CN108507484B - Bundled round steel multi-vision visual identifying system and method for counting - Google Patents
Bundled round steel multi-vision visual identifying system and method for counting Download PDFInfo
- Publication number
- CN108507484B CN108507484B CN201810303608.XA CN201810303608A CN108507484B CN 108507484 B CN108507484 B CN 108507484B CN 201810303608 A CN201810303608 A CN 201810303608A CN 108507484 B CN108507484 B CN 108507484B
- Authority
- CN
- China
- Prior art keywords
- round steel
- main phase
- phase machine
- light source
- radical
- 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.)
- Active
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
- G01B11/08—Measuring arrangements characterised by the use of optical techniques for measuring diameters
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/70—Determining position or orientation of objects or cameras
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/80—Analysis of captured images to determine intrinsic or extrinsic camera parameters, i.e. camera calibration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/10—Image acquisition modality
- G06T2207/10004—Still image; Photographic image
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30242—Counting objects in image
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
Bundled round steel multi-vision visual identifying system and method for counting, the system comprises: light source (1), main phase machine (2), main phase machine and light source bracket (3), auxiliary camera (4), auxiliary camera support (5), computer (6) and scaling board (7)；The method can obtain the image information of multiple angles of bundled round steel end face, image occlusion issue caused by effective solution is not spent concordantly because of end face, the calculated result in system meeting comprehensive visual recognition result and company information library, and conclusion reliability is higher.
Description
Technical field
The present invention relates to a kind of object visual identifying system and the methods of counting, more particularly, to a kind of bundled more mesh of round steel
Visual identifying system and method for counting.
Background technique
Round steel is the major product of present steel industry, even more the important production and processing raw material of industrial circle.In difference
Industry requires also to be different for extraordinary round steel, steel production enterprise produce and be supplied to the round steel of down-stream enterprise often with
Bundled form occurs.For the ease of the control of extraordinary round steel, need to carry out the total weight and radical of bundled round steel statistics and
Mark.According to the investigation to certain domestic special steel manufacturing enterprise, generallyd use in the production and management of domestic bundled round steel at present pair
Bundled round steel weighs to obtain bundle weight, single round steel and calculates list by type information (including material, density, diameter and length)
Then the root weight of root round steel obtains the radical of bundled round steel divided by root weight with bundle weight.It can be seen that the prior art is bundled
The radical information of round steel is that " calculating " obtains, and is not in calculating mistake generally for the big stick-means of intimidation being relatively large in diameter, and for straight
The lesser spillikin the method for diameter is it is possible that radical mistake of statistics.
In patent [CN201610193685.5], a kind of bundled round steel end face center coordinate of view-based access control model is proposed
System and method are obtained, is identified with end face of the monocular vision to bundled round steel and carries out plane positioning.In patent
In [CN201710878572.3], a kind of round steel end face binocular visual positioning method based on major-minor eye is proposed, with binocular vision
Feel system identifies the end face of bundled round steel and carries out space orientation, but also can recorde the quantity of round steel, realizes meter
Number function.But the method mentioned in two above patent be mainly used for bundled big stick-means of intimidation labeling when to the end face of round steel carry out identification and
Positioning, is not suitable for spillikin.Currently, although there is arrangement station when round steel bundling, but the end face of bundled round steel still have it is certain
Do not spend concordantly, in this way for spillikin with can be generated when visual imaging block so as to cause visual identity failure, to generate
Miscount.Therefore, it invents and a kind of can effectively solve the problem that image blocks the vision system and meter for leading to bundled round steel miscount
Counting method is particularly important for solving production practical problem.
Summary of the invention
Based on the above reasons, the present invention is directed to propose a kind of bundled round steel multi-vision visual identifying system and method for counting.
