CN106442543B - A kind of detection method of online recognition metal works continuous laser face of weld quality - Google Patents
A kind of detection method of online recognition metal works continuous laser face of weld quality Download PDFInfo
- Publication number
- CN106442543B CN106442543B CN201610900305.7A CN201610900305A CN106442543B CN 106442543 B CN106442543 B CN 106442543B CN 201610900305 A CN201610900305 A CN 201610900305A CN 106442543 B CN106442543 B CN 106442543B
- Authority
- CN
- China
- Prior art keywords
- image
- weld
- value
- grid
- checked
- 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.)
- Expired - Fee Related
Links
- 239000002184 metal Substances 0.000 title claims abstract description 36
- 238000001514 detection method Methods 0.000 title claims abstract description 32
- 238000000034 method Methods 0.000 claims abstract description 50
- 238000003466 welding Methods 0.000 claims abstract description 44
- 229910000679 solder Inorganic materials 0.000 claims abstract description 22
- 230000011218 segmentation Effects 0.000 claims abstract description 8
- 230000008569 process Effects 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 9
- 238000012545 processing Methods 0.000 claims description 9
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 241000208340 Araliaceae Species 0.000 claims 2
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims 2
- 235000003140 Panax quinquefolius Nutrition 0.000 claims 2
- 235000008434 ginseng Nutrition 0.000 claims 2
- 230000007547 defect Effects 0.000 abstract description 23
- 230000004069 differentiation Effects 0.000 abstract description 5
- CWYNVVGOOAEACU-UHFFFAOYSA-N Fe2+ Chemical compound [Fe+2] CWYNVVGOOAEACU-UHFFFAOYSA-N 0.000 abstract description 4
- 238000005286 illumination Methods 0.000 abstract description 3
- 239000007769 metal material Substances 0.000 abstract description 2
- 230000004044 response Effects 0.000 abstract description 2
- 230000006641 stabilisation Effects 0.000 abstract 1
- 238000011105 stabilization Methods 0.000 abstract 1
- 230000006870 function Effects 0.000 description 7
- 238000010586 diagram Methods 0.000 description 5
- 239000002956 ash Substances 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 4
- 238000004519 manufacturing process Methods 0.000 description 4
- 229910001220 stainless steel Inorganic materials 0.000 description 4
- 238000000605 extraction Methods 0.000 description 3
- 239000010935 stainless steel Substances 0.000 description 3
- 238000003384 imaging method Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 238000012372 quality testing Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 241000238097 Callinectes sapidus Species 0.000 description 1
- 235000002918 Fraxinus excelsior Nutrition 0.000 description 1
- 230000002159 abnormal effect Effects 0.000 description 1
- 238000009825 accumulation Methods 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000006399 behavior Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010894 electron beam technology Methods 0.000 description 1
- 239000003344 environmental pollutant Substances 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 239000003500 flue dust Substances 0.000 description 1
- 238000009472 formulation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000035772 mutation Effects 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 230000000149 penetrating effect Effects 0.000 description 1
- 230000000649 photocoagulation Effects 0.000 description 1
- 231100000719 pollutant Toxicity 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 230000005855 radiation Effects 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 230000001105 regulatory effect Effects 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
- 238000011179 visual inspection Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/8851—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges
- G01N2021/8887—Scan or image signal processing specially adapted therefor, e.g. for scan signal adjustment, for detecting different kinds of defects, for compensating for structures, markings, edges based on image processing techniques
Abstract
The present invention provides a kind of detection methods of online recognition metal works continuous laser face of weld quality: using high-resolution color area array cameras, high magnification micro-lens and double LED illumination Systems obtain face of weld image, then carry out welded seam area segmentation on one side and characteristic value calculates, pass through workpiece translational motion or rotation on one side, shoot the image of weld seam other positions, it constantly shoots weld image and completes to calculate in real time, until receiving solder joint or overlapped points to weld seam, finally the calculated result of each width image is integrated, it completes to continuous weld size, the online differentiation of position and surface defect.Method of the invention can complete the quality on-line checking of the long straight welds of laser of all kinds of ferrous metal and non-ferrous metal, curved welding seam or space weld, belong to the scope of non-contact vision-based detection, device is simple, compact, fast response time, strong antijamming capability, have the characteristics that stabilization, reliable, convenient, the laser welding field of metal material can be widely used in.
Description
Technical field
The present invention relates to technical field of laser welding, in particular to a kind of online recognition metal works continuous laser weld seam table
The detection method of face quality.
Background technique
Laser welding technology has been widely used for the connection of metal parts, and the thin plate of 3mm is especially less than to thickness, swashs
Photocoagulation has many advantages, such as that small deflection, high degree of automation, welding heat affected zone are narrow.But in high-volume welding production,
The process stability of laser welding control and the online quality-monitoring of laser welded seam be always determine product quality it is crucial because
Element.The factors such as beam quality, fit-up gap, workpieces surface condition, which fluctuate, can all be partially welded welding process generation interference, generation, hole
The weld defects such as hole, burn-through.Since laser beam energy density is high, naked eyes are invisible, there is potential safety to operator
It threatens, therefore laser beam welding is usually totally enclosed.In short, laser welding feature and supermatic production model are all
Need intelligentized on-line monitoring method.
Retrieval discovery, the representative achievement in the field includes: at present
1) patent " real-time detection method of butt weld of thin plates of view-based access control model " (application number: 200910083216.8), institute
It states method and calculates gray scale catastrophe point on the basis of image line scanning and column scan, so that it is determined that the position of weld seam in the picture,
There is no realize face of weld defect to be detected on differentiations for this method, and institute is spotweld to example, in patent specification not
Have and the processing and analysis of continuous weld image are illustrated.
