CN109682309A - Black crystal panel mechanical parameter intelligent device for measuring and method based on machine vision - Google Patents
Black crystal panel mechanical parameter intelligent device for measuring and method based on machine vision Download PDFInfo
- Publication number
- CN109682309A CN109682309A CN201811571439.4A CN201811571439A CN109682309A CN 109682309 A CN109682309 A CN 109682309A CN 201811571439 A CN201811571439 A CN 201811571439A CN 109682309 A CN109682309 A CN 109682309A
- Authority
- CN
- China
- Prior art keywords
- crystal panel
- camera
- black crystal
- black
- cameras
- 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
- 239000011011 black crystal Substances 0.000 title claims abstract description 133
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000001514 detection method Methods 0.000 claims abstract description 29
- 239000013078 crystal Substances 0.000 claims abstract description 9
- 238000000691 measurement method Methods 0.000 claims abstract description 6
- 239000000463 material Substances 0.000 claims description 15
- 238000004364 calculation method Methods 0.000 claims description 14
- 235000004443 Ricinus communis Nutrition 0.000 claims description 6
- 239000004973 liquid crystal related substance Substances 0.000 claims description 5
- 230000005540 biological transmission Effects 0.000 claims description 4
- 238000012935 Averaging Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000013011 mating Effects 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000012360 testing method Methods 0.000 claims description 3
- 230000000007 visual effect Effects 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims 1
- 238000005259 measurement Methods 0.000 description 9
- 238000010586 diagram Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 238000005070 sampling Methods 0.000 description 3
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000007688 edging Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 239000000779 smoke Substances 0.000 description 2
- 239000007787 solid Substances 0.000 description 2
- 241000288673 Chiroptera Species 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
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
- G01B11/02—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness
- G01B11/022—Measuring arrangements characterised by the use of optical techniques for measuring length, width or thickness by means of tv-camera scanning
-
- 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
- G01B11/26—Measuring arrangements characterised by the use of optical techniques for measuring angles or tapers; for testing the alignment of axes
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- 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
- 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
- G06T2207/30108—Industrial image inspection
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Theoretical Computer Science (AREA)
- Quality & Reliability (AREA)
- Length Measuring Devices By Optical Means (AREA)
Abstract
The present invention provides the black crystal panel mechanical parameter intelligent device for measuring based on machine vision, it is related to technical field of machine vision, using four cameras, camera fixing device is using movable type, the model that black crystal panel can be inputted according to interface is adjusted No. 1 camera, the position of No. 3 cameras is allowed to be constantly in the surface in black the crystal panel upper left corner and the lower right corner, uneven, the error caused by thickness is inconsistent to eliminate black crystal panel bevelling.The present invention provides the black crystal panel mechanical parameter intelligent measurement method based on machine vision during handling picture, to obtain most accurate numerical value, straight line and circle are drawn using least square method is carried out after Hough straight-line detection again, to solve the problems, such as that it is not positive that black crystal panel enters assembly line placement, it is corrected, to obtain accurate numerical value, error is largely reduced.
Description
Technical field
The invention belongs to technical field of machine vision, are mainly used for industrial automation detection, and in particular to a kind of accurate survey
Measure the machine vision intelligent device for measuring and method of black crystal panel mechanical parameter (length and width, chamfering, diagonal line).
Background technique
Nowadays domestic manufacturing industry level is gradually increased constantly aggravates with market competition, and enterprise and client get over the quality of product
More to pay attention to, sampling Detection can be gradually replaced to meet the control to product quality by detecting to product 100%, and life in China is horizontal
It is growing so that people more go to enjoy life, electromagnetic oven, fireplace, smoke exhaust ventilator etc. not only make people's lives more square
Just and environment will not be endangered, therefore be widely used.
It is exactly so-called black crystal panel in the glass of the black on the surfaces such as electromagnetic oven, smoke exhaust ventilator, fireplace.Manufacture these families
Electrical appliance be unable to do without black crystal panel, manufactures black crystal panel while considering safety, and normative and aesthetics is also critically important.It is black
There are many type of crystal panel, and different clients requires the mechanical parameter of black crystal panel different.The current detection side of black crystal panel
Method cannot be detected all, the method that can only use sampling Detection, and use manual measurement, and the result measured every time is not accurate,
And efficiency is very low.
(1) the technical issues of solving
In view of the above shortcomings of the prior art, the present invention provides a kind of black crystal panel mechanical parameter intelligence based on machine vision
It is capable of measuring device and method；It is able to achieve the contactless precise measurement of the mechanical parameter of the black crystal panel of plurality of specifications, measurement accuracy
Up to 0.01mm, the specification of panel is measured from 100*100 to 400*600 (unit: mm)；Skill is measured using non-contact vision
Art realizes the mechanical parameter that black crystal panel is dynamically measured on assembly line, can the realization when conveyor belt speed reaches 1m/s
To the precise measurement of black crystal panel, time of measuring is very short, and one black crystal panel of measurement only needs 2s or even shorter, sufficiently saving
Manpower；100% detection may be implemented, have the advantages that error is small, time-consuming short.
(2) technical solution
In order to achieve the above object, the present invention is achieved by the following technical programs:
Black crystal panel mechanical parameter intelligent device for measuring based on machine vision, including camera fixing device, Image Acquisition
Device, image measuring apparatus, panel delivery platform；Described image acquisition device is installed on above the panel delivery platform and leads to
Camera fixing device is crossed to fix；Described image measuring device is connected with image collecting device, for receiving described image acquisition
The image that device transmission comes, is handled on industrial personal computer, unqualified finally by whether He Ge signal is transmitted to alarm, is reported
It is alert.
Further, panel delivery platform include chassis, conveyer belt, liquid crystal display, display holder, electric box,
Lower margin, castor；The conveyer belt is located at the top of chassis 6, the lower margin and the ground mating presence of castor, is located at chassis for totally six sets
Lower section play a supporting role, the electric box is located at the inside of chassis, and the display holder is fixed on above conveyer belt
Section material frame on, fixed liquid crystal display.
Further, the camera fixing device includes section material frame, L-type support, camera connector, sliding rail, sliding block；Profile
Frame is fixed on the right and left of conveyer belt, and L-type support is fixed on section material frame, sliding rail and sliding block be it is a set of, be fixed on L-type support
Two-arm on, camera connector is installed above sliding block, has industrial camera to be fixed on sliding block by camera connector.
Further, described image acquisition device include three 8,900,000 industrial cameras and 2,300,000 industrial cameras, 4
A light source, light source fixing frame；Four cameras wherein No. 1 8,900,000 industrial cameras be fixed on conveyer belt direction of advance right side L-type
On one arm of bracket, the other three camera is fixed on another arm of L-type support, wherein No. 4 2,300,000 industrial cameras are located at No. 2
Among 8900000 industrial cameras and No. 3 8,900,000 industrial cameras, for identification the arrival of plank with leave；Four light sources pass through light source
Fixed frame is fixed on section material frame surrounding.
Further, the industrial camera includes camera power supply socket, network cable jack, camera body, camera lens exposure, mirror
Head, lens focus；Camera power supply socket, network cable jack are separately positioned on the top of camera body, and the lower part of camera body is successively
Camera lens exposure, camera lens, lens focus are installed.
Further, the conveyer belt includes driving wheel, belt, driven wheel, and driving wheel, driven wheel are mounted on conveyer belt
Both ends are attached between driving wheel, driven wheel by belt.
Black crystal panel mechanical parameter intelligent measurement method based on machine vision, comprising the following steps:
The picture that S1, shooting need；
By the mechanical parameter inputted in operation interface, adjustment 1 and No. 3 cameras guarantee camera pair to suitable position
Four angles of black crystal panel are vertically to shoot, and black crystal panel is moved along conveyer belt direction, and No. 4 cameras recognize the upper of black crystal panel
Bottom edge, which enters after about 600 pixel of field range, to be started 1,2, No. 3 cameras of triggering and starts to take pictures, until the bottom of black crystal panel
Triggering is taken pictures again when leaving the visual field there are also 600 pixels, is then handled picture；
S2, the picture taken is handled；
S2.1, row gaussian filtering is carried out；
Two-dimensional Gaussian function is:
Wherein the quadratic sum y of x square respectively indicate be in neighborhood other pixels at a distance from center pixel in neighborhood,
What σ was represented is standard deviation, generates the Gaussian filter template of a 5*5, is taken using the center of template as coordinate origin
Sample；It determines the coordinate of template at various locations, is brought into Gaussian function, obtained value is exactly the coefficient of template；For window
Template size is (2k+1) * (2k+1), and the calculation formula of each element value is as follows in template:
Normalizing obtained coefficient is:The namely inverse of coefficients sum；
S2.2, binaryzation is carried out to image, Hough straight-line detection then is carried out to the image after binaryzation；
Using the duality of point and line, in straight-line detection task, i.e., in the straight line and polar coordinate space in image space
Point be it is one-to-one, the point in the straight line and image space in polar coordinate space is also one-to-one；In image space
Every straight line correspond to an independent point all in polar coordinate space to indicate；Any a part on straight line in image space
It is the same point that line segment is corresponding in polar coordinate space, and straight-line detection problem in image space is transformed into polar coordinate space
In to the test problems of point, complete straight-line detection task by finding peak value in polar coordinate space；
S2.3, Accurate Curve-fitting straight line is carried out with least square method；
By Hough straight-line detection to straight line sign on another figure, which is traversed, do not consider be overlapped straight line,
Two straight lines are finally obtained, straight line is overlapped and just refers to the difference of the ρ of two straight lines in the same direction less than 40 pixels or two straight lines
θ difference less than 0.5 degree i.e. think two straight lines be overlapped；
S2.4, two straight-line intersections, angle are asked, limit radius area and is fitted fillet with least square method；
The intersection point and angle of two straight lines, two linear equations difference are calculated according to two linear equations that previous step obtains
For y=k*x+b1And y=l*x+b2, then the intersecting point coordinate of two straight lines beObtain the angle of two straight lines
θ meets:The size of intersecting point coordinate and angle is obtained, the length and width and diagonal line of black crystal panel are acquired；Root
The range of fitting fillet is limited according to the coordinate of intersection point, the range within intersection point is fitted, and the marginal point exceeded is then given up；
S3, the size for calculating black crystal panel:
S3.1, the upper bottom edge for calculating black crystal panel and right edge length: when camera carries out that triggering is taken pictures for the first time No. 1
Camera shoots the upper left corner of black crystal panel, and No. 2 cameras shoot the upper right corner of black crystal panel, and No. 3 cameras shoot the right side of black crystal panel
The intersection point calculation of inferior horn, two straight lines that the image shot by No. 1 and No. 2 obtains goes out the length of black crystal panel upper bottom edge, calculates public
Formula is ω1=(x1+x2)*ξ+L12；Wherein, ω1It is the length of upper bottom edge, x1It is the intersection point of No. 1 image shot by camera fitting a straight line
The abscissa of coordinate, x2It is the abscissa of the intersecting point coordinate of No. 2 image shot by camera fitting a straight lines, ξ is ratio, L12It is No. 1 and 2
Number camera calibration；The intersection point calculation for two straight lines that the image shot by No. 2 and No. 3 obtains goes out the length of black crystal panel right edge,
Calculation formula is l1=(y2+x3)*ξ+L23；Wherein l1It is the length of right edge, y2It is No. 2 image shot by camera fitting a straight lines
The ordinate of intersecting point coordinate, x3It is the abscissa of the intersecting point coordinate of No. 3 image shot by camera fitting a straight lines, ξ is ratio, L23It is 2
Number and No. 3 camera calibrations；The lower left corner that No. 1 camera when second of triggering is taken pictures shoots black crystal panel, No. 2 cameras are carried out in camera
The lower right corner for shooting black crystal panel calculates the length of bottom；
S3.2, calculate the black cornerwise length of crystal panel: three sides of known black crystal panel and each angle fit straight line
Angle, calculate cornerwise length, when being shot according to first time the black crystal panel upper right corner fit come two straight lines folder
Angle is α, then is exactly that diagonal line meets with the hypotenuse that upper bottom edge and right edge formWherein l1It is the length of right edge, ω1It is the length of upper bottom edge, DRIt is catercorner length；
Another catercorner length is found out by right edge and bottom Lai calculation formula isWherein
l1It is the length of right edge, ω2It is the length of bottom, DLIt is catercorner length, β is that the black crystal panel lower right corner fits straight line
Angle；
S3.3, the left side edge lengths for calculating black crystal panel: it according to two known diagonal lines, finds out on the left of black crystal panel twice
Edge lengths, then averaging is left side edge lengths that we want；Angle is found out according to the straight line that four angles fit, it is known that
The angle that the black crystal panel upper left corner fits straight line is γ, and the angle that the black crystal panel lower left corner fits straight line is φ, according to public affairs
Formula can export l2Length, then average:
It can obtain:
All mechanical parameters of black crystal panel can be obtained.
Further, when step S1 shooting picture, No. 1 and No. 3 cameras it is removable, No. 1 camera when shooting for the first time
The upper left corner of black crystal panel is shot, No. 2 cameras shoot the upper right corner of black crystal panel, and No. 3 cameras shoot the lower right corner of black crystal panel,
Thus the picture at black three angles of crystal panel is obtained, when shooting for the second time, No. 1 camera shoots the lower left corner, and No. 2 cameras shoot bottom rights
Angle, No. 3 cameras are not taken pictures, and the picture at the 4th angle has thus just been obtained.
(3) beneficial effect
Beneficial effects of the present invention: black crystal panel mechanical parameter intelligent device for measuring and method based on machine vision, this
Invention uses four cameras, one of them is to trigger its excess-three camera for receiving signal, takes full advantage of the work of assembly line
With, shooting is completed in the moving process of black crystal panel, camera placed using L-type, can save a camera, saved at
This；Black crystal panel measuring device of the invention, camera fixing device can input black crystalline substance according to interface using movable type
The model of panel is adjusted No. 1 camera, the position of No. 3 cameras is allowed to be constantly in the black crystal panel upper left corner and the lower right corner just
Top, so that it is uneven to eliminate black crystal panel bevelling, error caused by thickness is inconsistent；The present invention is in processing picture process
In, to obtain most accurate numerical value, straight line and circle are drawn using least square method is carried out after Hough straight-line detection again, to solve
Black crystal panel enters assembly line and places not positive problem, is corrected, to obtain accurate numerical value, is largely subtracted
Small error；The present invention is able to achieve the contactless precise measurement of the mechanical parameter of the black crystal panel of plurality of specifications, and measurement accuracy can
Up to 0.01mm, measure the specification of panel from 100*100 to 400*600 (unit: mm), realize dynamically measured on assembly line it is black
The mechanical parameter of crystal panel can realize the precise measurement to black crystal panel, time of measuring when conveyor belt speed reaches 1m/s
It is very short, manpower is sufficiently saved, 100% detection may be implemented, has the advantages that error is small, time-consuming short.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 is measuring device structure chart of the present invention；
Fig. 2 is industrial camera structural schematic diagram；
Fig. 3 is camera connector and industrial camera connection schematic diagram；
Fig. 4 is conveyer structure schematic diagram；
Fig. 5 is measuring method flow chart of the present invention；
Fig. 6 is coordinate transform schematic diagram；
Fig. 7 is 8 polar axis schematic diagrames.
Description of symbols:
1, section material frame；2, camera connector；3, industrial camera；4, light source；5, conveyer belt；6, chassis；7, lower margin；8, lower margin
Wheel；9, electric box；10, light source fixing frame；11, tripod；12, L-type support；31, camera power supply socket；32, network cable jack；
33, camera body；34, camera lens exposes；35, camera lens；36, lens focus；51, driving wheel；52, belt；53, driven wheel.
Specific embodiment
In order to make the object, technical scheme and advantages of the embodiment of the invention clearer, below in conjunction with the embodiment of the present invention
In attached drawing, technical scheme in the embodiment of the invention is clearly and completely described, it is clear that described embodiment is
A part of the embodiment of the present invention, instead of all the embodiments.Based on the embodiments of the present invention, those of ordinary skill in the art
Every other embodiment obtained without creative efforts, shall fall within the protection scope of the present invention.
In conjunction with Fig. 1, a kind of black crystal panel mechanical parameter intelligent device for measuring based on machine vision, including the fixed dress of camera
It sets, image collecting device, image measuring apparatus, panel delivery platform；Described image acquisition device is installed on the panel transmission
It is fixed above platform and by camera fixing device；Described image measuring device is connected with image collecting device, for receiving
The image that the transmission of described image acquisition device comes, is handled on industrial personal computer, finally by whether He Ge signal is transmitted to alarm
Device, it is unqualified, it alarms.
Camera fixing device includes section material frame 1, L-type support 12, camera connector 2, conveyer belt 5, sliding rail, sliding block；Profile
Frame 1 is fixed on the right and left of conveyer belt 5, is highly 70cm；L-type support 12 is fixed on section material frame 1
60cm；Sliding rail and sliding block be it is a set of, be fixed in the two-arm of L-type support 12, camera connector 2 be installed above sliding block, is had
Industrial camera 3 is fixed on sliding block by camera connector 2, can be mobile with sliding rail.
Image collecting device includes three 8,900,000 industrial cameras and 2,300,000 industrial cameras, and 4 light sources 4, light sources are solid
Determine frame 10；Four cameras wherein No. 1 8,900,000 industrial cameras be fixed on 5 direction of advance of conveyer belt right side L-type support an arm
On, the other three camera is fixed on another arm of L-type support, wherein No. 4 2,300,000 industrial cameras are located at No. 2 8,900,000 industrial phases
Among machine and No. 3 8,900,000 industrial cameras, for identification the arrival of plank with leave；Four light sources 4 are solid by light source fixing frame 10
It is scheduled on section material frame surrounding, the position apart from conveyer belt 5cm.
Panel delivery platform include chassis 6, conveyer belt 5, liquid crystal display, display holder, electric box 9, lower margin 7,
Ground castor 8；The conveyer belt 5 is located at the top of chassis 6, the lower margin 7 and the ground mating presence of castor 8, is located at chassis 6 for totally six sets
Lower section play a supporting role, the electric box 9 is located at the inside of chassis 6, can store public machine, power supply etc., the display
Device fixed frame is fixed on the section material frame above conveyer belt, fixed liquid crystal display.Pacify between section material frame 1 and panel delivery platform
Equipped with tripod 11.
In conjunction with Fig. 2, Fig. 3, industrial camera includes camera power supply socket 31, network cable jack 32, camera body 33, camera lens exposure
34, camera lens 35, lens focus 36；Camera power supply socket 31, network cable jack 32 are separately positioned on the top of camera body 33, phase
The lower part of machine ontology 33 is sequentially installed with camera lens and exposes 34, camera lens 35, lens focus 36.
In conjunction with Fig. 4, conveyer belt 5 includes driving wheel 51, belt 52, driven wheel 53, and driving wheel 51, driven wheel 53 are mounted on biography
The both ends of band are sent, are attached between driving wheel 51, driven wheel 53 by belt 52.
When specific works, the model of plank is inputted in operation interface, the movement by motor with movable slider, that is, control
The movement of camera, No. 2 camera positions are fixed, and No. 4 cameras and No. 2 cameras are adjacent, adjust No. 1 and 3 according to the size of black crystal panel
The position of number camera, so that realize that camera vertically shoots plate as far as possible, to eliminate black crystal panel edging and thickness is different
It is influenced caused by cause.
The triggering mode that the present invention uses is external trigger, and then four cameras mutually trigger, i.e., No. 4 cameras, which are in, connects
Discontinuous Conduction mode, 1,2, No. 3 camera are in triggering mode, recognize in No. 4 cameras and trigger No. 2 cameras bats when plate comes at once
According to No. 2 cameras trigger No. 1 camera again and take pictures, and No. 1 camera triggers No. 3 cameras again and takes pictures.The triggering mode can guarantee to touch simultaneously
Hair, it is more preferable compared to sensor external triggering effect, it avoids because black crystal panel has edging, occurs when sensor being caused to identify serious
The defect of error.
In conjunction with Fig. 5, the black crystal panel mechanical parameter intelligent measurement method based on machine vision, comprising the following steps:
The picture that S1, shooting need；
By the mechanical parameter inputted in operation interface, adjustment 1 and No. 3 cameras guarantee camera pair to suitable position
Four angles of black crystal panel are vertically to shoot, and black crystal panel is moved along conveyer belt direction, and No. 4 cameras recognize the upper of black crystal panel
Bottom edge, which enters after about 600 pixel of field range, to be started 1,2, No. 3 cameras of triggering and starts to take pictures, until the bottom of black crystal panel
Triggering is taken pictures again when leaving the visual field there are also 600 pixels, is then handled picture.
Described No. 1 and No. 3 cameras it is removable, No. 1 camera shoots the upper left corner of black crystal panel when shooting for the first time, No. 2
Camera shoots the upper right corner of black crystal panel, and No. 3 cameras shoot the lower right corner of black crystal panel, it is hereby achieved that black crystal panel three
The picture at angle, when shooting for the second time, No. 1 camera shoots the lower left corner, and No. 2 cameras shoot the lower right corner, and No. 3 cameras are not taken pictures, thus
The picture for just having obtained the 4th angle, obtain four angles information can be carried out processing be calculated black crystal panel length and width,
The specific value of diagonal line, fillet.
S2, the picture taken is handled；It is divided into the following steps:
S2.1, row gaussian filtering is carried out；
Two-dimensional Gaussian function is:
Wherein the quadratic sum y of x square respectively indicate be in neighborhood other pixels at a distance from center pixel in neighborhood,
What σ was represented is standard deviation, to generate the Gaussian filter template of a 5*5, is carried out by coordinate origin of the center of template
Sampling；The coordinate of template at various locations, as follows, wherein horizontally to the right, (and y-axis is anti-for j axis for i axis (identical with x-axis position)
Direction) straight down:
| (-2,2) | (-1,2) | (0,2) | (1,2) | (2,2) |
| (-2,1) | (-1,1) | (0,1) | (1,1) | (2,1) |
| (-2,0) | (-1,0) | (0,0) | (1,0) | (2,0) |
| (-2,-1) | (-1,-1) | (0,-1) | (1,-1) | (2,-1) |
| (-2,-2) | (-1,-2) | (0,-2) | (1,-2) | (2,-2) |
In this way, the coordinate of each position is brought into Gaussian function, obtained value is exactly the coefficient of template；For window
Template size is (2k+1) * (2k+1), and the calculation formula of each element value is as follows in template:
The template used herein is integer type, is exactly being normalized after being finished with above-mentioned formula calculating
It arrives, the coefficient normalized is:The namely inverse of coefficients sum；
S2.2, binaryzation is carried out to image, Hough straight-line detection then is carried out to the image after binaryzation；
The basic principle of Hough straight-line detection is using point and the duality of line, in straight-line detection task, i.e. image
Point in straight line and polar coordinate space in space is one-to-one, the point in the straight line and image space in polar coordinate space
It is also one-to-one.This means that it can be concluded that two highly useful conclusions: every straight line in image space is sat in pole
Mark corresponds to an independent point all in space to indicate；Any a part of line segment is in polar coordinate space on straight line in image space
Corresponding is the same point, therefore Hough line detection algorithm is exactly that straight-line detection problem in image space is transformed into pole
To the test problems of point in coordinate space, straight-line detection task is completed by finding peak value in polar coordinate space.
Such as: to straight line y=k*x+b, ρ=x*cos θ+y*sin θ is represented by polar coordinates, that is to say, that straight
Straight line y=k*x+b in angular coordinate system can indicate that is, polar each point (ρ, θ) is all right in polar coordinate system with (ρ, θ)
The straight line of rectangular co-ordinate is answered, rectangular co-ordinate a point just corresponds to a curve in polar coordinates in other words.If figure
Pixel as in constitutes straight line, then these pixel coordinate values (x, y) corresponding curve in polar coordinates centainly intersects at
One point, so the curve for only needing to be transformed into all pixels point (coordinate value) in image under polar coordinates, and in polar coordinates
Middle detection curve intersection point is assured that straight line.
In practical application, it is necessary to which the quantity (i.e. the direction of limited quantity) for limiting straight line can be calculated.Such as Fig. 6 institute
Show, turn to limited equidistant discrete value for the direction θ of straight line is discrete, parameter ρ also just it is corresponding it is discrete turn to limited value, in
Be parameter space no longer it is continuous, but is the sizes grid cell such as one by one by discrete quantized.By image space, (right angle is sat
Mark system) in after each pixel coordinate value transforms to parameter space (polar coordinate system), resulting value can be fallen in some grid, make this
The summary counter of grid cell adds 1.When pixel all in image space is all after Hough transformation, to grid cell progress
It checks, accumulated counts are worth maximum grid, coordinate value (ρ0,θ0) with regard to straight line required in correspondence image space.That is such as
Fruit accumulated value is greater than threshold value and then thinks there is straight line here.
S2.3, Accurate Curve-fitting straight line is carried out with least square method；
By Hough straight-line detection to straight line sign on another figure, which is traversed, do not consider be overlapped straight line,
Finally obtain two straight lines, it is so-called be overlapped straight line just refer to the difference of the ρ of two straight lines in the same direction less than 40 pixels or
The difference of the θ of two straight lines thinks that two straight lines are overlapped less than 0.5 degree.
But the straight line that actually Hough straight-line detection is found is inaccurate, by taking obtained straight line as an example: setting
Hough straight line is ρ=x*cos θ+y*sin θ, more accurate for the straight line that draws, and in the two sides of the straight line, mobile several pixels are again
Straight line is respectively drawn, ρ ± μ=x*cos θ+y*sin θ can be expressed as, wherein μ indicates mobile several pixels, then at this
It is traversed in a channel, finds all points, be fitted with least square method, more accurate fitting a straight line can be obtained, due to most
It is fine to non-perpendicular straight line fitting effect when small square law is fitted, but fit slope effect when 90 ° neighbouring
It is bad, therefore invert the transverse and longitudinal coordinate of each point when being fitted vertical line, then be fitted, at this time obtain
Linear equation beY is showed with x again.But the straight line found in this way will receive the influence of environment and produce
Raw mistake, can also measure not enough precisely in the case where black crystal panel sideling enters conveyer belt, not be able to satisfy error in ± 0.1mm
Within, it is provided with calibration thus, (tilt angle is not greater than when default plate enters conveyer belt by taking the fitting of upper bottom edge as an example
45 °), even if how upper bottom edge, which tilts, is still thought that this straight line is upper bottom edge, r and θ is returned when Hough straight line fitting in opencv,
R will appear positive and negative values, in the counterclockwise direction polar angleWhen polar diameter r be it is positive,When polar diameter r be negative value.
8 lines as shown in Figure 7 are all polar axis, can be divided into 4 kinds of situations herein:
A:Polar diameter r is negative (because θ black crystal panel inclination in the second quadrant is excessive, it is impossible to occur at this time
Such case, so r does not take just), the left and right side of black crystal panel is all tilted to the right when contour line is perpendicular to polar axis, intends at this time
The straight line closed out is calibrated so that it is tilted to the right, and inaccurate feelings are fitted when solving to be tilted to the right when black crystal panel enters
Condition；
B:Polar diameter r is negative (because θ black crystal panel inclination in the second quadrant is excessive, it is impossible to go out at this time
Existing such case, so r does not take just), the upper bottom of black crystal panel is all upper inclination to the right when contour line is perpendicular to polar axis, this
When the straight line that fits calibrated so that its upper inclination to the right, fitting is not when solving to be tilted to the left when black crystal panel enters
Quasi- situation；
C:Polar diameter r divides positive and negative at this time, is upper bottom edge or bottom certainly when contour line is perpendicular to polar axis
Lower inclination to the right；The straight line fitted at this time is calibrated so that its lower inclination to the right, solve when black crystal panel enters to
Inaccurate situation is fitted when right bank；
D:Polar diameter r divides positive and negative at this time, is left side or right edge certainly when contour line is perpendicular to polar axis
It is tilted to the left；The straight line fitted at this time is calibrated so that it is tilted to the left, and solves to incline to the left when black crystal panel enters
Fitting inaccurate situation when tiltedly；
Ideal fitting a straight line has been obtained in the case.
S2.4, two straight-line intersections, angle are asked, limit radius area and is fitted fillet with least square method；
The intersection point and angle of two straight lines, two linear equations can be calculated according to two linear equations that previous step obtains
Respectively y=k*x+b1And y=l*x+b2, then the intersecting point coordinate of two straight lines beObtain two straight lines
Angle theta meets:The size of intersecting point coordinate and angle is obtained, it can be in the hope of the length and width of black crystal panel and right
Linea angulata.The range of fitting fillet can also be limited according to the coordinate of intersection point, that is, the range within intersection point is fitted,
The marginal point exceeded is then given up, and the precision of fitting fillet is so just improved.
S3, the size for calculating black crystal panel:
S3.1, the upper bottom edge for calculating black crystal panel and right edge length: when camera carries out that triggering is taken pictures for the first time No. 1
Camera shoots the upper left corner of black crystal panel, and No. 2 cameras shoot the upper right corner of black crystal panel, and No. 3 cameras shoot the right side of black crystal panel
The intersection point of inferior horn, two straight lines that the image shot by No. 1 and No. 2 obtains can calculate the length of black crystal panel upper bottom edge, meter
Calculation formula is ω1=(x1+x2)*ξ+L12；Wherein, ω1It is the length of upper bottom edge, x1It is No. 1 image shot by camera fitting a straight line
The abscissa of intersecting point coordinate, x2It is the abscissa of the intersecting point coordinate of No. 2 image shot by camera fitting a straight lines, ξ is ratio, L12It is 1
Number and No. 2 camera calibrations.The intersection point for two straight lines that the image shot by No. 2 and No. 3 obtains can calculate on the right side of black crystal panel
The length on side, calculation formula l1=(y2+x3)*ξ+L23；Wherein l1It is the length of right edge, y2It is No. 2 image shot by camera
The ordinate of the intersecting point coordinate of fitting a straight line, x3It is the abscissa of the intersecting point coordinate of No. 3 image shot by camera fitting a straight lines, ξ is
Ratio, L23It is No. 2 and No. 3 camera calibrations.The lower-left that No. 1 camera when second of triggering is taken pictures shoots black crystal panel is carried out in camera
Angle, No. 2 cameras shoot the lower right corner of black crystal panel, can similarly calculate the length of bottom.
S3.2, calculate the black cornerwise length of crystal panel: three sides of currently known black crystal panel and each angle fit
The angle of straight line can calculate cornerwise length, when being shot according to first time the black crystal panel upper right corner fit come two
The angle of straight line is α, then is exactly that diagonal line meets with the hypotenuse that upper bottom edge and right edge formWherein l1It is the length of right edge, ω1It is the length of upper bottom edge, DRIt is catercorner length.
Another catercorner length can be found out by right edge and bottom Lai calculation formula isIts
Middle l1It is the length of right edge, ω2It is the length of bottom, DLIt is catercorner length, β is that the black crystal panel lower right corner fits directly
The angle of line.
S3.3, the left side edge lengths for calculating black crystal panel: it according to two known diagonal lines, finds out on the left of black crystal panel twice
Edge lengths, then averaging is left side edge lengths that we want.Angle is found out according to the straight line that four angles fit, it is known that
The angle that the black crystal panel upper left corner fits straight line is γ, and the angle that the black crystal panel lower left corner fits straight line is φ, according to public affairs
Formula can export l2Length, then average:
It can obtain:
Can be obtained all mechanical parameters of black crystal panel, precision within 0.1mm and ensure black crystal panel bevelling and
Thickness does not impact measurement result.
In conclusion the embodiment of the present invention, black crystal panel mechanical parameter intelligent device for measuring and side based on machine vision
Method, using four cameras, camera fixing device using mobile, can be inputted according to interface the model of black crystal panel into
Row adjusts No. 1 camera, the position of No. 3 cameras is allowed to be constantly in the surface in black the crystal panel upper left corner and the lower right corner, to disappear
In addition to black crystal panel bevelling is uneven, error caused by thickness is inconsistent.The present invention provides the black crystal panel based on machine vision
Mechanical parameter intelligent measurement method is during handling picture, to obtain most accurate numerical value, after Hough straight-line detection
Least square method is carried out again and draws straight line and circle, to solve the problems, such as that it is not positive that black crystal panel enters assembly line placement, is rectified
Just, to obtain accurate numerical value, error is largely reduced.
The above embodiments are merely illustrative of the technical solutions of the present invention, rather than its limitations；Although with reference to the foregoing embodiments
Invention is explained in detail, those skilled in the art should understand that: it still can be to aforementioned each implementation
Technical solution documented by example is modified or equivalent replacement of some of the technical features；And these modification or
Replacement, the spirit and scope for technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution.
Claims (8)
1. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision, which is characterized in that including camera fixing device,
Image collecting device, image measuring apparatus, panel delivery platform；Described image acquisition device is installed on the panel delivery platform
Top is simultaneously fixed by camera fixing device；Described image measuring device is connected with image collecting device, described for receiving
The image that image collecting device transmission comes, is handled on industrial personal computer, finally by whether He Ge signal is transmitted to alarm, no
It is qualified then alarm.
2. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision as described in claim 1, it is characterised in that:
Panel delivery platform include chassis, conveyer belt, liquid crystal display, display holder, electric box, lower margin, castor；The biography
It send band positioned at the top of chassis 6, the lower margin and the ground mating presence of castor, support work is played below chassis for totally six sets
With the electric box is located at the inside of chassis, and the display holder is fixed on the section material frame above conveyer belt, fixer
Crystal display.
3. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision as claimed in claim 2, it is characterised in that:
The camera fixing device includes section material frame, L-type support, camera connector, sliding rail, sliding block；Section material frame is fixed on conveyer belt
The right and left, L-type support are fixed on section material frame, sliding rail and sliding block be it is a set of, be fixed in the two-arm of L-type support, sliding block
Top is equipped with camera connector, has industrial camera to be fixed on sliding block by camera connector.
4. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision as claimed in claim 3, it is characterised in that:
Described image acquisition device includes three 8,900,000 industrial cameras and 2,300,000 industrial cameras, four light sources, light source fixing frames；
Four cameras wherein No. 1 8,900,000 industrial cameras be fixed on conveyer belt direction of advance right side L-type support an arm on, in addition
Three cameras are fixed on another arm of L-type support, wherein No. 4 2,300,000 industrial cameras are located at No. 2 8,900,000 industrial cameras and No. 3
Among 8900000 industrial cameras, for identification the arrival of plank with leave；Four light sources are fixed on section material frame by light source fixing frame
Surrounding.
5. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision as claimed in claim 4, it is characterised in that:
The industrial camera includes camera power supply socket, network cable jack, camera body, camera lens exposure, camera lens, lens focus；It is mutually electromechanical
Source socket, network cable jack are separately positioned on the top of camera body, and the lower part of camera body is sequentially installed with camera lens exposure, mirror
Head, lens focus.
6. the black crystal panel mechanical parameter intelligent device for measuring based on machine vision as claimed in claim 2, it is characterised in that:
The conveyer belt includes driving wheel, belt, driven wheel, and driving wheel, driven wheel are mounted on the both ends of conveyer belt, driving wheel, driven
It is attached between wheel by belt.
7. the black crystal panel mechanical parameter intelligent measurement method based on machine vision, it is characterised in that: the following steps are included:
The picture that S1, shooting need；
By the mechanical parameter inputted in operation interface, adjustment 1 and No. 3 cameras guarantee camera to black crystalline substance to suitable position
Four angles of panel are vertically to shoot, and black crystal panel is moved along conveyer belt direction, and No. 4 cameras recognize the upper bottom edge of black crystal panel
Start to take pictures into 1,2, No. 3 cameras of triggering are started after about 600 pixel of field range, until the bottom of black crystal panel also
Triggering is taken pictures again when 600 pixels leave the visual field, is then handled picture；
S2, the picture taken is handled；
S2.1, gaussian filtering is carried out；
Two-dimensional Gaussian function is:
Wherein the quadratic sum y of x square respectively indicate is that other pixels are at a distance from center pixel in neighborhood in neighborhood, σ generation
Table is standard deviation, generates the Gaussian filter template of a 5*5, is sampled using the center of template as coordinate origin；
It determines the coordinate of template at various locations, is brought into Gaussian function, obtained value is exactly the coefficient of template；For template window
Size is (2k+1) * (2k+1), and the calculation formula of each element value is as follows in template:
Normalizing obtained coefficient is:The namely inverse of coefficients sum；
S2.2, binaryzation is carried out to image, Hough straight-line detection then is carried out to the image after binaryzation；
Point using the duality of point and line, in straight-line detection task, i.e., in the straight line and polar coordinate space in image space
It is correspondingly, the point in the straight line and image space in polar coordinate space is also one-to-one；It is every in image space
Straight line corresponds to an independent point all in polar coordinate space to indicate；Any a part of line segment on straight line in image space
Corresponding in polar coordinate space is the same point, and it is right in polar coordinate space that straight-line detection problem in image space is transformed into
The test problems of point complete straight-line detection task by finding peak value in polar coordinate space；
S2.3, Accurate Curve-fitting straight line is carried out with least square method；
By Hough straight-line detection to straight line sign on another figure, which is traversed, do not consider be overlapped straight line, finally
Obtain two straight lines, be overlapped straight line just refer to the difference of the ρ of two straight lines in the same direction less than the θ of 40 pixels or two straight lines it
Difference is less than 0.5 degree, and since Hough straight-line detection result is not accurate enough, We conducted both sides corrections, first: to what is obtained
Range where point on two straight lines or so expands, that is, forms a pipeline, and it is quasi- that least square method is carried out inside the pipeline
Close, be fitted if straight line is horizontal according to y=kx+b, if straight line be it is vertical according toIt is fitted, supports
The least square method that disappears can not accurately indicate the problem of vertical line, secondly: entering device deflection that may be present for black crystal panel
Problem is divided into four kinds of different situations and is calibrated, to obtain accurate fitting a straight line；
S2.4, two straight-line intersections, angle are asked, limit radius area and is fitted fillet with least square method；
The intersection point and angle of two straight lines are calculated according to two linear equations that previous step obtains, two linear equations are respectively y
=k*x+b1And y=l*x+b2, then the intersecting point coordinate of two straight lines beThe angle theta for obtaining two straight lines is full
Foot:The size of intersecting point coordinate and angle is obtained, the length and width and diagonal line of black crystal panel are acquired；According to friendship
The coordinate of point is fitted the range of fillet to limit, and the range within intersection point is fitted, and the marginal point exceeded is then given up；
S3, the size for calculating black crystal panel:
S3.1, the upper bottom edge for calculating black crystal panel and right edge length: No. 1 camera when camera carries out that triggering is taken pictures for the first time
The upper left corner of black crystal panel is shot, No. 2 cameras shoot the upper right corner of black crystal panel, and No. 3 cameras shoot the lower right corner of black crystal panel,
The intersection point calculation for two straight lines that the image shot by No. 1 and No. 2 obtains goes out the length of black crystal panel upper bottom edge, and calculation formula is
ω1=(x1+x2)*ξ+L12；Wherein, ω1It is the length of upper bottom edge, x1It is the intersecting point coordinate of No. 1 image shot by camera fitting a straight line
Abscissa, x2It is the abscissa of the intersecting point coordinate of No. 2 image shot by camera fitting a straight lines, ξ is ratio, L12It is No. 1 and No. 2 phases
Machine calibration；The intersection point calculation for two straight lines that the image shot by No. 2 and No. 3 obtains goes out the length of black crystal panel right edge, calculates
Formula is l1=(y2+x3)*ξ+L23；Wherein l1It is the length of right edge, y2It is the intersection point of No. 2 image shot by camera fitting a straight lines
The ordinate of coordinate, x3It is the abscissa of the intersecting point coordinate of No. 3 image shot by camera fitting a straight lines, ξ is ratio, L23It is No. 2 and 3
Number camera calibration；The lower left corner that No. 1 camera when second of triggering is taken pictures shoots black crystal panel, No. 2 camera shootings are carried out in camera
The lower right corner of black crystal panel calculates the length of bottom；
S3.2, calculate the black cornerwise length of crystal panel: three sides of known black crystal panel and each angle fit the folder of straight line
Angle calculates cornerwise length, when being shot according to first time the black crystal panel upper right corner fit come the angles of two straight lines be
α is exactly then that diagonal line meets with the hypotenuse that upper bottom edge and right edge formIts
Middle l1It is the length of right edge, ω1It is the length of upper bottom edge, DRIt is catercorner length；Another catercorner length by right edge and
Bottom, which is found out, to be come, and calculation formula isWherein l1It is the length of right edge, ω2It is bottom
The length on side, DLIt is catercorner length, β is the angle that the black crystal panel lower right corner fits straight line；
S3.3, the left side edge lengths for calculating black crystal panel: according to two known diagonal lines, side length on the left of black crystal panel twice is found out
Degree, then averaging is left side edge lengths that we want；Angle is found out according to the straight line that four angles fit, it is known that black crystalline substance
The angle that the panel upper left corner fits straight line is γ, and the angle that the black crystal panel lower left corner fits straight line is φ, can according to formula
Export l2Length, then average:
It can obtain:
All mechanical parameters of black crystal panel can be obtained.
8. the black crystal panel mechanical parameter intelligent measurement method based on machine vision as claimed in claim 7, it is characterised in that:
When step S1 shooting picture, No. 1 and No. 3 cameras it is removable, No. 1 camera shoots the upper left of black crystal panel when shooting for the first time
Angle, No. 2 cameras shoot the upper right corner of black crystal panel, and No. 3 cameras shoot the lower right corner of black crystal panel, thus obtain black crystal panel three
The picture at a angle, when shooting for the second time, No. 1 camera shoots the lower left corner, and No. 2 cameras shoot the lower right corner, and No. 3 cameras are not taken pictures, by
This has just obtained the picture at the 4th angle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811571439.4A CN109682309B (en) | 2018-12-21 | 2018-12-21 | Intelligent measuring device and method for black crystal panel parameters based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811571439.4A CN109682309B (en) | 2018-12-21 | 2018-12-21 | Intelligent measuring device and method for black crystal panel parameters based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109682309A true CN109682309A (en) | 2019-04-26 |
| CN109682309B CN109682309B (en) | 2023-11-14 |
Family
ID=66188189
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811571439.4A Active CN109682309B (en) | 2018-12-21 | 2018-12-21 | Intelligent measuring device and method for black crystal panel parameters based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109682309B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112468719A (en) * | 2020-11-16 | 2021-03-09 | 中国农业大学 | Processed product image acquisition device |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0798214A (en) * | 1993-09-29 | 1995-04-11 | Nippondenso Co Ltd | Method and device for three dimensional position and attitude recognition method based on sense of sight |
| FR2921478A1 (en) * | 2007-09-24 | 2009-03-27 | 3D Ouest Sarl | SYSTEM AND METHOD FOR ACQUIRING THREE-DIMENSIONAL CHARACTERISTICS OF AN OBJECT FROM IMAGES TAKEN BY A PLURALITY OF MEASURING ORGANS |
| US20100245592A1 (en) * | 2009-03-31 | 2010-09-30 | Aisin Seiki Kabushiki Kaisha | Calibrating apparatus for on-board camera of vehicle |
| CN104458750A (en) * | 2013-09-25 | 2015-03-25 | 中国科学院沈阳自动化研究所 | Automatic aluminum profile surface defect detecting equipment based on machine vision |
| JP2017098859A (en) * | 2015-11-27 | 2017-06-01 | 株式会社明電舎 | Calibration device of image and calibration method |
| CN107665489A (en) * | 2017-09-18 | 2018-02-06 | 华中科技大学 | A kind of glass dihedral angle detection method based on computer vision |
| CN207573488U (en) * | 2017-12-26 | 2018-07-03 | 杭州友上智能技术有限公司 | A kind of industrial intelligent camera based on FPGA |
| CN108917594A (en) * | 2018-05-29 | 2018-11-30 | 广东理工学院 | A kind of machine vision device measuring household board size |
| CN209416276U (en) * | 2018-12-21 | 2019-09-20 | 北京安视中电科技有限公司 | Black crystal panel mechanical parameter intelligent device for measuring based on machine vision |
-
2018
- 2018-12-21 CN CN201811571439.4A patent/CN109682309B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH0798214A (en) * | 1993-09-29 | 1995-04-11 | Nippondenso Co Ltd | Method and device for three dimensional position and attitude recognition method based on sense of sight |
| FR2921478A1 (en) * | 2007-09-24 | 2009-03-27 | 3D Ouest Sarl | SYSTEM AND METHOD FOR ACQUIRING THREE-DIMENSIONAL CHARACTERISTICS OF AN OBJECT FROM IMAGES TAKEN BY A PLURALITY OF MEASURING ORGANS |
| US20100245592A1 (en) * | 2009-03-31 | 2010-09-30 | Aisin Seiki Kabushiki Kaisha | Calibrating apparatus for on-board camera of vehicle |
| CN104458750A (en) * | 2013-09-25 | 2015-03-25 | 中国科学院沈阳自动化研究所 | Automatic aluminum profile surface defect detecting equipment based on machine vision |
| JP2017098859A (en) * | 2015-11-27 | 2017-06-01 | 株式会社明電舎 | Calibration device of image and calibration method |
| CN107665489A (en) * | 2017-09-18 | 2018-02-06 | 华中科技大学 | A kind of glass dihedral angle detection method based on computer vision |
| CN207573488U (en) * | 2017-12-26 | 2018-07-03 | 杭州友上智能技术有限公司 | A kind of industrial intelligent camera based on FPGA |
| CN108917594A (en) * | 2018-05-29 | 2018-11-30 | 广东理工学院 | A kind of machine vision device measuring household board size |
| CN209416276U (en) * | 2018-12-21 | 2019-09-20 | 北京安视中电科技有限公司 | Black crystal panel mechanical parameter intelligent device for measuring based on machine vision |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112468719A (en) * | 2020-11-16 | 2021-03-09 | 中国农业大学 | Processed product image acquisition device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109682309B (en) | 2023-11-14 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2016138744A1 (en) | Display panel surface flatness detection device and method | |
| CN106097367B (en) | A kind of scaling method and device of binocular solid camera | |
| CN107454371B (en) | Method, device and system for adapting projection area | |
| CN103149789A (en) | Image modulation transfer function (MTF) evaluation-based module motor curve testing method | |
| CN107869954B (en) | Binocular vision volume weight measurement system and implementation method thereof | |
| CN109040728B (en) | Ultra-short-focus projection equipment with double-camera trapezoidal correction and method thereof | |
| CN105835045A (en) | Panel mounting truss mechanical arm and mounting method | |
| CN106612422B (en) | A kind of projection correction's method and device | |
| CN211504117U (en) | Telecentric optical system for tilt measurement | |
| CN103837097B (en) | A kind of workpiece angle self-operated measuring unit based on image procossing and measuring method | |
| CN206258081U (en) | A kind of 3-dimensional image measurement apparatus with real-time height calibrating function | |
| CN108377371A (en) | A kind of method and device of projection image correction | |
| CN104931421B (en) | The positioning focusing method of liquid crystal panel vision detection system | |
| CN103217108B (en) | A kind of spectacle frame geometric parameter detection method | |
| CN105783723A (en) | Machine vision-based precise die surface processing precision detection device and method | |
| CN204439022U (en) | A kind of capture mechanism and apply the automatic measuring equipment of this capture mechanism | |
| CN110398849B (en) | Optical detection system for liquid crystal display screen | |
| CN104713486B (en) | One kind is puted the palms together before one product width on-line measuring device and method | |
| CN107449586A (en) | A kind of fish-eye method of testing and test device | |
| CN109343578A (en) | A kind of TDI-CCD camera binocular vision system visual field alignment schemes of view-based access control model feedback | |
| CN109191527A (en) | A kind of alignment method and device based on minimum range deviation | |
| CN109682309A (en) | Black crystal panel mechanical parameter intelligent device for measuring and method based on machine vision | |
| CN201425471Y (en) | Terminal height difference detector | |
| CN115002429A (en) | Projector capable of automatically calibrating projection position based on camera calculation | |
| CN112179292B (en) | Projector-based line structured light vision sensor calibration 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 |