combination, or the serial number combines corresponding number, can also be other codings.Preferably, which can be with
Including the coordinate of record location point and the respective value of coding.That is, realizing the positioning and identification of coding maker.
The recognition positioning method of above-mentioned coding maker identifies that the coding maker of positioning can choose and meets aforementioned claim
Coding maker, specifically, as shown in Fig. 2, mark point includes the first mark point A, the second mark point B, third mark point C, positioning
Point is the crosspoint O of a plurality of line segment, and encoded point includes n point, and n is the integer more than or equal to 1, the first mark point A, the second label
Point B, third mark point C, crosspoint O are located on square four apex angle, and the first mark point A, crosspoint O are located at
On one diagonal line of square, the origin for defining coding maker design coordinate system is crosspoint O, and coding maker designs coordinate
X-axis, the Y-axis of system are parallel to the adjacent both sides of rectangle, and encoded point is interspersed to be designed in coordinate system in coding maker, different location
Encoded point is corresponding with unique coding number, and the corresponding coding number combination of each encoded point is unique encoded radio.
The recognition positioning method of above-mentioned coding maker, it is preferred that default feature includes cross, certainly can also be rice
Font, × type etc..
Since the recognition positioning method uses above-mentioned coding maker, by using the anchor point with default feature
(crosspoint of a plurality of line segment), anchor point easily identifies, and can quickly determine whether to realize there are anchor point by template matching
Fast accurate positioning, to know whether there are a coding maker, does not need to determine that (prior art is used uniformly dot one by one
Dot), calculation amount is small.And theoretically, three mark points and an encoded point (total quantity is more than or equal to 4) can realize volume
Code function, design structure is simpler, decreases the positional relationship in the quantity and decoding process of marked circle between each point
Determine, substantially reduces the resolving time；Meanwhile anchor point of the invention, when being calculated by sub-pix skeleton, it is fixed to can be improved
Position precision；The line segment length of the design template of the linear feature and selection of anchor point, can be convenient solution affine transformation matrix,
It is provided safeguard for decoding.In addition, anchor point is cross crosspoint, the fast of coding maker may be implemented using cross template matching
Speed search, and high accuracy positioning may be implemented in the coordinate solution of cross intersection point.Simultaneously because the first mark point A, the second mark point
B, third mark point C, crosspoint O are located on square four apex angle, and the first mark point A, crosspoint O are located at square
On one diagonal line of shape, which has stronger anti-interference ability to image deformation, guarantees the reliability of coding method.
Above-mentioned coding maker, encoded point number can be n=1, and 2,3 ..., the design position of encoded point can be according to need
It arranges, n is smaller, and encoded point arrangement is more sparse, and identification decoding accuracy rate is higher, but design position is just few, and n is bigger, can
With the quantity of abundant coding, it is preferred that n is greater than 2, and the line of at least two encoded points is parallel with X-axis, or/and, at least two compile
The line of code-point is parallel with Y-axis, it should be noted that herein parallel be not it is absolute parallel, due to making and encoding mark
The error that will is pasted, it is only necessary to substantially parallel.Further, each encoded point divides in coding maker design coordinate system
Row, point column arrangement.In Fig. 1,29 encoded points are illustrated, it, can be with including being parallel to 3 rows of X-axis and being parallel to 3 column of Y-axis
Understand, this is one of set-up mode, can also derive more modes, such as encoded point 18,25,36
It is a, 43 etc., including being parallel to 5 rows of X-axis and 4 column for being parallel to Y-axis etc..It will not enumerate and illustrate herein, as long as
Under design framework of the invention.
Specifically, with reference to shown in Fig. 2-3.Anchor point selects the decussation point with obvious graphic feature, mark point
Dot, the corresponding unique coding of different coding point combination are selected with encoded point.According to design principle it is found that code capacity N meets
Combinatorics rule, there is formula:
Wherein, n is the coding label sum of design, and m is the encoded point number (m >=1, integer) chosen.Work as n=29, m=
When 3, code capacity reaches 3654, and can satisfy all testing requirements substantially, (Fig. 3 (a)-(c) is the design for choosing 3 encoded points
Scheme, generation group is combined into tri- coding makers of 01-15-23,06-11-25 and 01-17-25 respectively, it is of course possible to select more kinds of
Coding maker).Criss-cross crosspoint O point designs the origin of coordinate system for coding maker, and X-axis, the Y-axis difference of coordinate system are parallel
In AB and AC, then, according to the ratio scale of design, it can be deduced that each mark point and the corresponding specified coordinate of encoded point.