2) patent " weld visual shape and detection method of surface flaw based on line laser structured light " (application number:
201510074062.1), the face of weld defect inspection method obtains the height of face of weld using line scan sensor
Data differentiate surface defect by weld profile fitting data and the difference being sized.This method relies only on weld seam height
Information is spent, the weld defect type that can be identified is less, and face of weld is usually rougher, surface scum, splashing, flue dust etc.
Disturbance seriously affects the accuracy of elevation carrection.In addition, the mode of line scanning is more time-consuming, deposited in continuous long weld seam detection field
Limiting to.
3) paper " unequal thickness plate laser weld defect sturcture light vision-based detection " (" laser technology " the 4th phase of volume 35,
2011) using the elevation information of structure light Active visual inspection method calculating face of weld, but this method can only obtain weld seam
Concavity or convexity state, and face of weld defect and position while welding can not be detected.
4) paper " the face of weld defect detecting system based on line laser structured light " (" welding " the 2nd phase in 2016) uses line
Laser scanning face of weld judges face of weld defect using the method for single contour fitting and more profile combinations, but this
Method can only the biggish arc weld of useable width, the narrow fine soldered seam relatively fine for weld width and flaw size (as swash
Flush weld seam, electron beam weld etc.) for and be not suitable for, incident laser at small size mutation due to light beam reflection, reflect and spread out
The problems such as penetrating easily leads to measurement failure, to be unable to complete the detection of surface state.
It follows that line scanning mode is mainly used to the identification of face of weld defect at present, the height based on face of weld
Degree exists in narrow fine soldered seam detection field and limits to according to condition discrimination is carried out, and line laser structured light mode is in detection speed side
Face is also difficult to meet actual production demand;On the other hand, the detection method based on weld seam two dimensional image is still in weld seam extraction algorithm
Optimization aspect, there is no the report for being directed to face of weld defects detection, the more systems without face of weld defect automatic detection
Method.In short, the effective, detection means with stronger adaptability is still lacked at present for face of weld quality testing, this
It is unable to satisfy the demand of extensive automatic welding production, or even causes serious security risk.
Summary of the invention
The technical problem to be solved in the present invention is to provide a kind of in the way of Surface scan, can quickly, accurately determine continuously to swash
The whether satisfactory method of flush weld seam surface quality.
In order to solve the above-mentioned technical problem, a kind of online recognition metal works continuous laser weld seam table is disclosed in the present invention
The detection method of face quality, the technical scheme is that be implemented:
A kind of detection method of online recognition metal works continuous laser face of weld quality, comprising the following steps:
S1: the laser welding workpiece of selection standard is taken pictures by camera system, obtains complete face of weld image conduct
Series standard image;
S2: it is treated by camera system and the face of weld of workpiece is examined to take pictures, obtain image to be checked;
S3: after pre-processing to the image to be checked, being carried out gridding segmentation, i.e., is divided into individual image to be checked
K × L grid, K are line number, and L is columns;
S4: calculating the accumulative gray value of each grid in step S3, obtains image to be checked and adds up gray areas distribution map, right
The change rate that each region adds up intensity profile is partitioned into background area, crystallizing field, weld metal zone;
S5: choosing series standard image corresponding with image to be checked, and the crystallizing field and weld metal zone to the two carry out similarity
It calculates, the similarity SiIt indicates, refers to the similarity of image to be checked Yu standard picture same position grid, calculate public
Formula is
In above formula, r is tonal gradation, and n indicates the tonal gradation of image, h1rIt is r's for tonal gradation in grid cell 1
Pixel number accounts for the ratio of 1 total pixel number of grid cell, h2rThe pixel number for being r for tonal gradation in grid cell 2 accounts for grid cell
The ratio of 2 total pixel numbers, abs function are ABS function, and max function is maximizing function;
In above formula, the value of n is decided by the digit A of image grayscale, i.e.,
N=2A-1
S6: established standards similarity value St, the image similarity value S to be checked that is obtained by step S5iWith StIt is compared,
To determine whether the corresponding face of weld quality of image to be checked meets the requirements.
Preferably, before executing step S6, following steps are first carried out:
S5.1: camera system, which is treated, examines workpiece continuously to be taken pictures, and obtains complete face of weld image as series
Image to be checked;
S5.2: successively executing step S2, S3, S4, S5 to series image to be checked, obtain series image to be checked crystallizing field and
The similarity value S of each grid in weld metal zonei;
In addition, in step s 6, determining the mode of face of weld quality are as follows: statistics SiValue is less than StGrid number R, and set
Fixed qualification permissible value Ra, determines the face of weld satisfactory quality of workpiece to be tested if R≤Ra;If it is not, then determining the weldering
Seam surface quality is undesirable.
