CN109187581A - The bearing finished products plate defects detection method of view-based access control model - Google Patents
The bearing finished products plate defects detection method of view-based access control model Download PDFInfo
- Publication number
- CN109187581A CN109187581A CN201810764930.2A CN201810764930A CN109187581A CN 109187581 A CN109187581 A CN 109187581A CN 201810764930 A CN201810764930 A CN 201810764930A CN 109187581 A CN109187581 A CN 109187581A
- Authority
- CN
- China
- Prior art keywords
- bearing
- image
- obtains
- detected
- region
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000007547 defect Effects 0.000 title claims abstract description 57
- 238000001514 detection method Methods 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 claims abstract description 44
- 230000011218 segmentation Effects 0.000 claims abstract description 14
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 12
- 230000003044 adaptive effect Effects 0.000 claims abstract description 9
- 238000004458 analytical method Methods 0.000 claims abstract description 8
- 230000004807 localization Effects 0.000 claims abstract description 8
- 230000002950 deficient Effects 0.000 claims abstract description 5
- 238000003709 image segmentation Methods 0.000 claims description 13
- 238000005070 sampling Methods 0.000 claims description 10
- 238000006243 chemical reaction Methods 0.000 claims description 6
- 238000007789 sealing Methods 0.000 claims description 5
- 238000000605 extraction Methods 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000000877 morphologic effect Effects 0.000 claims description 3
- 238000012216 screening Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 abstract description 6
- 230000007797 corrosion Effects 0.000 abstract description 6
- 238000005303 weighing Methods 0.000 abstract description 2
- 238000004519 manufacturing process Methods 0.000 description 13
- 238000010586 diagram Methods 0.000 description 11
- 238000005516 engineering process Methods 0.000 description 7
- 230000000694 effects Effects 0.000 description 5
- 230000008859 change Effects 0.000 description 4
- 230000004438 eyesight Effects 0.000 description 4
- 230000008569 process Effects 0.000 description 4
- 230000006870 function Effects 0.000 description 3
- 239000004744 fabric Substances 0.000 description 2
- 239000000203 mixture Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000035807 sensation Effects 0.000 description 2
- 208000037656 Respiratory Sounds Diseases 0.000 description 1
- 206010039509 Scab Diseases 0.000 description 1
- 229910000831 Steel Inorganic materials 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 230000004313 glare Effects 0.000 description 1
- 239000011521 glass Substances 0.000 description 1
- 239000003292 glue Substances 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000003475 lamination Methods 0.000 description 1
- 239000004973 liquid crystal related substance Substances 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 238000003909 pattern recognition Methods 0.000 description 1
- 238000005086 pumping Methods 0.000 description 1
- 238000013441 quality evaluation Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000006748 scratching Methods 0.000 description 1
- 230000002393 scratching effect Effects 0.000 description 1
- 239000010959 steel Substances 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 230000026676 system process Effects 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
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
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/952—Inspecting the exterior surface of cylindrical bodies or wires
-
- 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
- G01N21/8851—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
Landscapes
- Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Signal Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention belongs to technical field of machine vision, more particularly to a kind of bearing finished products plate defects detection method of view-based access control model, aim to solve the problem that corrosion that may be present on bearing finished products end face, problem on line detection the defects of scratch, weighing wounded, collide with, the method for the present invention includes: to obtain acquired original image；Obtain the first workpiece image of bearing face to be detected；Edge extracting is carried out, the accurate positioning of bearing component is obtained with round fitting algorithm, obtains second workpiece image；Based on the second workpiece image, annular workpieces figure is expanded into rectangular piece figure；The rectangular piece figure is divided by multiple end region figures using the method for one direction projection localization；Binarization segmentation is carried out by adaptive gray threshold, obtains defect candidate region；To the defect candidate region, defect area is obtained using connected domain analysis method, and according to defective locations and shape information.The present invention realizes the on-line checking of bearing face, improves detection efficiency, detection stability.
Description
Technical field
The invention belongs to technical field of machine vision, and in particular to a kind of bearing finished products plate defects detection of view-based access control model
Method.