Generally, recognition positioning method of the invention, for dynamic measurement and coarse scale structures stitching measure in coding
The actual requirement of mark improves feature recognition efficiency and coding calculation accuracy by step S10-S90, meanwhile, preferred
In scheme, a kind of sub-pix framework extraction method is proposed based on edge-smoothing and Curvature varying minimum thought, essence can be carried out
True mark positioning.Meanwhile the decoding of coding maker, the mark point methods are carried out based on affine transformation theory and look-up table matching
Coding is simple, readily identified, and feature location precision is high, decoding algorithm fast and stable.
These are only the preferred embodiment of the present invention, is not intended to restrict the invention, for those skilled in the art
For member, the invention may be variously modified and varied.All within the spirits and principles of the present invention, it is made it is any modification,
Equivalent replacement, improvement etc., should all be included in the protection scope of the present invention.
Claims (10)
1. a kind of recognition positioning method of coding maker, which is characterized in that the coding maker includes being located at coding maker to design
Mark point, anchor point and encoded point under coordinate system, the anchor point have difference dot default feature, the mark point and
Encoded point is dot, and the recognition positioning method includes the following steps:
S20: the image with coding maker is obtained；
S30: pre-processing the image of acquisition, and obtains the binary image comprising edge feature；
S40: carrying out Image Segmentation Methods Based on Features to binary image, isolate border circular areas, carries out feature templates matching to Uncircular field,
The shape for determining whether to have default feature；If nothing, return step S30；
S50: the skeleton of characteristic area where there is default character shape is obtained；
S60: the design coordinate under coordinate system is designed in coding maker according to the coordinate of skeleton and default feature, solves image
Coordinate system designs the affine transformation matrix of coordinate system to coding maker；
S70: searching for dot in the default certain contiguous range of feature, identifies dot and solves its centre coordinate, carries out inverse imitative
Transformation is penetrated, the image coordinate of dot is subjected to inverse transformation, is transformed under coding maker design coordinate system；
S80: dot corresponding with the matched all mark points of design template on image is found out, and coordinate is carried out to other dots
With decoding, the encoded radio of encoded point is obtained；
S90: the position according to corresponding to encoded point obtains coding maker and uniquely encodes.
2. the recognition positioning method of coding maker according to claim 1, which is characterized in that step S50 includes:
S51: the single pixel skeleton of characteristic area is obtained；
S52: seeking the normal orientation of each point on single pixel skeleton, and calculates in the optics for presetting the edge of feature in normal orientation
The heart obtains the sub-pix skeleton at edge.
3. the recognition positioning method of coding maker according to claim 2, which is characterized in that in step s 51,
The single pixel skeleton of characteristic area is obtained using K3M sequential iteration method.
4. the recognition positioning method of coding maker according to claim 1, which is characterized in that in step s 30, pretreatment
Including one of filtering, denoising, enhancing or a variety of.
5. the recognition positioning method of coding maker according to claim 1, which is characterized in that in step s 40, pass through shape
Shape parametric method isolates border circular areas.
6. the recognition positioning method of coding maker according to claim 1, which is characterized in that in step S80, find out figure
As the corresponding dot of upper matched with design template all mark points, and judge that affine matrix solves correctness, if incorrect,
Return step S30；Then, coordinate matching decoding is carried out to other dots by look-up table, obtains the encoded radio of encoded point.
7. the recognition positioning method of coding maker according to claim 1, which is characterized in that in step S90, also wrap
It includes, the coordinate of record location point and the respective value of coding.
8. the recognition positioning method of coding maker according to claim 1-7, which is characterized in that in step S20
Before, further include step S10: in testee surface mount coding maker, and carrying out Image Acquisition.
9. the recognition positioning method of coding maker according to claim 1-7, which is characterized in that the mark point
Including the first mark point (A), the second mark point (B), third mark point (C), the anchor point is the crosspoint of a plurality of line segment
(O), the encoded point includes n point, and n is the integer more than or equal to 1, first mark point (A), the second mark point (B), the
Three mark points (C), crosspoint (O) are located on square four apex angle, the first mark point (A), the position crosspoint (O)
It is crosspoint (O) on a diagonal line of square, defining coding maker to design the origin of coordinate system, and coding maker designs
X-axis, the Y-axis of coordinate system are parallel to the adjacent both sides of rectangle, and encoded point is interspersed in coding maker design coordinate system.