Bundled round steel multi-vision visual identifying system of the invention includes: light source 1, main phase machine 2, main phase machine and light source bracket 3,
Auxiliary camera 4, auxiliary camera support 5, computer 6 and scaling board 7.Light source 1 is annular light source, annular of the main phase machine 2 in annular light source 1
In hole and light source 1 is coaxially arranged, and light source 1 and main phase machine 2 are installed and fixed by main phase machine and light source bracket 3.Auxiliary camera 4 passes through
Auxiliary camera support 5 installs and fixes.The axis radial equipartition of multiple auxiliary cameras 4 and camera support 5 around main phase machine 2.Main phase
Machine 2 and multiple auxiliary cameras 4 form multi-vision visual image collecting device.When installation, make light by adjusting main phase machine and light source bracket 3
The axis of source 1 and main phase machine 2 is made more perpendicular to the center of the end face of bundled round steel by the multiple auxiliary camera supports 5 of adjusting
The axis of a auxiliary camera 4 and the axis of main phase machine 2 intersect at a point, and the focus is made to be located at the theory of bundled round steel entirety end face
Center.Light source 1, main phase machine 2, multiple auxiliary cameras 4 are connected with computer 6 by cable.Computer 6 is equipped with visual identity
With processing routine, and corporate intranet is accessed, production information can be read in real time online.Scaling board 7 is for main phase machine 2 and multiple
The vision calibration of auxiliary camera 4, calibration process need to only carry out primary after this system is correctly installed.
Included the following steps: using the bundled round steel identification of bundled round steel multi-vision visual system of the invention and method of counting
(1) the good bundled round steel multi-vision visual identifying system of installation and debugging；
(2) scaling board is arranged；
(3) monocular calibration is carried out to main phase machine and auxiliary camera, withdraws scaling board after the completion；
(4) Image Acquisition, image enhancement processing and image binaryzation is carried out respectively to main phase machine and auxiliary camera to handle；
(5) identification of image circle, centralized positioning are carried out respectively to main phase machine and auxiliary camera；
(6) the online radical and diameter information for reading bundled round steel in company information library in real time；
(7) synthesis is carried out to the round steel end face center point that main phase machine and auxiliary camera obtain according to the diameter information of round steel to sentence
It is disconnected, it determines the radical and reliability that bundled round steel visual identity goes out and carries out central point distance verification；
(8) calculated result in the visual identity result and company information library of more bundled round steel radical, comprehensive descision.
Advantageous effects of the invention include:
(1) image information of multiple angles of bundled round steel end face can be obtained, is effectively solved because concordantly degree is not led for end face
The image occlusion issue of cause；
(2) calculated result in system meeting comprehensive visual recognition result and company information library, conclusion reliability are higher.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the bundled round steel multi-vision visual identifying system of the present invention；
Fig. 2 is bundled round steel multi-vision visual identifying system main phase machine 2 of the invention and the radial arrangement signal of multiple auxiliary cameras 4
Figure；
Fig. 3 is that bundled round steel multi-vision visual identifying system main phase machine 2 of the invention and multiple 4 axis directions of auxiliary camera arrangement are shown
It is intended to.
Specific embodiment:
The structure and concrete operations of present system are introduced in conjunction with attached drawing 1-3.
Bundled round steel multi-vision visual identifying system of the invention includes: light source 1, main phase machine 2, main phase machine and light source bracket 3,
Auxiliary camera 4, auxiliary camera support 5, computer 6 and scaling board 7.Light source 1 is annular LED light source, and main phase machine 2 is in annular light source 1
In looping pit and light source 1 is coaxially arranged, and light source 1 and main phase machine 2 are installed and fixed by main phase machine and light source bracket 3.Auxiliary camera 4
It is installed and fixed by auxiliary camera support 5.The axis radial equipartition of 3 auxiliary cameras 4 and camera support 5 around main phase machine 2, folder
Angle is 120 °.The auxiliary camera 4 of main phase machine 2 and 3 forms multi-vision visual image collecting device.When installation, by adjust main phase machine and
Light source bracket 3 makes the axis of light source 1 and main phase machine 2 perpendicular to the center of the end face of bundled round steel, by adjusting 3 auxiliary phases
Machine support 5 makes the axis of 3 auxiliary cameras 4 and the axis of main phase machine 2 intersect at a point, and it is whole so that the focus is located at bundled round steel
The theoretical center position of end face.2,3 light source 1, main phase machine auxiliary cameras 4 are connected with computer 6 by cable.Computer 6 is installed
There are visual identity and processing routine, and access corporate intranet, production information can be read in real time online.Scaling board 7 is for leading
The vision calibration of the auxiliary camera 4 of camera 2 and 3, calibration process need to only carry out primary after this system is correctly installed.
Included the following steps: using the bundled round steel identification of bundled round steel multi-vision visual system of the invention and method of counting
(1) the good bundled round steel multi-vision visual identifying system of installation and debugging；
(2) scaling board is arranged；
(3) monocular calibration is carried out to main phase machine and auxiliary camera, withdraws scaling board after the completion；
(4) Image Acquisition, image enhancement processing and image binaryzation is carried out respectively to main phase machine and auxiliary camera to handle；
(5) identification of image circle, centralized positioning are carried out respectively to main phase machine and auxiliary camera；
(6) the online radical and diameter information for reading bundled round steel in company information library in real time；
(7) synthesis is carried out to the round steel end face center point that main phase machine and auxiliary camera obtain according to the diameter information of round steel to sentence
It is disconnected, it determines the radical and reliability that bundled round steel visual identity goes out and carries out central point distance verification；
(8) calculated result in the visual identity result and company information library of more bundled round steel radical, comprehensive descision.
If the bundled round steel end face diameter of certain special steel manufacturing enterprise is about 300mm, setting position measurement accuracy 0.5mm,
Camera and round steel end face distance about 1000mm.Pixel is estimated first, and measured object is the circle of 300mm, and camera target surface is usually
The rectangle of 4:3, in order to which by object, all intake target surfaces should be reference with target surface most bond length, pixel should be greater than 300/0.5=
600,1/3 inch of size of CCD camera MER-125-30UM target surface of Daheng (4.8 × 3.6mm) can choose according to the pixel of estimation,
Resolution ratio is 1292 × 964, and pixel dimension isu=3.75 μm of camera verifies precisionT, can be according to formulaT=u/βIt calculates, whereinβ=3.6/300 is camera lens magnifying power, is computedT=0.31mm meets required precision；Focal lengthfFormula can be passed throughf=L/ (1+1/β)
It calculates, whereinL=1000mm, is computedf=11.8mm, optional focal length is the camera lens of 12mm, such as Daheng's TG4Z2816FCS camera lens.
Light source selects the LED annular light source of the model VR144-B of thousand Electro-optical Technology, INC. (US) 62 Martin Road, Concord, Massachusetts 017, Shanghai side production, electricity matched with light source
Source model VPD-24S.In view of filtering veiling glare, light source is configured to blue light, camera Selecting Complete Set blue filter, model
BP470-30.5.Main phase machine and 3 auxiliary cameras are all made of identical industrial camera with blue filter, and illumination uses annular LED
Blue-light source, using industrial personal computer as system control unit.
1, the good bundled round steel multi-vision visual identifying system of installation and debugging
Bundled round steel is placed in visual identity and counts on station, is horizontally arranged, and whole round steel end face guarantees not spend concordantly
20mm.System makes main phase in use, bundled round steel multi-vision visual identifying system to be placed in the front of bundled round steel entirety end face
It is whole that the crossing point of axes of the coincidence of the ideal axis of the axis of machine 2 and bundled round steel, main phase machine 2 and 3 auxiliary cameras 4 is located at bundled round steel
The center in body end face.The array circular diameter of 3 auxiliary cameras is set as steel bundle diameter, i.e. 300mm.
2, scaling board is arranged
Arrangement scaling board 7 is located on the ideal plane of bundled round steel entirety end face, and makes the auxiliary camera 4 of main phase machine 2 and 3
Axis focuses on the center of scaling board 7.
3, monocular calibration is carried out to main phase machine and auxiliary camera, withdraws scaling board after the completion
Delaunay subdivision triangular interpolation value standardization is a kind of method of two-dimensional calibrations, for determining world coordinates
Transformational relation between system and image coordinate system.Select dot matrixes type target as scaling board, using in each border circular areas
For the heart as characteristic point, stated accuracy is higher than other types scaling board.
The step of Delaunay subdivision triangular interpolation value standardization:
1) plane target drone is placed perpendicular to ground, both sides are parallel to horizontal plane and guarantee plane target drone and circle target up and down
Steel end face is in the same plane, and the optical center of camera places and guarantee the position of camera position and plane target drone perpendicular to target face
It sets and immobilizes；
2) central point for defining plane target drone upper left corner dot is the origin of world coordinates, and world coordinatesZFace is flat
On the target of face,ZThe center of circle of the axis perpendicular to upper left corner dot；
3) acquire target image, through image procossing obtain the center of circle of all characteristic points pixel coordinate (x i ,y i )；
4) because plane target drone be precision instrument, target up and down adjacent two characteristic point absolute position it is known that
Therefore can obtain world coordinates corresponding to each target characteristic point (X wi ,Yw i )；
5) when obtain round steel end face center point pixel coordinate (x,y) after, according to above-mentioned pixel coordinate obtained by calibrating and
The transformational relation of world coordinates, can acquire corresponding round steel end face center point world coordinates (X W ,Y W )。
Centroid method seeks the image coordinates value of characteristic point:
The pixel coordinate and world coordinates value of each target dot center point of target are most importantly determined during this.Characteristic point
The acquisition of pixel coordinate: application removes the area etc. in irrelevant region, finally obtains target feature with Gauss weighted mass center method
The position of point.It extracts on target after the image coordinates of all characteristic points, it is known that the world coordinates of each characteristic point and each round steel
The image coordinates of end face center point can be obtained round steel end face with Delaunay subdivision triangular interpolation value camera calibration method
The world coordinates of central point.
When using Delaunay subdivision triangular interpolation value camera calibration method, it is desirable that the pixel coordinate of the characteristic point of target
Value and world coordinates value are one-to-one relationships, however the pixel coordinate for the target characteristic point for applying centroid method to obtain is by column
Arrangement, however the coordinate of each column is random, therefore uses data processing method, make the pixel coordinate of characteristic point by
It shows small to big arrangement.
The acquisition of characteristic point world coordinates value:
Scaling board up and down adjacent two characteristic point absolute position it is known that set target parameters line number asm；Columns isn；Row
Spacing isd m ；Column pitch isd n .World coordinates corresponding to the characteristic point center of circle (X wi ,Y wi ).
Many triangles will be formed after characteristic point triangulation, round steel end face center point pixel coordinate is included in composition
Triangle in.It can be acquired by the image coordinate value of the corresponding relationship and target point of characteristic point world coordinates and image coordinate
World coordinates value.
4, Image Acquisition, image enhancement processing and image binaryzation is carried out respectively to main phase machine and auxiliary camera to handle
Vision system carries out Image Acquisition to main phase machine and auxiliary camera first respectively, obtains the original graph of bundled round steel end face
Picture.
Image enhancement enhances method using greyscale transformation, carries out stereo pairs increasing using function imadjust in MATLAB
By force, the tonal range to be converted can be automatically obtained by function stretchlim in original image in the function, can also be used solid
Definite value can make system operational speed faster using fixed value in the case where image capture environment is constant, and this system is using solid
Definite value, it is verified [0.5;It is optimal when 1].The transformed tonal range of image uses as default i.e. 0-255.
Image binaryzation be in MATLAB by im2bw function by greyscale image transitions be bianry image, using
Threshold parameter must be set first when im2bw function, threshold value can be obtained by graythresh function automatically, can also be used
Fixed threshold.Since this system image zooming-out is carried out under light conditions, the image stabilization of acquisition, not using fixed threshold
Picture quality only can be improved, but also system operational speed can be improved, therefore this system uses fixed threshold, verified threshold
Value is optimal when being 0.6.
5, image is carried out to main phase machine and auxiliary camera respectively and carries out round identification, centralized positioning
Hough transform is to identify one of the basic skills of geometry, the function in MATLAB in image procossing
Imfindcircles is exactly that round identification is carried out with Hough transform.It need to be determined when imfindcircles function use several
A parameter: first is that round radius need to be detected, the radius of verified this system is set as [50 90] and is preferred；Two for and
Other background, there is ' bright ' and ' dark ', and this system is set as ' bright '；Three be parameter ' Sensitivity ' spirit
Sensitivity, the range of sensitivity is between [0,1], and sensitivity is bigger, and the circle that can be detected is more, then the risk of error detection is also therewith
Increase, it is that effect is preferable that verified sensitivity, which is 0.95,；The last one parameter is ' EdgeThreshold ' edge gradient threshold,
Its range is between [0,1], and the smaller circle that can be detected of edge gradient threshold is more, and error detection is also bigger therewith, works as edge
Effect is best when Grads threshold is 0.7.
The circle fitted in round identification is counted, the sum for recording the circle fitted is the root of round steel
Number.The circular centre coordinate come is fitted comprising all in the return value centers of function imfindcircles, is utilized
Size function can read circle number included in centers, this numerical value is the quantity of round steel, and record number is to reach
To the function of counting.
6, the online radical and diameter information for reading bundled round steel in company information library in real time
Corporate intranet is connected by the computer of system, the online production information read in company information library in real time is most closed
Key be current bundled round steel radical and diameter information.Wherein, the total weight of bundled round steel be practical weighing obtain, single circle
The weight of steel is calculated by material, diameter and length information, and the radical of bundled round steel is the total weight with bundled round steel
The estimation result obtained divided by the theoretical calculation weight of single round steel, may have 1-2 for the spillikin of different size
The error of root.
7, comprehensive descision is carried out to the round steel end face center point that main phase machine and auxiliary camera obtain according to the diameter information of round steel,
It determines the radical and reliability that bundled round steel visual identity goes out and carries out central point distance verification
Pass through the image recognition result of the image recognition operations of abovementioned steps available 1 main phase machine and 3 auxiliary cameras.
Wherein, the axis at the center of the bundled round steel of main phase machine face, 3 auxiliary cameras and bundled round steel is a certain included angle, that is to say, that 4
A camera has respectively obtained the image of the bundled round steel end face of different perspectives and has been identified and positioned.If all visual angles are all
It does not block, then all round steel can be identified and be positioned, and the end face center point of these round steel should be with uniformity.If
Some visual angle, which exists, blocks, then can not identify and position the round steel blocked on the image.Show Ge Gexiang by test of many times
Machine is less than 0.6mm, root of this precision for determining round steel using the error range of aforementioned image recognition and location of the core method
Number is enough.According to the diameter information of round steel, it can be determined which puts corresponding same root round steel, and also may determine that recognition result can
By property.
Such as: the specification of online bundled round steel is diameter 20mm, if 4 that 1 main phase machine and 3 auxiliary cameras identify
Central point, which is gathered in the circle of a small diameter, (can set greater than 0.6mm and be much smaller than some value of round steel diameter, generally
Can be set to the 10% of diameter, be here exactly 2mm), it may be considered that the central point of corresponding same root round steel, identifies at this time
Reliability is highest, can be set to 1.If there is the central point of 3 cameras identification is gathered in the circle of a small diameter, then may be used
To think the central point of corresponding same root round steel, the reliability identified at this time is time height, can be set to 0.8.And so on, it obtains
The radical and reliability of the round steel of identification.
The central point of all 4 cameras identifications come out is weighted summation and obtains the comprehensive identification of all round steel central points
Coordinate seeks distance to all synthesis coordinate identifications two-by-two, and all distance values should be greater than round steel diameter, if meeting
Pass through verification.In view of the location error of identification, the 90%(that all distance values are all larger than diameter is exactly 18mm here), also have
Standby higher reliability, can pass through verification.If all distance values are exactly 18mm in the presence of the 90%(less than diameter here)
Situation, then relevant two central points may be wrong.
8, the calculated result in the visual identity result and company information library of more bundled round steel radical, comprehensive descision
The actual test operating condition in conjunction with production scene is needed specifically to classify when comprehensive descision, it was therefore concluded that, such as:
(1) central point distance verification has been passed through, identification radical is consistent with the radical in company information library, then identification is correct；
(2) in all central point identification certainties there are reliability be 0.2 the case where, by central point apart from school
It tests, identifies that the radical in radical and company information library is inconsistent, then identify mistake, be subject to company information library；
(3) the case where there are reliability is 0.2 in all central point identification certainties has passed through central point apart from school
Test, identify that the radical in radical and company information library is inconsistent, then recognition result must by manpower intervention, correct recognition result or
The radical information in company information library.
Claims (1)
1. using the identification of bundled round steel, space orientation and the method for counting of multi-vision visual system, this method is more using bundled round steel
Mesh visual identifying system, for identification, space orientation and the counting of bundled round steel end face, the system comprises: light source (1), it is main
Camera (2), main phase machine and light source bracket (3), auxiliary camera (4), auxiliary camera support (5), computer (6) and scaling board (7), light source
It (1) is annular light source, main phase machine (2) is in the looping pit of annular light source (1) and light source (1) is coaxially arranged, light source (1) and main phase
Machine (2) is installed and fixed by main phase machine and light source bracket (3), and auxiliary camera (4) is installed and fixed by auxiliary camera support (5),
The axis radial equipartition of multiple auxiliary cameras (4) and camera support (5) around main phase machine (2), main phase machine (2) and multiple auxiliary cameras
(4) multi-vision visual image collecting device is formed, when installation, makes light source (1) and main phase by adjusting main phase machine and light source bracket (3)
The axis of machine (2) is made multiple perpendicular to the center of the whole end face of bundled round steel by the multiple auxiliary camera supports (5) of adjusting
The axis of auxiliary camera (4) and the axis of main phase machine (2) intersect at a point, and the point is made to be located at the theory of bundled round steel entirety end face
Center, light source (1), main phase machine (2), multiple auxiliary cameras (4) are connected with computer (6) by cable, computer (6) installation
There are visual identity and processing routine, and access corporate intranet, production information can be read in real time online, scaling board (7) is used for
The vision calibration of main phase machine (2) and multiple auxiliary cameras (4), calibration process need to only carry out primary after this system is correctly installed；
Described method includes following steps:
(1) the good bundled round steel multi-vision visual identifying system of installation and debugging；
(2) scaling board is arranged；
(3) monocular calibration is carried out using Delaunay subdivision triangular interpolation value standardization to main phase machine and auxiliary camera, after the completion
Withdraw scaling board；
(4) Image Acquisition, image enhancement processing and image binaryzation is carried out respectively to main phase machine and auxiliary camera to handle；
(5) identification of image circle, central space positioning are carried out to main phase machine and auxiliary camera respectively, main phase machine and auxiliary camera are distinguished
Carry out the identification of image circle, centralized positioning uses Hough transform, the return value of function imfindcircles in MATLAB
The circular centre coordinate come is fitted comprising all in centers, reads circle included in centers using size function
Shape number is the quantity of round steel；
(6) the online radical and diameter information for reading bundled round steel in company information library in real time；
(7) comprehensive descision is carried out to the round steel end face center point that main phase machine and auxiliary camera obtain according to the diameter information of round steel, really
Radical and reliability that fixed bundled round steel visual identity goes out simultaneously carry out central point distance verification, specifically: according to all main phase machines
With the diameter of a circle of the round steel end face center point aggregation of auxiliary camera identification and the relationship of round steel diameter, reliability step is set, is obtained
To the radical and reliability of the round steel of identification；The round steel end face center point coordinate identified to all main phase machines and auxiliary camera adds
Power summation obtains the synthesis coordinate identification of all round steel central points, distance is sought two-by-two to all synthesis coordinate identifications, if institute
Some distance values are all larger than setting value, then by verification, if all distance values are respectively less than setting value, in relevant two
Heart point may be wrong, completes central point distance verification；
(8) calculated result in the visual identity result and company information library of more bundled round steel radical, comprehensive descision obtain knot
By, specifically: if having passed through central point distance verification, identification radical is consistent with the radical in company information library, then identifies just
Really；If the case where being less than setting value there are reliability in all central point identification certainties does not pass through central point apart from school
It tests, identifies that the radical in radical and company information library is inconsistent, then identify mistake, be subject to company information library；If all
The case where there are reliability is less than setting value in central point identification certainty has passed through central point distance verification, identifies radical
Inconsistent with the radical in company information library, then recognition result must correct recognition result or company information library by manpower intervention
Radical information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810303608.XA CN108507484B (en) | 2018-04-08 | 2018-04-08 | Bundled round steel multi-vision visual identifying system and method for counting |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810303608.XA CN108507484B (en) | 2018-04-08 | 2018-04-08 | Bundled round steel multi-vision visual identifying system and method for counting |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108507484A CN108507484A (en) | 2018-09-07 |
| CN108507484B true CN108507484B (en) | 2019-09-24 |
Family
ID=63380536
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810303608.XA Active CN108507484B (en) | 2018-04-08 | 2018-04-08 | Bundled round steel multi-vision visual identifying system and method for counting |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108507484B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109741303B (en) * | 2018-12-25 | 2020-09-01 | 浙江中烟工业有限责任公司 | Filter stick counting method and device based on image acquisition and recognition |
| CN109470170B (en) * | 2018-12-25 | 2020-01-07 | 山东大学 | Stereoscopic vision space circular attitude high-precision measurement method and system based on optimal projection plane |
| CN109775055B (en) * | 2019-01-08 | 2021-06-04 | 河北科技大学 | Vision-based method for detecting label missing on end face of bundled bar materials and measuring error |
| CN112215846B (en) * | 2020-12-10 | 2021-03-23 | 江苏金恒信息科技股份有限公司 | Billet counting method and system based on three-dimensional point cloud |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105865329A (en) * | 2016-03-31 | 2016-08-17 | 河北科技大学 | Vision-based acquisition system for end surface center coordinates of bundles of round steel and acquisition method thereof |
| CN105953730A (en) * | 2016-06-22 | 2016-09-21 | 首航节能光热技术股份有限公司 | Multi-camera solar heat collector steel structure support assembling quality detection system |
| CN207946065U (en) * | 2018-04-08 | 2018-10-09 | 河北科技大学 | Bundled round steel multi-vision visual identifying system |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH09318351A (en) * | 1996-05-24 | 1997-12-12 | Kokusai Electric Co Ltd | Distance survey apparatus and size measuring method of object |
-
2018
- 2018-04-08 CN CN201810303608.XA patent/CN108507484B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105865329A (en) * | 2016-03-31 | 2016-08-17 | 河北科技大学 | Vision-based acquisition system for end surface center coordinates of bundles of round steel and acquisition method thereof |
| CN105953730A (en) * | 2016-06-22 | 2016-09-21 | 首航节能光热技术股份有限公司 | Multi-camera solar heat collector steel structure support assembling quality detection system |
| CN207946065U (en) * | 2018-04-08 | 2018-10-09 | 河北科技大学 | Bundled round steel multi-vision visual identifying system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108507484A (en) | 2018-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108507484B (en) | Bundled round steel multi-vision visual identifying system and method for counting | |
| CN110659636B (en) | Pointer instrument reading identification method based on deep learning | |
| CN102376089B (en) | Target correction method and system | |
| CN107133983B (en) | Bundled round steel end face binocular vision system and space orientation and method of counting | |
| CN110599541A (en) | Method and device for calibrating multiple sensors and storage medium | |
| CN102063718A (en) | Field calibration and precision measurement method for spot laser measuring system | |
| CN104683793B (en) | Detecting method for camera | |
| WO2006098258A1 (en) | Calibration method for 3-dimensional measurement and 3-dimensional measurement system using the method | |
| CN110146030A (en) | Side slope surface DEFORMATION MONITORING SYSTEM and method based on gridiron pattern notation | |
| JP2011182397A (en) | Method and apparatus for calculating shift length | |
| WO2014011182A1 (en) | Convergence/divergence based depth determination techniques and uses with defocusing imaging | |
| CN207946065U (en) | Bundled round steel multi-vision visual identifying system | |
| CN109163888A (en) | Optical centre test method, device and equipment | |
| CN105953741A (en) | Steel structure local geometric deformation measurement system and method | |
| CN211374003U (en) | Lens testing device | |
| CN105423975A (en) | Calibration system and method of large-size workpiece | |
| CN112767364B (en) | Image detection system for gate blade surface corrosion and corrosion area rapid measuring and calculating method | |
| US20190273845A1 (en) | Vibration monitoring of an object using a video camera | |
| CN112902869B (en) | Method and device for adjusting laser plane of rail profile measuring system | |
| CN103575742B (en) | One-touch optical measuring instrument | |
| Frobin et al. | Automatic Measurement of body surfaces using rasterstereograph | |
| CN107527323B (en) | Calibration method and device for lens distortion | |
| CN204944449U (en) | Depth data measuring system | |
| CN102826209A (en) | Method for realizing stereo shooting of ship draft image by using one-armed wall-climbing robot | |
| CN112361989B (en) | Method for calibrating parameters of measurement system through point cloud uniformity consideration |
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 |