Preferably, if face of weld quality is undesirable, carry out solder skip, be partially welded and weld seam hole in it is a kind of or one
Kind or more quality problems identification, comprising the following steps:
S7: crystallizing field and weld metal zone to image to be checked carry out gridding segmentation, and size of mesh opening is P × Q, and wherein P is row
Number, Q is columns;
S8: the calculation formula of the average gray value G of each grid described in setting procedure S7 is
In above formula, g 'i,jRefer to the gray value for falling in pixel in the grid;
S9: the entropy E calculation formula of each grid described in setting procedure S7 is
In above formula, x represents the tonal gradation of pixel, and n indicates the tonal gradation of image, and p (x) is the pixel of tonal gradation x
The ratio of the shared grid total pixel number of number;
S10: it carries out the identification of solder skip phenomenon: solder skip is identified by the average gray value G of grid;
S11: it carries out being partially welded phenomenon identification: finding S line by line since the first row gridiValue is less than StGrid, setting is following
Parameter:
If certain row SiValue is less than StGrid number be Rb,
Grid number is allowed to be Rc if being partially welded similarity,
Grid number is allowed to be Rd if being partially welded gray value,
If the row average gray value is Ga,
If being partially welded allows average gray value to be Gb,
If being partially welded allows gray value to be Gc,
If being partially welded allows entropy to be Ea;
If Rb >=Rc, the value of G, E, Ga, Gb, Gc and Ea are calculated according to the formula of step S8, S9;
If the grid number of Ga >=Gb or G >=Gc is at Rd or more, and the corresponding entropy E of these grids is more than or equal to Ea, then
Determine that the image to be checked exists to be partially welded;
S12: there is the phenomenon that hole to weld seam and identify: beginning look for S from weld metal zoneiThe steep drop region of value, setting with
Lower parameter:
If adjacent mesh SiThe absolute difference of value is Sa;
If absolute difference permissible value 1 is Sb;
If absolute difference operating value 2 is Sc;
If drop grid number is Re suddenly;
If drop region allows average gray value to be Gd suddenly;
If the Sa > Sc of Re grid of Sa > Sb or continuous, marking the region is steep drop region;
The calculating of average gray value is carried out after label to the region, and the curve of entropy E is calculated,
If the average gray value in the region occurs the saddle form that both sides are low, centre is high no more than the curve of Gd or entropy E,
Then determine that there are holes for weld seam corresponding position.
Preferably, before carrying out similarity value, gray value, entropy and calculating, the grey level of series image to be checked is adjusted
For uniform level, i.e., then the accumulative gray scale peak value for first unifying to limit every image to be checked calculates every image to be checked as Imax
Reality add up gray value I,
In above formula, gi,jFor the actual grey value of single pixel, i and j represent line number of the pixel in image to be checked and
Columns;
The regulation coefficient for finally obtaining this image to be checked is Z, i.e.,
Z=Imax/I
Conversion processing is carried out to image all pixels value to be checked, makes g 'i,j=gi,j/Z, subsequent all to calculate the ash that is related to
Angle value is all made of conversion treated gray value.
Preferably, to be accurately identified, following value is set:
St=0.3,
Ra=5,
Rc=3,
Rd=3,
Gb=180,
Gc=240,
Ea=5000,
Sb=0.25,
Sc=0.1,
Re=3,
Gd=100.
Preferably, the method also includes stablizing the determination method of welding: if being located at the somewhere of the crystallizing field of weld seam side
Width is more than 2mm, then determines occur unstable welding phenomena at this.
Preferably, the camera system includes that high-resolution color area array cameras, high magnification micro-lens and double LED shine
The relative position of bright device, the area array cameras and laser work head is kept constant in the welding process, and along welding direction cloth
It sets at incident laser rear, spends range 30 to 50 with the angle of incident laser axis.
Preferably, the method also includes receiving the identification of solder joint or overlapped points: blue point is extracted from weld seam color image
Spirogram picture, and morphology screening is carried out to its highlight regions, i.e. highlight regions should gradually be collapsed along welding direction, when it collapses angle
When between 10 to 30 degree, then the position is determined to receive solder joint or overlapped points position.
Preferably, double LED light devices include high power red LED light source, and it is special that the light source front end is provided with
Camera lens makes the control of beam spread angle within 15 degree, and the wave-length coverage of LED red light is 600 to 1100nm.
Preferably, the light source is placed in camera two sides, can be moved and be rotated in three-dimensional, to guarantee LED beam
It is irradiated in weld seam two sides respectively, incident angle controls between 25 ± 7 degree.
The beneficial effects of the practice of the present invention mainly has:
1, the present invention can be sentenced online for the position and surface quality of metal plate or metal tube continuous laser weld seam
Not, this method can be continuously shot weld seam using area array cameras and auxiliary LED light source in the welding process or after welding process
Area image is accurately and rapidly completed to exist to continuous laser face of weld quality by Digital Image Processing and distinguished number
Line identification;
2, party's subtraction unit is simple, and does not interfere with to laser beam welding, can be widely used in thin wall metalwork
Continuous laser weld seam quality testing field;
3, this method passes through the grey level of Universal Serial image to be checked, greatly reduces image brightness fluctuations to weld seam table
The influence of face quality judging accuracy;
4, weldering is determined by the characteristic parameter of computational gridding (similarity, average gray and entropy) and specific decision principle
Seam surface quality can effectively identify the defects of being partially welded with weld seam hole;
5, the Detection capability of different dimensional defects is had differences when using different size of mesh opening parameter (K, L), for
The higher occasion of testing requirements, thus it is possible to vary the value of K or L is computed repeatedly, and can effectively improve Detection capability.
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
A kind of embodiment 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 the positional diagram of photographic system and weld seam in one embodiment;
Fig. 2 is background area, crystallizing field and the structural schematic diagram of weld metal zone in one embodiment;
Fig. 3 is to receive the schematic diagram that angle is collapsed at solder joint;
Fig. 4 is qualified weld seam, partially weldering and surface hole defect schematic diagram in one embodiment;
Fig. 5 is the accumulation gradation data figure of acquisition in one embodiment;
Fig. 6 is the array of figure of qualified similarity, average gray and entropy in one embodiment;
Fig. 7 is to weld the array of figure of corresponding similarity, average gray and entropy partially in one embodiment;
Fig. 8 is the corresponding similarity of surface hole defect, the array of figure of average gray and entropy in one embodiment.
In above-mentioned attached drawing, each figure number label is respectively indicated:
1- camera system, 11- high-resolution color area array cameras, 12- high magnification micro-lens, the bis- LED illumination dresses of 13-
It sets,
2- workpiece, the weld metal zone 3-, 4- crystallizing field, the upper crystallizing field of 41-, crystallizing field under 42-, 5- background area, 6- collapse angle, 7-
Highlight regions, 8- imaging region, 9- surface hole defect.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.It is based on
Embodiment in the present invention, it is obtained by those of ordinary skill in the art without making creative efforts every other
Embodiment shall fall within the protection scope of the present invention.