10. the recognition positioning method of coding maker according to claim 1-7, which is characterized in that described default
Feature includes cross.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810875464.5A CN109215016B (en) | 2018-08-03 | 2018-08-03 | Identification and positioning method for coding mark |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810875464.5A CN109215016B (en) | 2018-08-03 | 2018-08-03 | Identification and positioning method for coding mark |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109215016A true CN109215016A (en) | 2019-01-15 |
| CN109215016B CN109215016B (en) | 2021-02-05 |
Family
ID=64987951
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810875464.5A Active CN109215016B (en) | 2018-08-03 | 2018-08-03 | Identification and positioning method for coding mark |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109215016B (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163916A (en) * | 2019-05-23 | 2019-08-23 | 西华大学 | The matching process of circulating type circular markers based on rolling target |
| CN111640083A (en) * | 2020-06-09 | 2020-09-08 | 中国科学院长春光学精密机械与物理研究所 | Color restoration method and system for object surface point cloud |
| CN113129394A (en) * | 2020-12-23 | 2021-07-16 | 合肥工业大学 | Parallelogram coding mark based on region division coding and coding method thereof |
| CN113188662A (en) * | 2021-03-16 | 2021-07-30 | 云南电网有限责任公司玉溪供电局 | Infrared thermal imaging fault automatic identification system and method |
| CN113313628A (en) * | 2021-06-29 | 2021-08-27 | 河南工业大学 | Affine transformation and mean pixel method-based annular coding point robustness identification method |
| CN113706621A (en) * | 2021-10-29 | 2021-11-26 | 上海景吾智能科技有限公司 | Mark point positioning and posture obtaining method and system based on marked image |
| CN114792104A (en) * | 2021-01-26 | 2022-07-26 | 中国科学院沈阳自动化研究所 | Method for identifying and decoding ring-shaped coding points |
| CN116687569A (en) * | 2023-07-28 | 2023-09-05 | 深圳卡尔文科技有限公司 | Coded identification operation navigation method, system and storage medium |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060291749A1 (en) * | 2005-06-07 | 2006-12-28 | Ralf Scheidhauer | Systems and methods for rendering symbols using non-linear scaling |
| CN103310215A (en) * | 2013-07-03 | 2013-09-18 | 天津工业大学 | Detecting and identifying method for annular coding mark point |
| CN103530889A (en) * | 2013-11-04 | 2014-01-22 | 天津工业大学 | Method for decoding coding mark points of measuring stick based on skeleton extraction |
| CN104154877A (en) * | 2014-09-03 | 2014-11-19 | 中国人民解放军国防科学技术大学 | Three-dimensional reconstruction and size measurement method of complex convex-surface object |
| US20180025479A1 (en) * | 2016-07-20 | 2018-01-25 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for aligning measurement data to reference data |
-
2018
- 2018-08-03 CN CN201810875464.5A patent/CN109215016B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20060291749A1 (en) * | 2005-06-07 | 2006-12-28 | Ralf Scheidhauer | Systems and methods for rendering symbols using non-linear scaling |
| CN103310215A (en) * | 2013-07-03 | 2013-09-18 | 天津工业大学 | Detecting and identifying method for annular coding mark point |
| CN103530889A (en) * | 2013-11-04 | 2014-01-22 | 天津工业大学 | Method for decoding coding mark points of measuring stick based on skeleton extraction |
| CN104154877A (en) * | 2014-09-03 | 2014-11-19 | 中国人民解放军国防科学技术大学 | Three-dimensional reconstruction and size measurement method of complex convex-surface object |
| US20180025479A1 (en) * | 2016-07-20 | 2018-01-25 | Toyota Motor Engineering & Manufacturing North America, Inc. | Systems and methods for aligning measurement data to reference data |
Non-Patent Citations (3)
| Title |
|---|
| 倪松章，成垒等: "视觉测量中环形编码标志点的精确识别算法研究", 《新技术新工艺实验研究》 * |
| 殷永凯，刘晓丽等: "圆形标志点的亚像素定位及其应用", 《红外与激光工程》 * |
| 许策，马杰，赵全明，唐红梅: "改进的圆形标志亚像素级中心检测方法", 《计算机工程》 * |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110163916A (en) * | 2019-05-23 | 2019-08-23 | 西华大学 | The matching process of circulating type circular markers based on rolling target |
