CN109977714A - A kind of integrated vision positioning method of the more QR codes of warehoused cargo - Google Patents
A kind of integrated vision positioning method of the more QR codes of warehoused cargo Download PDFInfo
- Publication number
- CN109977714A CN109977714A CN201910006108.4A CN201910006108A CN109977714A CN 109977714 A CN109977714 A CN 109977714A CN 201910006108 A CN201910006108 A CN 201910006108A CN 109977714 A CN109977714 A CN 109977714A
- Authority
- CN
- China
- Prior art keywords
- positioning
- code
- profile
- codes
- candidate 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1408—Methods for optical code recognition the method being specifically adapted for the type of code
- G06K7/1417—2D bar codes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K7/00—Methods or arrangements for sensing record carriers, e.g. for reading patterns
- G06K7/10—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation
- G06K7/14—Methods or arrangements for sensing record carriers, e.g. for reading patterns by electromagnetic radiation, e.g. optical sensing; by corpuscular radiation using light without selection of wavelength, e.g. sensing reflected white light
- G06K7/1404—Methods for optical code recognition
- G06K7/1439—Methods for optical code recognition including a method step for retrieval of the optical code
- G06K7/1443—Methods for optical code recognition including a method step for retrieval of the optical code locating of the code in an image
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Electromagnetism (AREA)
- General Health & Medical Sciences (AREA)
- Toxicology (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of integrated vision positioning methods of the more QR codes of warehoused cargo, are related to field of image detection, more particularly, to a kind of method of the integrated vision positioning of more QR codes.The present invention solve to cause due to factors such as QR code local environment complexity and multi-angled shootings existing detection method can not precise positioning QR code the technical issues of.The present invention is first filtered image, turn grayscale image, the pretreatment operations such as binaryzation, then according to the position of the closed contour information coarse localization out position detection figure in bianry image, detection figure is set according to the similar feature contraposition of the area that 3 position sensing figures are presented in the picture in the same QR code and carries out fine positioning, and then finally orients QR code.This method can accurately orient multiple two dimensional codes in scene from multiple shooting angle in complex environment, have stronger robustness, have performance well in actual use.
Description
Technical field
The present invention relates to field of image detection, more particularly, to a kind of integrated vision positioning of the more QR codes of warehoused cargo
Method.
Background technique
Two dimensional code is also known as two-dimensional bar code, is popular a kind of coding mode in mobile device in recent years.It uses certain
The specific geometric figure of kind is according to certain rules in the chequered with black and white graphic recording data symbol of plane (on two-dimensional directional) distribution
Information.Two dimensional code has many types, and one of the most common is exactly QR code, its full name is that Quick Response Code(is fast
Fast reaction code).QR code has many advantages, such as that memory capacity is big, discrimination is high compared with other two dimensional codes.
At present QR code in warehousing and logistics industry using more and more extensive, by integrated vision collecting two dimensional code to goods
More than one piece cargo on frame is made an inventory and data input, can greatly improve the working efficiency of storage sector.This just needs to realize
The integrated vision positioning of multiple QR codes and identification under complex background.In current QR code detection method, visited mostly by position
Survey graphic feature positioned, there are three the position sensing figures of a QR code, be located at the upper left corner, the lower left corner of QR code with
And the lower right corner.For convenience of narration, it will simply be referred to as hereafter as positioning pattern.Existing method usually utilizes QR code positioning pattern to be had
The proportionate relationship of chequered with black and white 1:1:3:1:1 detected, but since actual photographed inevitably adulterates interference, so that the ratio is special
Sign is not easy to be detected completely, and then influences the detection effect of QR code.
Open (bulletin) number " a kind of more QR codes while extracting detection for the Chinese invention patent application of 106991354 A of CN
Algorithm " from another angle the extraction algorithm to more QR codes is proposed, aforementioned proportion feature has been abandoned, has been briefly described below: first
Multiple positioning patterns are detected by the special profile inclusion relation of positioning pattern, calculate separately its center position, further according to
Whether the distance between positioning pattern central point meets the feature of right angled triangle to judge which positioning pattern (region) belongs to
In the same two dimensional code, and finally, positioning extracts multiple QR codes.However, in the practical application scenes such as warehouse shelf, vision
The two dimensional code ambient background of acquisition is usually more complex, and the region of some non-locating figures may also have above-mentioned special profile packet
Containing relationship, if being mistakenly considered positioning pattern, subsequent treatment effect undoubtedly will affect.Further, in practical applications, it claps
Angle is taken the photograph also it is difficult to ensure that vertical with subject, this, which will lead to the shape that two dimensional code is showed in figure, has centainly abnormal
Become.Therefore, the judgment method based on right angled triangle is also just no longer applicable in conventional method and above-mentioned patent application.
Summary of the invention
A kind of integrated visions of the more QR codes of warehoused cargo are provided it is an object of the invention to the deficiency for prior art to determine
Position method, this method can not only solve multiple QR codes under storage rack complex environment on multiple storage container surfaces
Orientation problem applies also for the case where shooting angle is with subject out of plumb.
In order to achieve the above objectives, the technical scheme is that carrying out the image of following steps by image processing algorithm
Treatment process:
S1: image preprocessing: carrying out gray processing and filtering operation to the original image comprising multiple two dimensional codes, obtain gray level image,
Threshold segmentation is carried out to gray level image again, obtains whole bianry image.
Following steps S2 and S3 parallel processing, in no particular order.
S2.1: edge detection: carrying out edge detection to whole bianry image, and all closed contours obtained in image are (total
Referred to as I).
S2.2: the foundation of level tree structure: according to the inclusion relation that closed contour I is mutual, the tree-like knot of level is established
Structure, each node indicates a profile in structure.
For multiple closed contours in image, a case where contoured interior surrounds other profiles there will necessarily be.In
It is that the present invention will not have besieged closed contour as the ceiling of hierarchical structure in image, will be directly surrounded by them
Closed contour as time high-rise, and so on, until using the internal closed contour without any profile as the bottom, completion
The foundation of level tree structure.
S2.3: the screening of positioning pattern coarse positioning: in the level tree structure of closed contour I, according to positioning pattern spy
3 layers of different profile inclusion relation, preliminary screening go out to meet the profile of this relationship；Calculate the respective central point of these profiles, will in
Coarse positioning of the heart point profile on its interior as positioning pattern.
Since the profile of positioning pattern belongs to convex polygon, central point must be on its interior.If calculated in some
Heart point position is not in its corresponding positioning pattern coarse positioning profile, then it represents that this coarse positioning profile is not convex polygon, less
It may be the profile of positioning pattern, this coarse positioning can be rejected.
S3.1: Morphological scale-space: the operation of morphology opening operation is carried out to whole bianry image obtained in S1, is had
The bianry image of multiple connected domains.
S3.2: edge detection: split operation result figure carries out edge detection, and all closed contours obtained in image are (total
Referred to as II).
S3.3: the foundation of level tree structure: according to the inclusion relation that closed contour II is mutual, the tree-like knot of level is established
Structure.
The generation of S3.4:QR code candidate region: by the differently contoured inside of each branch's bottom of the tree structure of profile II
Different gray values is assigned, is distinguished with showing.Each region will be formed by as QR code candidate region, obtained image is referred to as
For QR code candidate region image.
It is the two dimensional code in shelf on cargo, the area presented in captured figure since the present invention is targeted
It is smaller, so can be completely filled inside a two dimensional code after opening operation operates.After edge detection, for these
Closed contour detected by two dimensional code, inside will not exist other profiles, so be in each branch of level tree structure
The bottom.
Based on the above analysis, this step can find out the profile of each branch's bottom of hierarchical structure of closed contour II, by it
It is internal be assigned to different gray values, form the mutually different region of multiple gray values, i.e. QR code candidate region.
S4: the matching of coarse positioning and QR code candidate region: for each positioning pattern coarse positioning, according to its central point in QR
Gray value on the image of code candidate region, is divided into different groups, to represent which candidate region it belongs to.
Since there are three positioning patterns for a QR code tool, so after extracting positioning pattern coarse positioning needing which is learnt
A little coarse positionings belong to the same QR code.If single QR code pattern is in a plane, from the same angle, this QR code
On three positioning pattern central points still in this QR code within.
Based on the above analysis, this step carries out gray value of each coarse positioning central point on the image of QR code candidate region
Analysis: if the gray value of certain some center position is identical, being partitioned into the same group for their corresponding coarse positionings, indicates
They belong to the same candidate region.This method can more be competent at shooting angle compared with using the method for right angled triangle feature
The actual conditions of degree and subject out of plumb.
S5: the screening of positioning pattern fine positioning: being greater than wherein coarse positioning number 1 QR code candidate region, calculates every
The area that a coarse positioning is surrounded, and the feature similar according to the positioning pattern area of the same QR code carry out positioning pattern essence
The screening of positioning.
Since QR code is generally in a plane, no matter so being shot from which angle, in obtained figure
As in, the area that three positioning patterns in the same QR code are showed all is not much different.Even if because ambient enviroment is multiple
The wrong positioning of positioning pattern coarse positioning, i.e., be also determined as positioning pattern for the pattern of non-locating figure caused by the factors such as miscellaneous,
Those will not show similar feature by the area that the positioning pattern of erroneous detection is surrounded.
Based on the above analysis, this step is each thick in more same group on the basis of carrying out group division to coarse positioning
The area surrounded is positioned, the coarse positioning that area difference ratio is no more than certain threshold value is remained, the essence as positioning pattern
Positioning.
The final positioning of S6:QR code: using the convex closure of the candidate region profile where fine positioning as the final positioning of QR code.
Compared with prior art, the present invention has following obvious prominent substantive distinguishing features and significant technological progress:
The present invention is still able to maintain preferable detection robustness under complex environment, and when shooting angle is not vertical with subject
Multiple QR codes in image can be accurately oriented, there is good practical value.
Detailed description of the invention
Specific embodiments of the present invention will be described in further detail with reference to the accompanying drawing.
Fig. 1 is the specific flow chart that the present invention carries out positioning QR code.
Fig. 2 is the original image comprising more QR codes to be processed in the embodiment of the present invention.
Fig. 3 is the gray level image of original image in the embodiment of the present invention.
Fig. 4 is the bianry image in the embodiment of the present invention.
Fig. 5 is the sample display diagram for the closed contour level tree structure that the present invention is utilized.
Fig. 6 is to carry out the effect picture after opening operation in the embodiment of the present invention to bianry image.
Fig. 7 is to carry out the closed contour figure after edge detection in the embodiment of the present invention to bianry image.
Fig. 8 is 3 layers of special profile inclusion relation schematic diagram of the positioning pattern of the invention utilized.
Fig. 9 is the upper left corner closed contour figure (Fig. 7) enlarged drawing in the embodiment of the present invention.
Figure 10 is the effect picture screened in the embodiment of the present invention to positioning pattern coarse positioning.
Figure 11 is the QR code candidate region image constituted in the embodiment of the present invention.
Figure 12 is the magnified partial view of positioning pattern coarse positioning effect picture (Figure 10) in the embodiment of the present invention.
Figure 13 is the fine positioning image of positioning pattern in the embodiment of the present invention.
Figure 14 is final more QR code precise positioning figures of the embodiment of the present invention.
Figure 15 is QR code positioning figure of the present invention from front shooting image.
Figure 16 is QR code positioning figure of the present invention from upper right side shooting image.
Figure 17 is QR code positioning figure of the present invention from upper left side shooting image.
Specific embodiment
With reference to the accompanying drawing and preferred embodiment, invention is further described in detail, but protection scope of the present invention
It is not limited to following specific embodiments.Obviously, described embodiment is only some embodiments of the present application, rather than whole
Embodiment.Based on the embodiment in the application, those of ordinary skill in the art are obtained without making creative work
The every other embodiment obtained, shall fall within the protection scope of the present application.
Unless otherwise defined, all technical terms used hereinafter are generally understood meaning phase with those skilled in the art
Together.Technical term used herein is intended merely to the purpose of description specific embodiment, and it is of the invention to be not intended to limitation
Protection scope.
Embodiment one: referring to Fig. 1 to Figure 17, the integrated vision positioning method of the more QR codes of this warehoused cargo, it is characterised in that
Operating procedure is as follows:
S1: pretreatment: whole bianry image is obtained；
S2.1: edge detection is carried out to whole binary map, is obtained closed contour (collectively referred to as I)；
S2.2: closed contour I level tree structure is established: according to its mutual inclusion relation；
S2.3: the screening of position sensing figure coarse positioning: according to its 3 layers of special profile inclusion relation and central point whether
Contoured interior is screened；
S3.1: Morphological scale-space: opening operation is carried out to whole binary map, forms multiple connected domains；
S3.2: split operation result figure carries out edge detection, obtains closed contour (collectively referred to as II)；
S3.3: closed contour II level tree structure is established: according to its mutual inclusion relation；
The generation of S3.4:QR code candidate region: by the differently contoured internal imparting of each branch's bottom of the tree structure of profile II
Different gray values is distinguished with showing；
S4: the matching of coarse positioning and QR code candidate region: according to the gray value of each coarse positioning central point；
S5: the screening of position sensing figure fine positioning: according to the similar feature of the position sensing graphics area on the same QR code；
S6:QR code finally positions: the convex closure of the QR code candidate region profile where fine positioning.
Embodiment two: the present embodiment is basically the same as the first embodiment, and special feature is as follows:
The big step S2 — that is, S2.1 to S2.3 and big step S3 — that is, S3.1 to S3.4, parallel processing, in no particular order.
The step S3.4 further include:
1) all profiles of the closed contour II level tree structure Zhong Ge branch bottom are found out；
2) these contoured interiors are assigned to different gray values, are distinguished with showing, assigned gray value is incremented by one by one since 0.
The step S4 further include:
1) for the position sensing figure coarse positioning filtered out, its respective center position is calculated；
2) for each coarse positioning center position, its gray value in QR code candidate region is taken；
If 3) gray value in certain center positions is identical, it is partitioned into same group, is completed candidate with corresponding QR code
The matching in region.
The step S5 further include:
1) each QR code candidate region is retained, is otherwise given up if the coarse positioning number being wherein matched to is greater than 1；
2) for the QR code candidate region of reservation, the area that wherein each coarse positioning profile is surrounded is calculated；
3) if these coarse positionings are sieved there are the coarse positioning profile that area difference is no more than 20% in some QR code candidate region
It is selected as fine positioning.
The step S6 further include:
1) QR code candidate region of all inside there are position sensing figure fine positioning is found out；
2) convex closure of these QR code candidate regions is calculated, and using these convex closures as the final positioning result of QR code.
Embodiment three:
The specific flow chart of the present embodiment is as shown in Figure 1.
The present embodiment will be 1916*894 with resolution ratio shown in Fig. 2, for the image of simulated warehouse shelf actual scene,
Detailed description uses technical solution provided by the present invention, detects the process of multiple QR codes wherein on storage container,
The following steps are included:
S1: by camera or the collected original image read-in programme of other image capture devices, the present embodiment is with original shown in Fig. 2
For image.Gaussian filtering is carried out to original image and turns gray scale graphic operation, obtains grayscale image as shown in Figure 3.Grayscale image is carried out
The threshold value of Threshold segmentation, segmentation is determined using the method for adaptive threshold fuzziness, obtains bianry image as shown in Figure 4.
Following steps S2 and S3 parallel processing, in no particular order.
S2.1: carrying out edge detecting operation to bianry image, obtain the line of demarcation of its black picture element and white pixel, i.e., whole
The closed contour I of width image, the result of acquisition are as shown in Figure 7.
S2.2: while detecting edge, the inclusion relation between closed contour is got, is built according to this inclusion relation
The level tree structure of vertical closed contour I.
To elaborate, by taking profile shown in fig. 5 as an example.In figure, the closed line of black indicates profile, in figure
4 black blockade lines, i.e. 4 profiles, indicate 1 to 4 number to be mutually distinguishable.For profile 1, internal directly includes wheel
Exterior feature 2 and profile 3.And the region that profile 2 and profile 3 respectively surround has no intersection, then end out line 2 and profile 3 are coordination.?
The inside of profile 3 also includes profile 4.Inclusion relation between above-mentioned 4 profiles can use the level tree structure on the right side of Fig. 5
Represented, wherein profile 2 and profile 4 are exactly the profile of the bottom in the branch where each.
The example of profile inclusion relation in above-mentioned Fig. 5 at large illustrates the original that closed contour level tree structure is constituted
Reason.
Positioning pattern in QR code has 3 layers of special profile inclusion relation.As shown in figure 8,3 closing white wires in figure
Item is the profile in positioning pattern between black picture element and white pixel.It includes to close that this 3 profiles, which meet following special 3 layers of profile,
System: including in one profile of outermost layer and only includes another profile, and wherein, also include and only includes the 3rd article of profile,
Any profile is not present in 3rd article of contoured interior.
S2.3: in the level tree structure of closed contour I, lookup meets the special 3 layers of profile inclusion relation of positioning pattern
Profile: since top layer's profile, judged one by one along each branch to bottom profile, judged in each profile A
Portion whether other profiles B that one and only one is surrounded by the profile, will meet the condition profile A and its inside profile B
It records respectively, if also meeting the condition, and for the profile inside profile B when the subsequent above-mentioned judgement of progress to profile B
C, inside without any profile, then profile A is extracted, whether inside it judges its central point, if inside it,
Then as the coarse positioning of positioning pattern.The effect for carrying out aforesaid operations to Fig. 7 is as shown in Figure 10.
More fully to illustrate, by taking the enlarged drawing Fig. 9 in the upper left corner Fig. 7 as an example.There is the closure wheel of reference number in figure
Exterior feature 1 to 3, since profile 1 is in the upper layer of other 2 profiles in the profile level tree structure of foundation, so first to wheel
Exterior feature 1 judged, judges wherein to have really and only a profile, i.e. profile 2, then records profile 1 and profile 2 respectively
Get off, and continues to judge other profiles.It when judging to profile 2, finds also to have inside it and an only profile, i.e.,
Profile 3, and for profile 3, it finds that any profile has been not present inside it.Then profile 1 is extracted, calculates its center
Point position, discovery is really on its interior, then the coarse positioning by profile 1 as positioning pattern, and effect also has in Figure 10
It embodies.
But the not all profile central point for meeting 3 layers of profile inclusion relation is all inside it.By taking Figure 10 as an example, cargo 2
The overstriking profile got on the car in pattern meets 3 layers of profile inclusion relation, but its obvious central point (being marked with stain) does not exist
Contoured interior, so being rejected.And remaining overstriking profile all meets 3 layers of profile inclusion relation simultaneously and its central point all exists
Contoured interior, so being retained as positioning pattern coarse positioning.
S3.1: the operation of morphology opening operation is carried out to the bianry image that S1 is obtained, is used particularly as the template size of opening operation
It is determined according to size of the two dimensional code in figure, template size used by this example is 9*9.It is obtained shown in Fig. 6 by opening operation
Result figure.
S3.2: split operation result Fig. 6, it carries out edge detection and establishes the level tree structure of closed contour II in figure.
It assigns the contoured interior of each branch's bottom of tree structure to mutually different gray value, constitutes multiple QR codes candidate region, shape
At new images be referred to as QR code candidate region image.As shown in figure 11, wherein grey lines are each branch of tree structure most bottom
The profile of layer.In these contoured interiors, it is assigned to different gray values all to show and distinguish, gray value has used number to mark in figure
Out.
80 to 255 gray values (or gray-scale level) can satisfy requirement under normal circumstances.If bottom profile
Number then can be used 16 0 to 65635 gray-scale levels and be marked more than 256.
S4: the position for positioning pattern coarse positioning, where representing them with its central point.Judge that its position is in
In which candidate region of QR code candidate region image (Figure 11), it is thick fixed in the same candidate region that position is in
Position, divides them into one group, represents them and completes to match with the QR code candidate region.
More fully to illustrate, by taking Figure 12 as an example.Figure 12 is 1 enlarged drawing of cargo of the shelf second layer rightmost side in Figure 10.
Being can see in the two dimensional code on cargo 1 in figure has 4 overstriking profiles, i.e. 4 positioning pattern coarse positionings.Their own center
Point has also been marked in figure with black dot.For the position where each coarse positioning central point, the position is obtained in Figure 11
In gray value.It is at due to the position of resulting 4 coarse positioning central points in the QR code candidate region that gray value is 40, institute
With for above-mentioned 4 coarse positioning central points, the gray value got is all 40, then by the coarse positioning of this 4 positioning patterns point
It is one group, completes the matching with the QR code candidate region.
S5: on the basis of completing matched to coarse positioning and QR code candidate region, for internal coarse positioning number greater than 1
Each candidate region all calculates the area that wherein coarse positioning is surrounded, and takes area similar 2 or 3 coarse positionings conduct essences
Positioning.
More fully to illustrate, still by taking Figure 12 as an example.In the two dimensional code on cargo 1, there are 4 be marked out
A positioning pattern coarse positioning.In S4, this 4 coarse positionings have been divided into same group, then calculate separately their inside and are surrounded
Pixel number, i.e. surround the area.Surround the area difference is no more than 20% coarse positioning if it exists, then can be as fixed
Bit pattern fine positioning.Can intuitively it find out very much in Figure 10, coarse positioning profile packet representated by coarse positioning central point 1,2,3
It is much smaller to enclose coarse positioning profile surround the area representated by area ratio coarse positioning central point 4, and 1,2,3 institute of coarse positioning central point
All meet the condition that difference is no more than 20% between the coarse positioning surround the area of representative, then as positioning pattern fine positioning.
In addition, there is also the positioning pattern coarse positionings of some mistakes in the automobile pattern on cargo 3 in Figure 10, but by
In in QR code candidate region of the central point not in Figure 11 of these profiles, the operation of S5 would not be also participated in, into
And it is removed.Even if there are the coarse positionings of some mistakes to be in the same candidate region, the area that they are surrounded also is difficult
Meet the condition that difference is no more than 20%, and then can also be removed away.
The result for carrying out the choosing of positioning pattern fine positioning brush to Figure 10 is as shown in figure 13.Wherein overstriking grey profile is to position
Figure fine positioning.
S6: the QR code candidate region for inside containing positioning pattern fine positioning calculates the convex closure of its profile, by this convex closure
As final QR code precise positioning result.
More fully to illustrate, by taking Figure 13 as an example.Candidate region in figure where positioning pattern fine positioning is all that QR code is true
Real region calculates the convex closure of these candidate region profiles, the QR code precise positioning result final as the embodiment of the present invention.
And for such as candidate regions such as automobile pattern, due to wherein and there is no fine positionings as a result, can be removed in turn.Final positioning
Effect is as shown in figure 14, and wherein white contours are the final positioning result of QR code.
For show the present invention from multiple shooting angle can precise positioning QR code ability, Figure 15 to 17 respectively show from
Shelf front, upper right side, upper left side are to the positioning result of QR code, it can be seen that the present invention is really to from multiple shooting angles
The case where degree is shot has extraordinary adaptability.
Claims (6)
1. a kind of integrated vision positioning method of the more QR codes of warehoused cargo, it is characterised in that operating procedure is as follows:
S1: pretreatment: whole bianry image is obtained；
S2.1: edge detection is carried out to whole binary map, is obtained closed contour (collectively referred to as I)；
S2.2: closed contour I level tree structure is established: according to its mutual inclusion relation；
S2.3: the screening of position sensing figure coarse positioning: according to its 3 layers of special profile inclusion relation and central point whether
Contoured interior is screened；
S3.1: Morphological scale-space: opening operation is carried out to whole binary map, forms multiple connected domains；
S3.2: split operation result figure carries out edge detection, obtains closed contour (collectively referred to as II)；
S3.3: closed contour II level tree structure is established: according to its mutual inclusion relation；
The generation of S3.4:QR code candidate region: by the differently contoured internal imparting of each branch's bottom of the tree structure of profile II
Different gray values is distinguished with showing；
S4: the matching of coarse positioning and QR code candidate region: according to the gray value of each coarse positioning central point；
S5: the screening of position sensing figure fine positioning: according to the similar feature of the position sensing graphics area on the same QR code；
S6:QR code finally positions: the convex closure of the QR code candidate region profile where fine positioning.
2. the integrated vision positioning method of the more QR codes of a kind of warehoused cargo according to claim 1, wherein big step S2-
That is S2.1 to S2.3 and big step S3 — that is, S3.1 to S3.4, parallel processing, in no particular order.
3. the integrated vision positioning method of the more QR codes of a kind of warehoused cargo according to claim 1, the step S3.4 is also
Include:
1) all profiles of the closed contour II level tree structure Zhong Ge branch bottom are found out；
2) these contoured interiors are assigned to different gray values, are distinguished with showing, assigned gray value is incremented by one by one since 0.
4. the integrated vision positioning method of the more QR codes of a kind of warehoused cargo according to claim 1, the step S4 are also wrapped
It includes:
1) for the position sensing figure coarse positioning filtered out, its respective center position is calculated；
2) for each coarse positioning center position, its gray value in QR code candidate region is taken；
If 3) gray value in certain center positions is identical, it is partitioned into same group, is completed candidate with corresponding QR code
The matching in region.
5. the integrated vision positioning method of the more QR codes of a kind of warehoused cargo according to claim 1, the step S5 are also wrapped
It includes:
1) each QR code candidate region is retained, is otherwise given up if the coarse positioning number being wherein matched to is greater than 1；
2) for the QR code candidate region of reservation, the area that wherein each coarse positioning profile is surrounded is calculated；
3) if these coarse positionings are sieved there are the coarse positioning profile that area difference is no more than 20% in some QR code candidate region
It is selected as fine positioning.
6. the integrated vision positioning method of the more QR codes of a kind of warehoused cargo according to claim 1, the step S6 are also wrapped
It includes:
1) QR code candidate region of all inside there are position sensing figure fine positioning is found out；
2) convex closure of these QR code candidate regions is calculated, and using these convex closures as the final positioning result of QR code.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910006108.4A CN109977714B (en) | 2019-01-04 | 2019-01-04 | Multi-QR-code integrated visual positioning method for warehoused goods |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910006108.4A CN109977714B (en) | 2019-01-04 | 2019-01-04 | Multi-QR-code integrated visual positioning method for warehoused goods |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109977714A true CN109977714A (en) | 2019-07-05 |
| CN109977714B CN109977714B (en) | 2022-06-10 |
Family
ID=67076524
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910006108.4A Active CN109977714B (en) | 2019-01-04 | 2019-01-04 | Multi-QR-code integrated visual positioning method for warehoused goods |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109977714B (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111369188A (en) * | 2020-02-27 | 2020-07-03 | 广州中智融通金融科技有限公司 | Machine vision-based bundled note warehousing method, system and medium |
| CN111428530A (en) * | 2020-02-29 | 2020-07-17 | 厦门华联电子股份有限公司 | Two-dimensional code image detection and identification equipment, device and method |
| CN111783493A (en) * | 2020-06-18 | 2020-10-16 | 福州富昌维控电子科技有限公司 | Identification method and identification terminal for batch two-dimensional codes |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260693A (en) * | 2015-12-01 | 2016-01-20 | 浙江工业大学 | Laser two-dimensional code positioning method |
| CN106485183A (en) * | 2016-07-14 | 2017-03-08 | 深圳市华汉伟业科技有限公司 | A kind of Quick Response Code localization method and system |
| CN106529370A (en) * | 2016-10-18 | 2017-03-22 | 福州觉感视觉软件科技有限公司 | Multi-QR code identification method |
| CN106991354A (en) * | 2017-01-23 | 2017-07-28 | 中山大学 | A kind of many QR codes extract detection algorithm simultaneously |
| WO2017157034A1 (en) * | 2016-03-15 | 2017-09-21 | 中兴通讯股份有限公司 | Method, device, and storage medium for identifying two-dimensional code |
| CN108985129A (en) * | 2018-08-02 | 2018-12-11 | 谷东科技（广州）有限公司 | The positioning of two dimensional code and method for detecting |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
-
2019
- 2019-01-04 CN CN201910006108.4A patent/CN109977714B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260693A (en) * | 2015-12-01 | 2016-01-20 | 浙江工业大学 | Laser two-dimensional code positioning method |
| WO2017157034A1 (en) * | 2016-03-15 | 2017-09-21 | 中兴通讯股份有限公司 | Method, device, and storage medium for identifying two-dimensional code |
| CN106485183A (en) * | 2016-07-14 | 2017-03-08 | 深圳市华汉伟业科技有限公司 | A kind of Quick Response Code localization method and system |
| CN106529370A (en) * | 2016-10-18 | 2017-03-22 | 福州觉感视觉软件科技有限公司 | Multi-QR code identification method |
| CN106991354A (en) * | 2017-01-23 | 2017-07-28 | 中山大学 | A kind of many QR codes extract detection algorithm simultaneously |
| CN108985129A (en) * | 2018-08-02 | 2018-12-11 | 谷东科技（广州）有限公司 | The positioning of two dimensional code and method for detecting |
| CN109724990A (en) * | 2019-01-08 | 2019-05-07 | 上海大学 | The quick positioning and detection method in coding region in a kind of packing box label |
Non-Patent Citations (1)
| Title |
|---|
| 张勇等: "基于凸包及最小面积外接矩形的", 《电子测量技术》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111369188A (en) * | 2020-02-27 | 2020-07-03 | 广州中智融通金融科技有限公司 | Machine vision-based bundled note warehousing method, system and medium |
| CN111369188B (en) * | 2020-02-27 | 2024-04-26 | 广州运通科金技术有限公司 | Machine vision-based banknote bundling and warehousing method, system and medium |
| CN111428530A (en) * | 2020-02-29 | 2020-07-17 | 厦门华联电子股份有限公司 | Two-dimensional code image detection and identification equipment, device and method |
| CN111428530B (en) * | 2020-02-29 | 2022-07-15 | 厦门华联电子股份有限公司 | Two-dimensional code image detection and identification equipment, device and method |
| CN111783493A (en) * | 2020-06-18 | 2020-10-16 | 福州富昌维控电子科技有限公司 | Identification method and identification terminal for batch two-dimensional codes |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109977714B (en) | 2022-06-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110427905B (en) | Pedestrian tracking method, device and terminal | |
| CN110210409B (en) | Method and system for detecting form frame lines in form document | |
| CN106485183B (en) | A kind of Quick Response Code localization method and system | |
| CN107507167B (en) | Cargo tray detection method and system based on point cloud plane contour matching | |
| CN115082419B (en) | Blow-molded luggage production defect detection method | |
| CN102609686B (en) | Pedestrian detection method | |
| CN107038424A (en) | A kind of gesture identification method | |
| CN108009460A (en) | The recognition methods of various characters and the superposition Quick Response Code of numeral mixing under complex background | |
| CN102254144A (en) | Robust method for extracting two-dimensional code area in image | |
| CN109977714A (en) | A kind of integrated vision positioning method of the more QR codes of warehoused cargo | |
| CN110321769A (en) | A kind of more size commodity on shelf detection methods | |
| CN110097596A (en) | A kind of object detection system based on opencv | |
| CN106997459A (en) | A kind of demographic method split based on neutral net and image congruencing and system | |
| CN107092871A (en) | Remote sensing image building detection method based on multiple dimensioned multiple features fusion | |
| Deng et al. | Detection and recognition of traffic planar objects using colorized laser scan and perspective distortion rectification | |
| CN102930251A (en) | Two-dimensional collection data recording and discriminating device and method | |
| CN115830359A (en) | Workpiece identification and counting method based on target detection and template matching in complex scene | |
| US20230038286A1 (en) | Imaging system and method using a multi-layer model approach to provide robust object detection | |
| CN113657264A (en) | Forest fire smoke root node detection method based on fusion of dark channel and KNN algorithm | |
| CN110378337B (en) | Visual input method and system for drawing identification information of metal cutting tool | |
| CN111402185B (en) | Image detection method and device | |
| CN112330659A (en) | Geometric tolerance symbol segmentation method combining LSD (least squares) linear detection and connected domain marking method | |
| Lafuente-Arroyo et al. | Traffic sign classification invariant to rotations using support vector machines | |
| Kopf et al. | Robust character recognition in low-resolution images and videos | |
| CN102073868A (en) | Digital image closed contour chain-based image area identification method |
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 |