In a specific embodiment, a kind of detection side of online recognition metal works continuous laser face of weld quality
Method, comprising the following steps:
S1: the laser welding workpiece of selection standard is taken pictures by camera system, obtains complete face of weld image conduct
Series standard image;
S2: it is treated by camera system and the face of weld of workpiece is examined to take pictures, obtain image to be checked;
S3: after pre-processing to the image to be checked, being carried out gridding segmentation, i.e., is divided into individual image to be checked
K × L grid, K are line number, and L is columns;
S4: calculating the accumulative gray value of each grid in step S3, obtains image to be checked and adds up gray areas distribution map, right
The change rate that each region adds up intensity profile is partitioned into background area, crystallizing field, weld metal zone;
S5: choosing series standard image corresponding with image to be checked, and the crystallizing field and weld metal zone to the two carry out similarity
It calculates, the similarity SiIt indicates, refers to the similarity of image to be checked Yu standard picture same position grid, calculate public
Formula is
In above formula, r is tonal gradation, and n indicates the tonal gradation of image, h1rIt is r's for tonal gradation in grid cell 1
Pixel number accounts for the ratio of 1 total pixel number of grid cell, h2rThe pixel number for being r for tonal gradation in grid cell 2 accounts for grid cell
The ratio of 2 total pixel numbers, abs function are the functions for seeking absolute value, and max function is the function of maximizing;
In above formula, the value of n is decided by the digit A of image grayscale, i.e.,
N=2A-1
S6: established standards similarity value St, the image similarity value S to be checked that is obtained by step S5iWith StIt is compared,
To determine whether the corresponding face of weld quality of image to be checked meets the requirements.
Method of the invention can complete the long straight weld of laser, curved welding seam or the sky of all kinds of ferrous metal and non-ferrous metal
Between the quality of weld seam determine that belong to the scope of non-contact vision-based detection, device is simple, compact, and fast response time is anti-interference online
Ability is strong, has the characteristics that stable, reliable, convenient, can be widely used in the laser welding field of metal material.
The camera system can use existing equipment, and those skilled in the art can also be imaged according to existing shooting
Field devices carry out any combination, as long as satisfactory, clearly face of weld image can be obtained;It is understood that
It is that the clarity of face of weld image is more clear, it is higher in the subsequent accuracy meeting for carrying out face of weld quality Identification.
The mode of complete face of weld image, general way are obtained about camera system are as follows: clamp workpiece is used, and
Workpiece is translated or is rotated, so that weld seam is completely passed through the range of taking pictures of camera system, it is complete finally to get whole weld seam
Whole image series.
Due to that may have the influence of environment or equipment in image acquisition process to be checked, such as illumination bright-dark degree and
The superiority and inferiority etc. of camera system performance, often there are noise, the disadvantages of contrast is insufficient, it is therefore desirable to image to be checked be carried out pre-
Processing, to obtain the higher image of quality for subsequent surface quality measurement, pretreated normal method is to be filtered
It makes an uproar.
Background area described in method refers to the parent metal background area of workpiece;Crystallizing field refers to laser beam welding metal
Steam falls in workpiece surface after spraying and is formed by the darker region of color, special according to the behavior of metallic vapour in welding process
Sign, crystallizing field are usually located at the 1-2mm of weld seam two sides.Background area 5, crystallizing field 4, weld metal zone 3 positional relationship can refer to Fig. 2 into
Row understands.
The image grayscale grade n being related in above-mentioned calculating process, is determined by the digit A of image grayscale, such as 8 ashes
The image of degree, n value are 255.
Similarity value SiRange between 0 to 1, value it is bigger indicate two images same position grid similarity degree
Higher, corresponding grids all to two images carry out similarity calculation, available SiValue array;And standard similarity value St
Value be decided by the height for workpiece weldquality requirement to be tested.
In a preferred embodiment, before executing step S6, following steps are first carried out:
S5.1: camera system, which is treated, examines workpiece continuously to be taken pictures, and obtains complete face of weld image as series
Image to be checked;
S5.2: successively executing step S2, S3, S4, S5 to series image to be checked, obtain series image to be checked crystallizing field and
The similarity value S of each grid in weld metal zonei;
In addition, in step s 6, determining the mode of face of weld quality are as follows: statistics SiValue is less than StGrid number R, and set
Fixed qualification permissible value Ra, determines the face of weld satisfactory quality of workpiece to be tested if R≤Ra;If it is not, then determining the weldering
Seam surface quality is undesirable.The value of Ra can be 0 herein, when it is 0, it is meant that the S of each gridiValue is both needed to small
In StWhen, which just reaches requirement.
In a preferred embodiment, it if face of weld quality is undesirable, carries out solder skip, be partially welded and weld seam hole
One or more kinds of quality problems identification in hole, comprising the following steps:
S7: crystallizing field and weld metal zone to image to be checked carry out gridding segmentation, and size of mesh opening is P × Q, and wherein P is row
Number, Q is columns;
S8: the calculation formula of the average gray value G of each grid described in setting procedure S7 is
In above formula, g 'i,jRefer to the gray value for falling in pixel in the grid;The gray value can be true gray value,
It can refer to the gray value after being adjusted by specific formulation.
S9: the entropy E calculation formula of each grid described in setting procedure S7 is
In above formula, x represents the tonal gradation of pixel, and n indicates the tonal gradation of image, and p (x) is the pixel of tonal gradation x
The ratio of the shared grid total pixel number of number;
S10: it carries out the identification of solder skip phenomenon: identifying that (when solder skip, imaging region 8 does not have solder skip by the average gray value G of grid
There is weld seam, the corresponding region of image to be checked is in highlighted state, only can be carried out determining from G value);
S11: it carries out being partially welded phenomenon identification: finding S line by line since the first row gridiValue is less than StGrid, setting is following
Parameter:
If certain row SiValue is less than StGrid number be Rb,
Grid number is allowed to be Rc if being partially welded similarity,
Grid number is allowed to be Rd if being partially welded gray value,
If the row average gray value is Ga,
If being partially welded allows average gray value to be Gb,
If being partially welded allows gray value to be Gc,
If being partially welded allows entropy to be Ea;
If Rb >=Rc, the value of G, E, Ga, Gb, Gc and Ea are calculated according to the formula of step S8, S9;
If the grid number of Ga >=Gb or G >=Gc is at Rd or more, and the corresponding entropy E of these grids is more than or equal to Ea, then
Determine that the image to be checked exists to be partially welded;
S12: there is the phenomenon that hole to weld seam and identify: beginning look for S from weld metal zoneiThe steep drop region of value, setting with
Lower parameter:
If adjacent mesh SiThe absolute difference of value is Sa;
If absolute difference permissible value 1 is Sb;
If absolute difference operating value 2 is Sc;
If drop grid number is Re suddenly;
If drop region allows average gray value to be Gd suddenly;
If the Sa > Sc of Re grid of Sa > Sb or continuous, marking the region is steep drop region;
The calculating of average gray value is carried out after label to the region, and the curve of entropy E is calculated,
If the average gray value in the region occurs the saddle form that both sides are low, centre is high no more than the curve of Gd or entropy E,
Then determine that there are holes for weld seam corresponding position.
Further, before carrying out similarity value, gray value, entropy and calculating, by grey level's tune of series image to be checked
Section is uniform level, i.e., then the accumulative gray scale peak value for first unifying to limit every image to be checked calculates every figure to be checked as Imax
The reality of picture adds up gray value I,
In above formula, gi,jFor the actual grey value of single pixel, i and j represent line number of the pixel in image to be checked and
Columns;
The regulation coefficient for finally obtaining this image to be checked is Z, i.e.,
Z=Imax/I
Conversion processing is carried out to image all pixels value to be checked, makes g 'i,j=gi,j/Z, subsequent all to calculate the ash that is related to
Angle value is all made of conversion treated gray value.
Unitized processing is carried out to series image to be checked by adjusting coefficient Z, makes the accumulative ash of the highest of single width image to be checked
Degree is kept constant, to substantially reduce the influence of accuracy of the image brightness fluctuations to quality problem identification.In some embodiments
In, the peak I max that image can be added up to gray scale is uniformly set as 1000, thus can calculate regulated quantity Z, is used for different works
The grey level of part image to be checked is adjusted to uniform level, lays the foundation for subsequent calculating.
In a preferred embodiment, to carry out accurately identifying quality problems, following value is set:
St=0.3,
Ra=5,
Rc=3,
Rd=3,
Gb=180,
Gc=240,
Ea=5000,
Sb=0.25,
Sc=0.1,
Re=3,
Gd=100.
The above value is preferable value, does not limit application mode of the invention, other people are stated using the present invention
Method on the basis of, even if the value to above-mentioned parameter is all modified, will also fall into protection scope of the present invention.
In a preferred embodiment, the method also includes stablizing the determination method of welding: if being located at weld seam side
The somewhere width of crystallizing field is more than 2mm, then determines occur unstable welding phenomena at this.
As shown in Figure 1, in a preferred embodiment, the camera system 1 include high-resolution color area array cameras 11,
It is welding the relative position of high magnification micro-lens 12 and double LED light devices 13, the area array cameras and laser work head
It keeps constant in the process, and is arranged in incident laser rear along welding direction, the angle with incident laser axis is in 30 to 50 degree
Range.This method may be implemented to obtain series image to be checked on one side, carries out data processing on one side, substantially increases face of weld matter
Measure the efficiency determined.Further, shooting speed of the area array cameras when picture size is 300 pixels × 300 pixel should be equal to
Or it is greater than 50 frames/second.
In a preferred embodiment, the method also includes receiving the identification of solder joint or overlapped points: from weld seam color image
Middle extraction blue component image, and morphology screening is carried out to its highlight regions, i.e. highlight regions should gradually be received along welding direction
Hold together, when it collapses angle between 10 to 30 degree, then determines the position to receive solder joint or overlapped points position.Fig. 3 is to carry out blue
Component extraction treated schematic diagram, weld metal zone is between upper crystallizing field 41 and lower crystallizing field 42, and highlight regions 7 are along welding
Direction is gradually collapsed, and collapses angle 6 in 10 to 30 degree ranges, which is to receive solder joint or overlapped points position.
The cromogram of rgb format includes red, three layers of green and blue, and each component is all the range of 0-255;Wherein for
Identification/judgement of face of weld quality can use the red component of color image, and adopt for receiving the differentiation of solder joint/overlapped points
Use blue component.Gathering angle 6 exists with laser power and speed of welding to be associated with, generally between 10 to 30 degree, art technology
Personnel can refer to Fig. 3 and understand.
In a preferred embodiment, double LED light devices 13 include high power red LED light source, the light source
Front end is provided with special lens, makes the control of beam spread angle within 15 degree, the wave-length coverage of LED red light for 600 to
1100nm。
Further, the light source is placed in camera two sides, can be moved and be rotated in three-dimensional, to guarantee LED light
Beam is irradiated in weld seam two sides respectively, and incident angle controls between 25 ± 7 degree.With low angle radiation modality, weld seam can be increased
The border contrast in area 3 and crystallizing field 4 can get the higher image of quality.
It is possible to further carry out operation below: in the processing and judgement for completing all weld images, and receiving solder joint
After differentiation, the surface quality of whole weld seam is determined, provide weld width, welding position, surface crater judgement as a result,
And a record is formed, it is stored in database together with fault location image.It will be appreciated by persons skilled in the art that institute of the present invention
Method is stated, face of weld quality judging/identification result can be exported by modes such as text, charts, or
The various ways such as display, prompt tone are marked by picture and carry out prompt defective locations, do not make to be unfolded herein.
One embodiment is disclosed again below:
Use power for 0.6kW, focus diameter is the laser of 0.2mm, and the workpiece material being examined is 304 stainless steels,
With a thickness of 0.8mm.Examined workpiece is stainless steel tube, having a size of 70mm (length) × 8mm (diameter).In welding process, enter
It penetrates laser to keep constant, fixture drives stainless steel tube rotation, and workpiece rotation process is swung, or since chucking power deficiency causes
When revolving speed unevenness or welding region there are when the pollutants such as water oil, it may appear that be partially welded, be imperfect, weld width is uneven, surface
The defects of pit or protrusion, it should identify these situations and be intervened.
In this example, detected object is stainless steel tube girth joint, and examined workpiece at the uniform velocity rotates in detection process, face battle array phase
Machine obtains 31 width images altogether, wherein last 1 width image and the 1st width picture registration, that is, have 30 width effective images;Area array cameras shooting
The weld image in specified region, having a size of 300 pixels × 260 pixels, shooting speed is equal to or more than 50 frames/second.
In this example, the weld image of shooting such as Fig. 4, including qualified weld seam (weld seam of the laser welding workpiece of standard), position
It sets and deviates weld seam (being partially welded) and surface hole defect weld seam (surface hole defect 9 is schematic forms).To the mesh segmentation of single width weld image
Parameter are as follows: line number (K)=16, columns (L)=30.
In this example, separate threads development is respectively adopted in image taking and image procossing differentiation;While workpiece starts turning
Triggering shooting, captured picture are stored in specified memory, until shooting stops when workpiece stops rotating;It detects program and finds image
Start to calculate after having image in memory, the processing time of single image is 20-35ms, and the Data Integration time is 2-5ms, all numbers
After having handled and having saved required picture and data, output test result simultaneously empties specified memory, is the bat of next workpiece
It takes the photograph and detection is ready.
In this example, add up grayscale image as shown in figure 5, the entropy array and similarity array of qualified weld seam are as shown in fig. 6, deviate
The entropy array and similarity array of weld seam as shown in fig. 7, the entropy array and similarity array of hole weld seam as shown in figure 8, therefrom
It receives solder joint as can be seen that this detection method can accurately reflect, be partially welded, the states such as surface hole defect, it can be according to characteristic parameter
Abnormal distribution prepares the defect of identification face of weld quality.
The above-mentioned various embodiments enumerated can be combined with each other implementation, those skilled in the art under the premise of reconcilable
In combination with attached drawing and above to the explanation of embodiment, as the foundation being combined to the technical characteristic in different embodiments.
It should be pointed out that the foregoing is merely illustrative of the preferred embodiments of the present invention, it is not intended to limit the invention, it is all
Within the spirit and principles in the present invention, any modification, equivalent replacement, improvement and so on should be included in guarantor of the invention
Within the scope of shield.
Claims (8)
1. a kind of detection method of online recognition metal works continuous laser face of weld quality, it is characterised in that:
The following steps are included:
S1: the laser welding workpiece of selection standard is taken pictures by camera system, obtains complete face of weld image as series
Standard picture;
S2: it is treated by camera system and the face of weld of workpiece is examined to take pictures, obtain image to be checked;
S3: after pre-processing to the image to be checked, being carried out gridding segmentation, i.e., individual image to be checked is divided into K × L
A grid, K are line number, and L is columns;
S4: calculating the accumulative gray value of each grid in step S3, obtains image to be checked and adds up gray areas distribution map, according to each
The change rate that region adds up intensity profile is partitioned into background area, crystallizing field, weld metal zone;
S5: choosing series standard image corresponding with image to be checked, and the crystallizing field and weld metal zone to the two carry out similarity calculation,
The similarity SiIt indicates, refers to the similarity of image to be checked Yu standard picture same position grid, its calculation formula is
In above formula, r is tonal gradation, and n indicates the tonal gradation of image, h1rThe pixel number for being r for tonal gradation in grid cell 1
Account for the ratio of 1 total pixel number of grid cell, h2rThe pixel number for being r for tonal gradation in grid cell 2 accounts for the total picture of grid cell 2
The ratio of prime number, abs function are ABS function, and max function is maximizing function;
In above formula, the value of n is decided by the digit A of image grayscale, i.e.,
N=2A-1
S6: established standards similarity value St, the image similarity value S to be checked that is obtained by step S5iWith StIt is compared, to sentence
Determine whether the corresponding face of weld quality of image to be checked meets the requirements;
Before executing step S6, following steps are first carried out:
S5.1: camera system, which is treated, examines workpiece continuously to be taken pictures, and it is to be checked as series to obtain complete face of weld image
Image;
S5.2: successively executing step S3, S4, S5 to series image to be checked, obtains crystallizing field and the weld metal zone of series image to be checked
The similarity value S of each gridi;
In addition, in step s 6, determining the mode of face of weld quality are as follows: statistics SiValue is less than StGrid number R, and set conjunction
Lattice permissible value Ra determines the face of weld satisfactory quality of workpiece to be tested if R≤Ra;If it is not, then determining the weld seam table
Face quality is undesirable;
If face of weld quality is undesirable, carry out solder skip, be partially welded and the quality problems of weld seam hole identification, including with
Lower step:
S7: crystallizing field and weld metal zone to image to be checked carry out gridding segmentation, and size of mesh opening is P × Q, and wherein P is line number, Q
For columns;
S8: the calculation formula of the average gray value G of grid described in setting procedure S7 is
In above formula, g 'i,jRefer to the gray value for falling in pixel in the grid;
S9: the entropy E calculation formula of grid described in setting procedure S7 is
In above formula, x represents the tonal gradation of pixel, and n indicates the tonal gradation of image, p (x) for tonal gradation x pixel number institute
Account for the ratio of grid total pixel number;
S10: it carries out the identification of solder skip phenomenon: solder skip is identified by the average gray value G of grid;
S11: it carries out being partially welded phenomenon identification: finding S line by line since the first row gridiValue is less than StGrid, set following ginseng
Number:
If certain row SiValue is less than StGrid number be Rb,
Grid number is allowed to be Rc if being partially welded similarity,
Grid number is allowed to be Rd if being partially welded gray value,
If the row average gray value is Ga,
If being partially welded allows average gray value to be Gb,
If being partially welded allows gray value to be Gc,
If being partially welded allows entropy to be Ea;
If Rb >=Rc, the value of G, E are calculated according to the formula of step S8, S9, and calculate the value of Ga;
If the grid number of Ga >=Gb or G >=Gc is at Rd or more, and the corresponding entropy E of these grids is more than or equal to Ea, then determines
The workpiece to be tested, which exists, to be partially welded;
S12: there is the phenomenon that hole to weld seam and identify: beginning look for S from weld metal zoneiThe steep drop region of value, sets following ginseng
Number:
If adjacent mesh SiThe absolute difference of value is Sa;
If absolute difference permissible value 1 is Sb;
If absolute difference permissible value 2 is Sc;
If drop grid number is Re suddenly;
If drop region allows average gray value to be Gd suddenly;
If the Sa > Sc of Re grid of Sa > Sb or continuous, marking the region is steep drop region;
The calculating of average gray value is carried out after label to the region, and the curve of entropy E is calculated,
If the average gray value in the region occurs the saddle form that both sides are low, centre is high no more than the curve of Gd or entropy E, sentence
Determining weld seam corresponding position, there are holes.
2. detection method according to claim 1, it is characterised in that: calculated carrying out similarity value, gray value, entropy
Before, the grey level of series image to be checked is adjusted to uniform level, i.e., first unifies the accumulative gray scale for limiting every image to be checked
Peak value is Imax, and the reality for then calculating every image to be checked adds up gray value I,
In above formula, gi,jFor the actual grey value of single pixel, i and j represent line number and column of the pixel in image to be checked
Number;
The regulation coefficient for finally obtaining this image to be checked is Z, i.e.,
Z=Imax/I
Conversion processing is carried out to image all pixels value to be checked, makes g 'i,j=gi,j/ Z, the gray value that subsequent all calculating are related to
It is all made of conversion treated gray value.
3. detection method according to claim 1, it is characterised in that:
To be accurately identified, following value is set:
St=0.3,
Ra=5,
Rc=3,
Rd=3,
Gb=180,
Gc=240,
Ea=5000,
Sb=0.25,
Sc=0.1,
Re=3,
Gd=100.
4. detection method according to claim 1, it is characterised in that:
The method also includes stablizing the determination method of welding: if the somewhere width for being located at the crystallizing field of weld seam side is more than 2mm,
Then determine occur unstable welding phenomena at this.
5. detection method according to claim 1, it is characterised in that:
The camera system includes high-resolution color area array cameras, high magnification micro-lens and double LED light devices, described
The relative position of area array cameras and laser work head is kept constant in the welding process, and is arranged in incident laser along welding direction
Range is spent 30 to 50 with the angle of incident laser axis in rear.
6. detection method according to claim 1, it is characterised in that:
The method also includes receiving the identification of solder joint or overlapped points: blue component image is extracted from weld seam color image, and right
Its highlight regions carries out morphology screening, i.e. highlight regions should gradually be collapsed along welding direction, when it collapses angle in 10 to 30 degree
Between when, then determine the position to receive solder joint or overlapped points position.
7. detection method according to claim 5, it is characterised in that:
Double LED light devices include high power red LED light source, and the light source front end is provided with special lens, make light beam
Angle of flare controls within 15 degree, and the wave-length coverage of LED red light is 600 to 1100nm.
8. detection method according to claim 7, it is characterised in that:
The light source is placed in camera two sides, can be moved and be rotated in three-dimensional, to guarantee that LED beam is irradiated in weldering respectively
Two sides are stitched, incident angle controls between 25 ± 7 degree.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610900305.7A CN106442543B (en) | 2016-10-14 | 2016-10-14 | A kind of detection method of online recognition metal works continuous laser face of weld quality |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610900305.7A CN106442543B (en) | 2016-10-14 | 2016-10-14 | A kind of detection method of online recognition metal works continuous laser face of weld quality |
Publications (2)
Publication Number | Publication Date |
---|---|
CN106442543A CN106442543A (en) | 2017-02-22 |
CN106442543B true CN106442543B (en) | 2019-10-15 |
Family
ID=58175068
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610900305.7A Expired - Fee Related CN106442543B (en) | 2016-10-14 | 2016-10-14 | A kind of detection method of online recognition metal works continuous laser face of weld quality |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106442543B (en) |
Families Citing this family (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108645865A (en) * | 2018-05-10 | 2018-10-12 | 中国石油天然气集团有限公司 | A kind of measurement method of the submerged-arc welding steel pipe weld seam amount of the being partially welded parameter based on CCD |
CN109239081B (en) * | 2018-09-18 | 2021-01-15 | 广东省特种设备检测研究院珠海检测院 | Weld quality parameter detection method based on structured light and visual imaging |
DE102018129425B4 (en) * | 2018-11-22 | 2020-07-30 | Precitec Gmbh & Co. Kg | System for recognizing a machining error for a laser machining system for machining a workpiece, laser machining system for machining a workpiece by means of a laser beam comprising the same, and method for detecting a machining error in a laser machining system for machining a workpiece |
CN110145318A (en) * | 2019-05-17 | 2019-08-20 | 中国铁建重工集团股份有限公司 | Shield body deformation detection and the method corrected online on a kind of shield machine |
CN110608684B (en) * | 2019-08-12 | 2022-02-11 | 广东工业大学 | Single-layer multi-channel weld accumulation deposition effect detection method and system |
CN110658160B (en) * | 2019-10-09 | 2022-06-07 | 湖南工学院 | Aluminum oxide film uniformity degree detection device based on line laser scanning |
CN110987944B (en) * | 2019-12-19 | 2022-07-01 | 武汉新耐视智能科技有限责任公司 | Laser brazing weld surface defect detection method based on envelope line identification |
CN110992362A (en) * | 2019-12-25 | 2020-04-10 | 宁波江丰电子材料股份有限公司 | Method for detecting air holes in welding type sputtering target |
CN111665247A (en) * | 2020-07-08 | 2020-09-15 | 浙江力驰雷奥环保科技有限公司 | Welding seam detection method for plate type fin and detection tool adopting method |
CN112091480B (en) * | 2020-08-06 | 2022-02-11 | 苏州实创德光电科技有限公司 | Auxiliary internal lighting system for identifying axial weld joint of straight pipe and welding method |
CN112317335A (en) * | 2020-09-15 | 2021-02-05 | 苏州通富超威半导体有限公司 | Tray anti-mixing system and method |
CN112132807B (en) * | 2020-09-23 | 2024-02-23 | 泉州装备制造研究所 | Weld joint region extraction method and device based on color similarity segmentation |
CN112730428A (en) * | 2020-12-23 | 2021-04-30 | 联合汽车电子有限公司 | Extension rod weld joint detection system and method |
CN113450323B (en) * | 2021-06-22 | 2022-12-06 | 深圳盈天下视觉科技有限公司 | Quality detection method and device, electronic equipment and computer readable storage medium |
CN113916884A (en) * | 2021-10-09 | 2022-01-11 | 宿州学院 | Hard turning workpiece surface quality online prediction method |
CN115144399B (en) * | 2022-08-30 | 2023-01-13 | 吉林省信息技术研究所 | Assembly quality detection method and device based on machine vision |
CN115980092B (en) * | 2023-03-20 | 2023-06-23 | 宁波吉宁汽车零部件有限公司 | Welding part detection equipment |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1058170A (en) * | 1996-08-19 | 1998-03-03 | Mazda Motor Corp | Method and device for judging quality of laser beam welding |
CN101730607A (en) * | 2007-05-26 | 2010-06-09 | 通快机床两合公司 | Method for detecting faults at a weld seam during a laser-welding process |
CN104512579A (en) * | 2014-10-22 | 2015-04-15 | 中国电子科技集团公司第四十一研究所 | Defect detecting method for bracing wire on transparent packing material |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN105938563A (en) * | 2016-04-14 | 2016-09-14 | 北京工业大学 | Weld surface defect identification method based on image texture |
-
2016
- 2016-10-14 CN CN201610900305.7A patent/CN106442543B/en not_active Expired - Fee Related
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JPH1058170A (en) * | 1996-08-19 | 1998-03-03 | Mazda Motor Corp | Method and device for judging quality of laser beam welding |
CN101730607A (en) * | 2007-05-26 | 2010-06-09 | 通快机床两合公司 | Method for detecting faults at a weld seam during a laser-welding process |
CN104512579A (en) * | 2014-10-22 | 2015-04-15 | 中国电子科技集团公司第四十一研究所 | Defect detecting method for bracing wire on transparent packing material |
Also Published As
Publication number | Publication date |
---|---|
CN106442543A (en) | 2017-02-22 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106442543B (en) | A kind of detection method of online recognition metal works continuous laser face of weld quality | |
JP3123146B2 (en) | Weld bead quality inspection equipment | |
US7570794B2 (en) | System and method for evaluating a machined surface of a cast metal component | |
CN109001224B (en) | Welding seam detection method and detection device | |
CN104636760A (en) | Positioning method for welding seam | |
CN102063718A (en) | Field calibration and precision measurement method for spot laser measuring system | |
CN105006261B (en) | Nuclear fuel assembly video detecting method and detection means | |
CN102175700A (en) | Method for detecting welding seam segmentation and defects of digital X-ray images | |
CN109559304A (en) | Image quality online evaluation method, apparatus and application for industrial vision detection | |
JP5158406B2 (en) | Welding situation analysis apparatus and method | |
CN108247179A (en) | I shape grooves CO based on crater image2Weld deviation detecting method and device | |
CN102735611A (en) | Method for detecting resistance welding strength of lithium ion battery, and apparatus thereof | |
CN115631138A (en) | Zirconium alloy plate laser cutting quality monitoring method and device | |
CN115100127A (en) | Circular weld defect detection method based on visual identification | |
CN108596173A (en) | One camera full view wire size real-time distinguishing apparatus and its detection method | |
CN116309409A (en) | Weld defect detection method, system and storage medium | |
CN115601359A (en) | Welding seam detection method and device | |
CN113743311B (en) | Device and method for detecting welding spots of battery and connecting sheet based on machine vision | |
CN111397529A (en) | Complex surface shape detection method based on binocular vision structured light | |
TW419634B (en) | Automatic detection system and method using bar code positioning | |
JP4981433B2 (en) | Inspection device, inspection method, inspection program, and inspection system | |
TW201812705A (en) | Optical measurement of step size and plated metal thickness | |
CN112858341B (en) | Detection method, shooting system and detection system | |
JP2000009880A (en) | Device and method for inspecting fuel assembly | |
US20180089824A1 (en) | Method of calibrating a direct laser deposition process |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | 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 | ||
CF01 | Termination of patent right due to non-payment of annual fee | ||
CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20191015 |