| CN110163916B (en) * | 2019-05-23 | 2022-06-28 | 西华大学 | Rolling target-based surrounding type circular mark point matching method |
| CN111640083A (en) * | 2020-06-09 | 2020-09-08 | 中国科学院长春光学精密机械与物理研究所 | Color restoration method and system for object surface point cloud |
| CN113129394A (en) * | 2020-12-23 | 2021-07-16 | 合肥工业大学 | Parallelogram coding mark based on region division coding and coding method thereof |
| CN113129394B (en) * | 2020-12-23 | 2022-09-06 | 合肥工业大学 | Parallelogram coding mark based on region division coding and coding method thereof |
| CN114792104A (en) * | 2021-01-26 | 2022-07-26 | 中国科学院沈阳自动化研究所 | Method for identifying and decoding ring-shaped coding points |
| CN113188662A (en) * | 2021-03-16 | 2021-07-30 | 云南电网有限责任公司玉溪供电局 | Infrared thermal imaging fault automatic identification system and method |
| CN113313628A (en) * | 2021-06-29 | 2021-08-27 | 河南工业大学 | Affine transformation and mean pixel method-based annular coding point robustness identification method |
| CN113706621A (en) * | 2021-10-29 | 2021-11-26 | 上海景吾智能科技有限公司 | Mark point positioning and posture obtaining method and system based on marked image |
| CN116687569A (en) * | 2023-07-28 | 2023-09-05 | 深圳卡尔文科技有限公司 | Coded identification operation navigation method, system and storage medium |
| CN116687569B (en) * | 2023-07-28 | 2023-10-03 | 深圳卡尔文科技有限公司 | Coded identification operation navigation method, system and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109215016B (en) | 2021-02-05 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109215016A (en) | A kind of recognition positioning method of coding maker | |
| CN108827316A (en) | Mobile robot visual orientation method based on improved Apriltag label | |
| CN104751187B (en) | Meter reading automatic distinguishing method for image | |
| US9483830B2 (en) | Depth map generation method, related system and computer program product | |
| CN100377167C (en) | Two-dimensional code area precisive positioning method for two-dimensional recognition | |
| CN100390807C (en) | Trilateral poly-dimensional bar code easy for omnibearing recognition and reading method thereof | |
| CN107292922B (en) | A method of it is registrated for optics with diameter radar image | |
| CN103530889A (en) | Method for decoding coding mark points of measuring stick based on skeleton extraction | |
| CN109902527B (en) | Automatic QR code recognizing and reading method and device | |
| CN109978871B (en) | Fiber bundle screening method integrating probability type and determination type fiber bundle tracking | |
| CN109190742B (en) | Decoding method of coding feature points based on gray feature | |
| CN110096920A (en) | A kind of high-precision high-speed positioning label and localization method towards visual servo | |
| KR101742115B1 (en) | An inlier selection and redundant removal method for building recognition of multi-view images | |
| CN108960267A (en) | System and method for model adjustment | |
| CN110009670A (en) | The heterologous method for registering images described based on FAST feature extraction and PIIFD feature | |
| CN108022245B (en) | Facial line primitive association model-based photovoltaic panel template automatic generation method | |
| CN113506276B (en) | Marker and method for measuring structural displacement | |
| CN113793309A (en) | Sub-pixel level ellipse detection method based on morphological characteristics | |
| CN109214486B (en) | Three-dimensional code, three-dimensional code generation method and device, and three-dimensional code identification method and device | |
| CN110031471B (en) | Method, system and device for analyzing surface defect growth of large-caliber optical element | |
| CN110363101A (en) | A kind of flowers recognition methods based on CNN Fusion Features frame | |
| CN110084117A (en) | Document table line detecting method, system based on binary map segmented projection | |
| CN109115181A (en) | A kind of coding maker | |
| CN115496657A (en) | Pathological section digitalization method and system for low-resolution imaging | |
| CN112215319B (en) | Two-dimensional code of color mark characteristic graph and identification method thereof |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |