CN106705884A - Product dimension detection method - Google Patents
Product dimension detection method Download PDFInfo
- Publication number
- CN106705884A CN106705884A CN201710009549.0A CN201710009549A CN106705884A CN 106705884 A CN106705884 A CN 106705884A CN 201710009549 A CN201710009549 A CN 201710009549A CN 106705884 A CN106705884 A CN 106705884A
- Authority
- CN
- China
- Prior art keywords
- detected
- size
- product
- image
- testing 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.)
- Pending
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
- G01B11/24—Measuring arrangements characterised by the use of optical techniques for measuring contours or curvatures
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
Provided is a product dimension automatic detection method, which comprises the following steps: carrying out system initialization and providing at least two camera devices; presetting coordinates of the at least two camera devices; increasing a detection position for dimension detection for a product standard part; driving the at least two camera devices to be corresponding to at least two end points of a position to be detected of the product standard part respectively; obtaining coordinates of the at least two end points of the position to be detected through shooting of the camera devices, and calculating detection dimensions according to the coordinates of the end points to obtain a standard value of a detection dimension; providing a dimension to be detected of a member to be detected, and driving the dimension to be corresponding to the camera devices corresponding to the at least two end points of the position to be detected; and obtaining coordinates of at least two end points of a position corresponding to the dimension to be detected through shooting of the camera devices, calculating the value of the dimension to be detected of the member to be detected according to the coordinate values of the end points and comparing the value with the standard value of the corresponding detection dimension. The method can measure various dimensions of a product through the double-camera-device principle, and thus the test result is allowed to be more precise.
Description
Technical field
The present invention relates to field of visual inspection, more particularly to product size visible detection method.
Background technology
In modern mechanical industrial automation production, for the product produced, it is related to various inspections, example
Such as the various appearance and dimension checks of product, product printing quality check, the bar code in product packaging and character recognition.This kind of product
Common feature be it is continuous produce in enormous quantities, the requirement to presentation quality it is very high.But, it is this with high reproducibility at present
Completed by visually with intelligent work, but at some in particular cases, accurate quick measurement such as to microsize,
Form fit and color-identifying etc., cannot continuously and stably be carried out at all by naked eyes, and other physical quantity transducers are also difficult to
It is competent.And artificial independent measurement time-consuming, efficiency is low, detection error largely influences product percent of pass, and measurement is multiple
Speed is slower during size, and the mode of this manual detection can cause high, poor quality of product cost etc..
The content of the invention
The technical problem to be solved in the present invention is to provide a kind of method that automatic detection can be carried out to product size.
In order to solve the above technical problems, the present invention provides following technical scheme:
A kind of product size automatic testing method, comprises the following steps:
Step one:System initialization, there is provided at least two image-taking devices;
Step 2:The coordinate of default at least two image-taking devices;
Step 3:Increase a product standard part test position of detected size;
Step 4:At least two image-taking devices are driven with least two-end-point of corresponding product standard component position to be detected respectively;
Step 5:Image-taking device shoots the coordinate for obtaining position to be detected at least two-end-point, and the coordinate value according to end points calculates inspection
Size is surveyed, the standard figures of detected size is obtained;
Step 6:The size that one part to be detected need to be detected is provided, and drives detected size correspondence test position at least two-end-point
Image-taking device;
Step 7:Image-taking device shoots the coordinate for obtaining size correspondence position to be detected at least two-end-point, according to extreme coordinates value
The size of part detected size to be detected is calculated, and is compared with the standard figures of corresponding detected size.
Wherein, it is further comprising the steps:An error range value X is preset on the basis of detected size standard figures,
If the size of part detected size to be detected with the error range value X of the standard figures of corresponding detected size, it is to be detected
Part is qualified, and otherwise, part to be detected is unqualified.
Wherein, it is further comprising the steps after the step 5:Parameter optimization, shape are carried out to detected size standard figures
Into detected size canonical parameter, if the error model of the size of part detected size to be detected and the canonical parameter of corresponding detected size
Enclose within value X, then part to be detected is qualified, otherwise, part to be detected is unqualified.
Wherein, the parameter optimization is profile optimiged index.
Wherein, the parameter optimization is contrast optimization.
Wherein, the parameter optimization is image smoothing optimization.
Wherein, following steps are further included:Preserve detected size canonical parameter and error range value X.
The beneficial effects of the invention are as follows:The needles of various sizes of product is tested by using double image-taking device principles, is made
Obtain test result more accurate.
Brief description of the drawings
Fig. 1 is specific embodiment of the invention product size automatic testing method flow chart.
Specific embodiment
Fig. 1 is referred to, is the specific implementation method flow chart of product size detection method of the present invention, present invention specific implementation
Example is mainly used in the various sizes of test product, and such as power hole length and width of measurement mobile phone, volume key length and width, the back side day linear slit is wide
The interior length and inner width equidimension of degree, SIM holes length and width and whole product, to make the result of detection more accurate, the present invention is specific real
Example is applied for the method using at least two image-taking device combined measurements, is comprised the following steps that.
S100:System initialization, there is provided at least two image-taking devices(Such as machine phase).And fill such as capture of various types of hardware equipment
Put, laser aid, motion controller, serial communication IO cards(Such as control device alarm and tri coloured lantern indication device)Deng with it is main
System is connected.
S110:The plane coordinates of default at least two image-taking devices.By round dot scaling method to all image-taking devices
Image physical coordinates(That is plane coordinates)Demarcated, in units of millimeter, i.e., represented image pixel positions with physical unit.
In order that the size of detection is more accurate, large scale of the present embodiment in detection product(More than single image-taking device coverage)
When, all shot using at least two image-taking devices, i.e.,:The two ends for treating detected size are shot, then further according to
The positional information that two image-taking devices shoot obtains the coordinate value of two-end-point respectively, during if necessary to the N number of size for detecting product, can
To set 2N image-taking device, and the plane coordinates to all image-taking devices is preset respectively.Lifted with the USB holes of mobile phone
Example:Go to shoot a USB hole with an image-taking device and gone to shoot a same USB hole with two image-taking devices, for USB holes
Edge for if, hence it is evident that be two image-taking devices go shoot a complete USB hole it is more accurate, taken by using two
As the coordinate in the image on the both sides in the USB holes of device shooting, then they are converted to corresponding physical coordinates.Further according to these
Coordinate obtains the interior length in USB holes, inner width equidimension.
S120:Increase a product standard part test position of detected size.That is, the product standard part tool to be detected is selected
Width at the top of body size, such as handset shell, the position that can be directed to case top two ends sets test position respectively.
S130:At least two image-taking devices are driven with least two-end-point of corresponding product standard component position to be detected respectively.System
System can preset the corresponding image-taking device numbering of each size two-end-point according to the position of size to be detected, specifically shoot
When, complete to take pictures by need to only selecting the corresponding image-taking device numbered, certainly, size detecting method of the present invention is not limited to only determine
One air line distance, it is also possible to the multiple distances of test, such as girth of triangle, can be respectively provided with three cameras at three ends
Point, then by three, the face plane coordinates of end points of taking pictures to obtain, and then the girth of triangle is drawn, equally, it is also possible to for testing
The size of more Multiple Shape, if the plane coordinates for drawing each point is shot by multipoint positioning, then logical coordinate values calculate corresponding
Size.
S140:Image-taking device shoots the coordinate for obtaining position to be detected at least two-end-point, and the coordinate value according to end points is calculated
Detected size size, obtains the standard figures of detected size.I.e.:When certain size exceedes the coverage of single image-taking device
When, two end values that detected size is treated by least two image-taking devices are taken pictures, and obtain the plane coordinates of two-end-point
Value, the plane coordinate value further according to the two-end-point calculates the distance between size endpoints thereof to be detected, and then draws to be detected
The numerical values recited of size, i.e.,:Demarcated by by two image-taking devices, such that it is able to obtain the figure of one of image-taking device
As the physical distance at edge to the image border of another image-taking device.
S150:Parameter optimization is carried out to detected size standard figures, detected size canonical parameter is formed.Test position is entered
Line parameter optimizes, and default parameters can be generated when test position is added, and whether the system can observe measurement result with real-time detection
Correctly, such as the situation that edge offsets or can not find, profile index, contrast and smoothing parameter can be changed.Detection is set again
The margin of tolerance of position, can call tolerance set in advance to be compared during output result.
Wherein, profile index:Contour detecting refers to the object bounds in detection image, and it is right in concern upper strata semanteme to be more biased towards
As.Such as OpenCV(Open Source Computer Vision Library, cross-platform computer vision library)In
findContours(Contour extraction)Function, it can obtain each profile and be stored with a vector mode, except this also obtains one
The index of latter profile of the profile of the topology information of individual image, i.e., previous profile, father's profile and embedded profile is compiled
Number.
Contrast:It is the ratio of B&W, that is, the gradual change level from black to white, ratio is bigger, from black to white gradually
Become level more, computing formula c=∑ r (i, j) * r (i, j) * p (i, j) of contrast, r (i, j)=| i-j | i.e. adjacent pixels
Between gray scale difference, p (i, j) is that adjacent pixel gray scale difference is the pixel distribution probability of r, and picture contrast is higher, then contrast
It is more stable that parameter can set more conference.
Image smoothing:The system uses Gaussian Blur filtering process, and Gaussian Blur is a kind of convolution blur operation of bidimensional,
Gaussian Blur is completed for average is fuzzy in image, and amount of calculation can increase, but Gaussian Blur can realize that some are special
Special effect really, particularly in picture noise(Non- salt-pepper noise)Cancellation aspect, even more there is extraordinary effect, the smooth ginseng of the system
Number represents the standard variance in Gaussian function, and recommendation is 1.0 ~ 5.5.
Certainly, if product test required precision is not high, it is also possible to not to the criterion numeral of the detected size of product standard part
Value is optimized, and the standard figures of the detected size for directly being obtained using step S140 is basic as comparing.
S160:An error range value X is preset on the basis of detected size canonical parameter.Assuming that the product size standard value
Parameter value is A, and error range value is X, and measured value is D, and the result of determination of D is:(A-X)<D<(A+X), if not in this model
In enclosing, then it represents that test result does not pass through.By that after the canonical parameter that measurement draws detected size, can be produced according to detection is actually subjected to
The quality requirements of product, set the value range of X on the basis of the canonical parameter.
S170:Preserve detected size canonical parameter and error range value X.
S180:Whether product standard part detected size is added in continuation, and the system can add product standard part multiple sizes
Test position.If necessary to continue to add, then the step of repeat S120 to S170.Need not such as add again, then into S190
Step.
S190:The size that one part to be detected need to be detected is provided, and drives detected size correspondence test position at least two ends
The image-taking device of point.This process is similar to the process of S130, will not be repeated here.
S200:Image-taking device shoots the coordinate for obtaining size correspondence position to be detected at least two-end-point, according to extreme coordinates
Value calculates the size of part detected size to be detected, and compares with the canonical parameter of corresponding detected size.I.e.:According to selected
The particular location that two image-taking devices are taken pictures obtains two plane coordinate values of end points of the size, is calculated by the plane coordinate value
The actual size of part detected size to be detected, and result is compared with the canonical parameter of corresponding detected size.Can also
Directly the standard figures with detected size is compared.
S210:Judge whether comparison result is in error range value X.If comparison result is in error range value X,
The part to be detected is qualified, if it exceeds error range value X, then the part to be detected is unqualified.
Further, to make test result more accurate, can be further comprising the steps before step S140:
The light source that position to be detected uses is provided, in order to be directed to the retrochromism of product, light source uses combined light source, the system
Any light source can be called, for different test positions, single source and multiple light sources can be opened while using, so that
To picture quality higher, certainty of measurement and stability are improved.
The specific embodiment of the invention is tested the needles of various sizes of product by using double image-taking device principles so that
Test result is more accurate.Certainly, size detecting method of the present invention is not limited to only determine an air line distance, it is also possible to which test is multiple
The girth of distance, such as triangle, can be respectively provided with three cameras in three end points, then by take pictures three, face end points it is flat
Areal coordinate, and then the girth of triangle is drawn, equally, it is also possible to the size for testing more Multiple Shape, as long as fixed by multiple spot
Position shoots the plane coordinates for drawing each point, then logical coordinate values calculate corresponding size.
Presently preferred embodiments of the present invention is the foregoing is only, is not intended to limit the invention, it is all in essence of the invention
Any modification, equivalent and improvement made within god and principle etc., should be included within the scope of the present invention.
Claims (10)
1. a kind of product size automatic testing method, it is characterised in that comprise the following steps:
Step one:System initialization, there is provided at least two image-taking devices;
Step 2:The coordinate of default at least two image-taking devices;
Step 3:Increase a product standard part test position of detected size;
Step 4:At least two image-taking devices are driven with least two-end-point of corresponding product standard component position to be detected respectively;
Step 5:Image-taking device shoots the coordinate for obtaining position to be detected at least two-end-point, and the coordinate value according to end points calculates inspection
Size is surveyed, the standard figures of detected size is obtained;
Step 6:The size that one part to be detected need to be detected is provided, and drives detected size correspondence test position at least two-end-point
Image-taking device;
Step 7:Image-taking device shoots the coordinate for obtaining size correspondence position to be detected at least two-end-point, according to extreme coordinates value
The size of part detected size to be detected is calculated, and is compared with the standard figures of corresponding detected size.
2. product size automatic testing method as claimed in claim 1, it is characterised in that:It is further comprising the steps:
An error range value X is preset on the basis of detected size standard figures, if the size of part detected size to be detected and corresponding inspection
Within the error range value X of the standard figures for surveying size, then part to be detected is qualified, and otherwise, part to be detected is unqualified.
3. product size automatic testing method as claimed in claim 2, it is characterised in that:Further included after the step 5
Following steps:Parameter optimization is carried out to detected size standard figures, detected size canonical parameter is formed, if part to be detected is detected
The size of size is with the error range value X of the canonical parameter of corresponding detected size, then part to be detected is qualified, otherwise, to be checked
Survey part unqualified.
4. product size automatic testing method as claimed in claim 3, it is characterised in that:The parameter optimization is indexed for profile
Optimization.
5. product size automatic testing method as claimed in claim 3, it is characterised in that:The parameter optimization is that contrast is excellent
Change.
6. product size automatic testing method as claimed in claim 3, it is characterised in that:The parameter optimization is image smoothing
Optimization.
7. product size automatic testing method as claimed in claim 6, it is characterised in that:Further include following steps:Protect
Deposit detected size canonical parameter and error range value X.
8. product size automatic testing method as claimed in claim 1, it is characterised in that:Further included after the step 5
Following steps:Whether continuation addition product standard part detected size, if desired for continuing to add, then repeat step three is to step 5,
Need not such as add again, then into step 6.
9. product size automatic testing method as claimed in claim 1, it is characterised in that:Can further be wrapped before the step 5
Include following steps:The light source that position to be detected uses is provided.
10. product size automatic testing method as claimed in claim 1, it is characterised in that:The coordinate is plane coordinates.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710009549.0A CN106705884A (en) | 2017-01-06 | 2017-01-06 | Product dimension detection method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710009549.0A CN106705884A (en) | 2017-01-06 | 2017-01-06 | Product dimension detection method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106705884A true CN106705884A (en) | 2017-05-24 |
Family
ID=58908202
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710009549.0A Pending CN106705884A (en) | 2017-01-06 | 2017-01-06 | Product dimension detection method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106705884A (en) |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107389139A (en) * | 2017-08-03 | 2017-11-24 | 尤立荣 | Micrometeor vision measurement device and vision measuring method |
CN108036721A (en) * | 2017-12-06 | 2018-05-15 | 东莞盛翔精密金属有限公司 | A kind of high accuracy size detecting method |
CN108716890A (en) * | 2018-08-17 | 2018-10-30 | 苏州富鑫林光电科技有限公司 | A kind of high-precision size detecting method based on machine vision |
CN109141333A (en) * | 2018-07-23 | 2019-01-04 | 北京电子工程总体研究所 | Collecting method and system |
CN110345876A (en) * | 2019-06-10 | 2019-10-18 | 重庆惠科金渝光电科技有限公司 | Polarisation plate detection device and polarisation plate detection method, readable storage medium storing program for executing |
CN110500950A (en) * | 2018-05-16 | 2019-11-26 | Juki株式会社 | Check device |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101685002A (en) * | 2008-09-25 | 2010-03-31 | 比亚迪股份有限公司 | Method of length detection and system thereof |
CN102116610A (en) * | 2010-11-29 | 2011-07-06 | 科达斯特恩(常州)汽车塑件系统有限公司 | Automatic on-line detection method and device for size of automobile parts based on machine vision |
CN102305597A (en) * | 2011-04-29 | 2012-01-04 | 无锡众望四维科技有限公司 | Method for automatically detecting abnormal shape of injector part by using mechanical vision system |
CN103438803A (en) * | 2013-09-18 | 2013-12-11 | 苏州晓创光电科技有限公司 | Method for performing view-field-across accurate measurement on size of rectangular part through computer vision technology |
CN105509653A (en) * | 2015-11-30 | 2016-04-20 | 广州超音速自动化科技股份有限公司 | Method and system for visual measurement of profile of machine part |
-
2017
- 2017-01-06 CN CN201710009549.0A patent/CN106705884A/en active Pending
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101685002A (en) * | 2008-09-25 | 2010-03-31 | 比亚迪股份有限公司 | Method of length detection and system thereof |
CN102116610A (en) * | 2010-11-29 | 2011-07-06 | 科达斯特恩(常州)汽车塑件系统有限公司 | Automatic on-line detection method and device for size of automobile parts based on machine vision |
CN102305597A (en) * | 2011-04-29 | 2012-01-04 | 无锡众望四维科技有限公司 | Method for automatically detecting abnormal shape of injector part by using mechanical vision system |
CN103438803A (en) * | 2013-09-18 | 2013-12-11 | 苏州晓创光电科技有限公司 | Method for performing view-field-across accurate measurement on size of rectangular part through computer vision technology |
CN105509653A (en) * | 2015-11-30 | 2016-04-20 | 广州超音速自动化科技股份有限公司 | Method and system for visual measurement of profile of machine part |
Cited By (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107389139A (en) * | 2017-08-03 | 2017-11-24 | 尤立荣 | Micrometeor vision measurement device and vision measuring method |
CN107389139B (en) * | 2017-08-03 | 2023-01-24 | 尤立荣 | Micro-flow vision measuring device and vision measuring method |
CN108036721A (en) * | 2017-12-06 | 2018-05-15 | 东莞盛翔精密金属有限公司 | A kind of high accuracy size detecting method |
CN110500950A (en) * | 2018-05-16 | 2019-11-26 | Juki株式会社 | Check device |
CN109141333A (en) * | 2018-07-23 | 2019-01-04 | 北京电子工程总体研究所 | Collecting method and system |
CN108716890A (en) * | 2018-08-17 | 2018-10-30 | 苏州富鑫林光电科技有限公司 | A kind of high-precision size detecting method based on machine vision |
CN110345876A (en) * | 2019-06-10 | 2019-10-18 | 重庆惠科金渝光电科技有限公司 | Polarisation plate detection device and polarisation plate detection method, readable storage medium storing program for executing |
CN110345876B (en) * | 2019-06-10 | 2022-01-25 | 重庆惠科金渝光电科技有限公司 | Polarizing plate detection device, polarizing plate detection method and readable storage medium |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106705884A (en) | Product dimension detection method | |
CN106485275B (en) | A method of realizing that cover-plate glass is bonded with liquid crystal display positioning | |
CN104657711B (en) | A kind of readings of pointer type meters automatic identifying method of robust | |
CN109308722B (en) | Space pose measurement system and method based on active vision | |
Huang et al. | A fast and flexible projector-camera calibration system | |
CN105069789B (en) | Structure light dynamic scene depth acquisition methods based on coding grid template | |
CN105303514A (en) | Image processing method and apparatus | |
WO2005024720A2 (en) | Color edge based system and method for determination of 3d surface topology | |
CN104390591A (en) | Accurate positioning method for circular marker in large-sized curved plate measurement | |
CN105303573B (en) | The pin detection method and system of acupuncture needle class component | |
CN104766309A (en) | Plane feature point navigation and positioning method and device | |
CN111637834B (en) | Three-dimensional data measuring device and method | |
CN107816942A (en) | A kind of planar dimension measurement method based on cross structure light vision system | |
CN106033614B (en) | A kind of mobile camera motion object detection method under strong parallax | |
CN110966956A (en) | Binocular vision-based three-dimensional detection device and method | |
CN109765240A (en) | A kind of detection industrial part stitch defect device and method | |
CN113724193A (en) | PCBA part size and clearance high-precision visual measurement method | |
CN112001973B (en) | Quick three-dimensional human head measuring method based on digital speckle correlation | |
CN116168072A (en) | Multi-camera large-size vision measurement method and system | |
CN111738971A (en) | Circuit board stereo scanning detection method based on line laser binocular stereo vision | |
CN105631846A (en) | Detection method for circular figure | |
CN108198226B (en) | Ceramic color identification method, electronic equipment, storage medium and device | |
CN109948630A (en) | Recognition methods, device, system and the storage medium of target sheet image | |
CN116563391B (en) | Automatic laser structure calibration method based on machine vision | |
CN117152727A (en) | Automatic reading method of pointer instrument for inspection robot |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170524 |
|
WD01 | Invention patent application deemed withdrawn after publication |