Background technique
For the bearing manufacturing in current China still based on traditional product, technical level is not high, and thus bring influence is,
Domestic Production of bearing be still it is labor-intensive based on, overall productivity is high, but it is with low content of technology, precision is low, added value of product
It is low, and quality not can guarantee.Although Production of bearing enterprise of a small number of country, some automations, semi-automatic axis are introduced from abroad
Manufacturing equipment is held, forms the production production capacity of the high-accuracy bearing products in part, but since detection means is with artificial detection, offline pumping
Based on looking into, so that product quality is unable to get guarantee, it is unable to satisfy the quality requirement of Foreign User, when there are quality problems,
Often it is faced with huge fine.Simultaneously as cannot achieve the real-time online detection of precision bearing, production line automation, intelligence
Controlled level, which can be changed, to be improved, and production efficiency can not be compared with overseas enterprise, and domestic Ji Jia domestic enterprise on the cutting edge is raw
Producing benefit can not also be compared with an external large enterprise.Online measuring technique can realize the automation and one of manufacture, detection
Change, can also realize the standardization of detection means.With the increase of the high-accuracy bearing products demand in China, the expansion of production capacity, with
And the increasing of export volume, especially bearing products field international standard are carried out and implemented, so that the online inspection to high-accuracy bearing
The demand of survey technology is more more and more urgent.
In the manufacture process of bearing and its accessory, in fact it could happen that burr, color difference, excessive glue, corrosion, crackle, scratch,
Scab, hole, skin lamination, point, lack pearl, chamfering is undressed, chamfering is excessive or it is too small, weigh wounded, the defects of end face is undressed,
This will reduce the performances such as the corrosion resistance, wearability and fatigue resistance of bearing.Bearing Manufacturing Enterprise is usually adopted both at home and abroad at present
The mode manually visually inspected by random samples carries out quality evaluation, lacks the consistency and science of detection, and efficiency is lower, and with letter
The intelligent detection equipment developed based on breath technology can solve problems.Along with high precision image senser element, number letter
The theory study and application study of the related fieldss such as number processor technology, computer vision, pattern-recognition become better and approaching perfection day by day with it is practical
Change, attention of the development of the bearing defect automatic checkout system of view-based access control model by more and more bearing producers.
Defects detection based on computer vision has non-contact, quick, accurate and high reliability, therefore
Defects detection on surfaces such as food grade, printed matter, glass, steel, liquid crystal displays obtains extensive research and application.
However, domestic existing vision detection system can not be applied to the on-line checking of bearing finished products, and main reason is that: 1) finished product
Bearing face model comparision is complicated, including the parts such as chamfering, inner and outer ring end face, sealing cover；2) defect present on bearing face
There are many type, including corrosion, scratch, weigh wounded, collide with；3) industry spot has higher requirement of real-time；4) bearing face
The required precision of defects detection is very high, and detection accuracy is usually required that in 0.1mm2It is even higher, it is therefore desirable to use high-resolution
Imaging sensor (more than million pixels)；This just proposes higher requirement to the efficiency of visible sensation image processing process.
Summary of the invention
In order to solve the above problem in the prior art, in order to solve on bearing finished products end face it is that may be present corrosion,
Scratch, the problem on line detection for the defects of weighing wounded, colliding with, the invention proposes a kind of bearing finished products plate defects of view-based access control model
Detection method, comprising the following steps:
Step S1 obtains the acquired original image comprising bearing face image to be detected；
Step S2 obtains the first workpiece image of bearing face to be detected based on the acquired original image；
Step S3 carries out edge extracting to the first workpiece image, and to extracted marginal point with the fitting algorithm of circle
The accurate positioning for obtaining bearing component, obtains the second workpiece image of bearing face to be detected；
Step S4 is based on the second workpiece image, the annular workpieces figure of bearing face to be detected is expanded into rectangle work
Part figure；
The rectangular piece figure is divided into multiple end region figures using the method for one direction projection localization by step S5；
Step S6 carries out binarization segmentation by adaptive gray threshold, obtains according to the different gamma characteristics in each region
Defect candidate region；
Step S7, to the defect candidate region, using connected domain analysis method, and according to defective locations and shape information
Obtain final defect area.
In the preferred embodiment of the present invention, the method used in step S2 obtains bearing end to be detected for image segmentation algorithm
First workpiece image in face, comprising the following steps:
Step S21 obtains the acquired original image using adaptive threshold fuzziness and connected component analysis method
First candidate artifacts regional ensemble；
Step S22, based on the size in each candidate artifacts region, location information pair in the first candidate artifacts regional ensemble
Candidate artifacts region is screened and is merged, and the second candidate artifacts regional ensemble is obtained；
Step S23 determines the first workpiece image based on the second candidate artifacts regional ensemble.
In the preferred embodiment of the present invention, " the first candidate artifacts regional ensemble is obtained " in step S21, method are as follows:
Step S211 carries out down-sampling to the acquired original image；
Step S212, the grey level histogram of statistical picture, and threshold required for segmented image is acquired using ISODATA algorithm
Value t2a1；
Step S213 calculates the peak value t of the grey level histogram of image2a2, and further calculate the threshold value for image segmentation
t2a3, t2a3=A t2a1-Bt2a2, wherein A, B are predetermined coefficient；
Step S214 is based on t2a3Image segmentation is carried out, the first candidate artifacts regional ensemble is obtained.
In the preferred embodiment of the present invention, " candidate artifacts region is screened and merged " in step S22, comprising: from
The candidate artifacts region that size meets given threshold is screened in the first candidate artifacts regional ensemble, and further will sieve
There is the candidate artifacts region of lap to merge in candidate artifacts region after choosing, generates the second candidate artifacts region collection
It closes.
In the preferred embodiment of the present invention, the obtaining step of second workpiece image described in step S3 includes:
Step S31 uses edge extraction techniques to the first workpiece image, obtain its outermost edge under polar coordinate system and
Innermost edge information obtains the marginal information of the inside and outside circle of bearing to be detected；
Step S32, based on the marginal information of the inside and outside circle of bearing to be detected, with least square circle approximating method obtain to
The accurate positioning of the inside and outside circle of bearing is detected, and then obtains the second workpiece image.
In the preferred embodiment of the present invention, " accurate positioning for obtaining the inside and outside circle of bearing to be detected ", side in step S32
Method are as follows:
Step S321, based on the marginal information of the inside and outside circle of bearing to be detected, the edge for treating the detection inside and outside circle of bearing is adopted
Sampling point does down-sampling；Initializing all edge sample points is interior point, and carries out Least Square Circle fitting based on this；
Step S322 judges the edge sample point of all inside and outside circles of bearing to be detected: if edge sample point and
The distance of the circle of fitting is less than C, then determines that edge sample point is interior point, be otherwise judged as exterior point；C is given threshold；
Step S323 carries out Least Square Circle fitting using the interior point set obtained after judging in step S322 again, obtains
To the accurate positioning of the inside and outside circle of bearing to be detected.
In the preferred embodiment of the present invention, " the annular workpieces figure of bearing face to be detected is expanded into rectangle in step S4
Workpiece figure ", method are as follows:
Step S41 obtains the center of bearing face to be detected based on the accurate positioning of the inside and outside circle of bearing to be detected；
Step S42 obtains coordinate conversion parameter based on the rectangular image dimension information to be stretched；
The annular workpieces figure of bearing face to be detected is expanded into rectangle work according to the coordinate conversion parameter by step S43
Part figure.
In the preferred embodiment of the present invention, the method for one direction projection localization described in step S5, comprising:
Rectangular piece figure is projected to short-axis direction, calculates mean value, obtains the gray scale point in workpiece height direction by step S51
Cloth curve；
Step S52 carries out Threshold segmentation according to gradient transformation to the intensity profile curve, obtains point in each region of workpiece
Boundary line.
In the preferred embodiment of the present invention, " defect candidate region is obtained " in step S6, method are as follows: obtain to step S5
Each described end region figure, execute following steps:
Step S61 calculates the average brightness value of end region figure；
Step S62 carries out Threshold segmentation according to average brightness value, if current grayvalue f (i, j) < mean (i, j)-T,
Current point is determined as defect point, wherein mean (i, j) is the gray average at coordinate (i, j), and T is preset constant offset
Amount；
Step S63 filters out too small defect point using morphologic filtering method, obtains defect candidate region.
In the preferred embodiment of the present invention, multiple end regions described in step S5 include: inner ring end region, outer ring end
Face region, sealing cover area and chamfered area.
Technical effect of the invention: technical solution of the present invention is with the difference manually inspected by random samples offline, by using height
Resolution ratio industrial camera and visible sensation image processing process overcome artificial offline sampling observation method low efficiency, are easy fatigue, quality
The shortcomings that not can guarantee can meet the needs that Production of bearing producer promotes automatization level.By the way that the image acquired in real time is transported
With image segmentation and target association, workpiece image of the bearing finished products end face at camera field of view center is obtained；Then, pass through work
Part segmentation, coordinate transform, region segmentation, gray scale adaptive threshold fuzziness, determining defects obtain corrosion present in workpiece, press
The location information for the defects of hurting, collide with and scratching solves the problem on line detection of bearing finished products end face.
Detailed description of the invention
Fig. 1 is that the present invention is based on the flow diagrams of the bearing finished products plate defects detection method of vision；
Fig. 2 is bearing face image segmentation schematic diagram；
Fig. 3 is coordinate transform effect diagram；
Fig. 4 is final defects detection result schematic diagram.
Specific embodiment
The preferred embodiment of the present invention described with reference to the accompanying drawings.It will be apparent to a skilled person that this
A little embodiments are used only for explaining technical principle of the invention, it is not intended that limit the scope of the invention.
The bearing finished products plate defects detection method of view-based access control model of the invention, main thought is: utilizing machine vision
Light source (RL-70-90-W) and glare shield controls ambient brightness, with high resolution industrial video camera (MV-2000UC, 200 everythings
Element) image that acquires in real time is chief source of information, it is handled using visual pattern and workpiece is positioned and analyzed, realize finished product axis
The online automatic detection of socket end planar defect.
The bearing finished products plate defects detection method of view-based access control model of the invention, comprising the following steps:
Step S1 obtains the acquired original image comprising bearing face image to be detected；
Step S2 obtains the first workpiece image of bearing face to be detected based on the acquired original image；
Step S3 carries out edge extracting to the first workpiece image, and to extracted marginal point with the fitting algorithm of circle
The accurate positioning for obtaining bearing component, obtains the second workpiece image of bearing face to be detected；
Step S4 is based on the second workpiece image, the annular workpieces figure of bearing face to be detected is expanded into rectangle work
Part figure；
The rectangular piece figure is divided into multiple end region figures using the method for one direction projection localization by step S5；
Step S6 carries out binarization segmentation by adaptive gray threshold, obtains according to the different gamma characteristics in each region
Defect candidate region；
Step S7, to the defect candidate region, using connected domain analysis method, and according to defective locations and shape information
Obtain final defect area.
In order to be more clearly illustrated to technical solution of the present invention, carried out below according to each step of the method for the present invention
Extensibility description.
Step S1 obtains the acquired original image comprising bearing face image to be detected.
In order to obtain more accurate detection effect, high resolution industrial video camera real-time image acquisition can use.
Step S2 obtains the first workpiece image of bearing face to be detected based on the acquired original image.
The image segmentation algorithm used in the step obtains the first workpiece image of bearing face to be detected.Fig. 2 is this hair
Bright bearing face image segmentation schematic diagram, wherein left figure be by the collected acquired original image of high resolution industrial camera,
Middle figure is the area schematic that Threshold segmentation obtains, and right figure is finally to divide the first obtained workpiece image.The tool of image segmentation
Body method is as follows:
Step S21 obtains the acquired original image using adaptive threshold fuzziness and connected component analysis method
First candidate artifacts regional ensemble.
Step S211 carries out down-sampling to the acquisition image to reduce calculation amount.
Step S212, the grey level histogram of statistical picture, and threshold required for segmented image is acquired using ISODATA algorithm
Value t2a1。
The step can also increase the judgement of the workpiece image with the presence or absence of bearing face to be detected: if there is workpiece figure
Picture, then t2a1Should be between background and workpiece brightness, proportion is not too big in the certain bandwidth of threshold value, if therefore t2a1In 20 bands
Pixel ratio in wide is greater than 70%, then illustrates that there is no workpiece image, direct return step S1 in current time image.Increase
The step of judgement, can terminate the execution of following steps when the workpiece image of bearing face to be detected is not present, and it is invalid to eliminate
The resource and time that step expends, improve detection efficiency.
Step S213 calculates the peak value t of the grey level histogram of image2a2, and further calculate the threshold value for image segmentation
t2a3, t2a3=A t2a1-Bt2a2, wherein A, B are predetermined coefficient.A=1.1 in the present embodiment, B=0.1.
Step S214 is based on t2a3Image segmentation is carried out, the first candidate artifacts regional ensemble is obtained.
Step S22, based on the size in each candidate artifacts region, location information pair in the first candidate artifacts regional ensemble
Candidate artifacts region is screened and is merged, and the second candidate artifacts regional ensemble is obtained.
The step method particularly includes: screen size from the first candidate artifacts regional ensemble and meet setting threshold
The candidate artifacts region of value D, and will further have in the candidate artifacts region after screening the candidate artifacts region of lap into
Row merges, and generates the second candidate artifacts regional ensemble.In the present embodiment, given threshold D is 200 pixels.
Step S23 determines the first workpiece image based on the second candidate artifacts regional ensemble.
In the step, according to the distribution situation of brightness of image, by treated the connection region step S21, step S22 by
Three concentric loop compositions, specifically include outer end face region, metal sealing cover area and inner face region, resulting according to dividing
The information such as connected region position shape, the final location information for obtaining bearing face.If the centre coordinate of three connected regions point
It Wei not (Xi,Yi), i=1,2,3, then bearing face regional center is (X, Y), whereinAxis
Socket end face zone radius size is taken as 5 pixels of increasing radius of maximum annulus, and widened pixel value can in other embodiments
To be set according to demand.
Step S3 carries out edge extracting to the first workpiece image, and to extracted marginal point with the fitting algorithm of circle
The accurate positioning for obtaining bearing component, obtains the second workpiece image of bearing face to be detected.
After bearing face approximate region range obtains, edge extracting will be used, the technologies such as circle fitting are to obtain bearing face
Accurate positioning.The specific method is as follows for effective bearing face image zooming-out:
Step S31 uses edge extraction techniques to the first workpiece image, obtain its outermost edge under polar coordinate system and
Innermost edge information obtains the marginal information of the inside and outside circle of bearing to be detected.
The step specifically includes: (1) using the first workpiece image center as origin, angle direction uniformly takes 720 sampled points；
(2) along each sampling angle direction, from outside to inside scanning, first aim point is the point on outer edge；(3) with the first work
Part picture centre is origin, is scanned from the inside to the outside, and first aim point is the point on inward flange.
Step S32, based on the marginal information of the inside and outside circle of bearing to be detected, with least square circle approximating method obtain to
The accurate positioning of the inside and outside circle of bearing is detected, and then obtains the second workpiece image.The step specifically includes:
Step S321, based on the marginal information of the inside and outside circle of bearing to be detected, the edge for treating the detection inside and outside circle of bearing is adopted
Sampling point does down-sampling；Initializing all edge sample points is interior point, and carries out Least Square Circle fitting based on this；
Step S322 judges the edge sample point of all inside and outside circles of bearing to be detected: if edge sample point and
The distance of the circle of fitting is less than C, then determines that edge sample point is interior point, be otherwise judged as exterior point；C is given threshold, this implementation
The radius of the circle of C=0.02* fitting in example；
Step S323, after judging the edge sample point of all inside and outside circles of bearing to be detected, using in step S322
The interior point set obtained after judgement carries out Least Square Circle fitting again, obtains the accurate positioning of the inside and outside circle of bearing to be detected, packet
Include the radius and central information of bearing face Internal and external cycle.
Second workpiece image refer to bearing face position and size information corresponding to geometric areas image, namely according to
The geometric parameter of outer circle and inner circle obtains the corresponding image-region of bearing face.
Step S4 is based on the second workpiece image, the annular workpieces figure of bearing face to be detected is expanded into rectangle work
Part figure.Fig. 3 is coordinate transform effect diagram, and upper figure is second workpiece image, and the following figure is the effect diagram after stretching.
The step specifically includes:
Step S41 obtains the center of bearing face to be detected based on the accurate positioning of the inside and outside circle of bearing to be detected
(x0,y0)；It obtains simultaneously and stretches start angle θ0, start radius r0, terminate radius r1Etc. parameters；
Step S42 obtains coordinate conversion parameter based on the rectangular image dimension information to be stretched.
In the present embodiment, tensile diagram image height is H=r1-r0, tensile diagram image width is W, then for stretching any point in image
(x, y), the then angle corresponded in polar coordinates areIt is r=r corresponding to the radius in polar coordinates0+ y is right
Should be in the coordinate in polar coordinatesIn more embodiment, tensile diagram image width is that W is axis to be detected
The perimeter of outer circle is held, W=2 π R, R are the radius of bearing top circle to be detected.
The annular workpieces figure of bearing face to be detected is expanded into rectangle work according to the coordinate conversion parameter by step S43
Part figure.
The rectangular piece figure is divided into multiple end region figures using the method for one direction projection localization by step S5.
After obtaining rectangular piece figure, by with one direction projection localization technology to obtain the outer chamfer region, outer of workpiece
Ring, sealing element, inner ring and inner chamfer region line of demarcation, thus realize workpiece area divide.Specific step is as follows:
Square workpiece shape figure is projected to short-axis direction, calculates mean value, obtains the gray scale point in workpiece height direction by step S51
Cloth curve；If drawing the width of rectangular piece figure for W, a height of H, then the intensity profile of short transverse isI=
1 ..., H, wherein f (i, j) is gray value of the image at coordinate (i, j).
Step S52 carries out Threshold segmentation according to gradient transformation to the intensity profile curve, obtains point in each region of workpiece
Boundary line.
Step S6 carries out binarization segmentation by adaptive gray threshold, obtains according to the different gamma characteristics in each region
Defect candidate region.Fig. 4 is final defects detection result schematic diagram.
The step needs each the described end region figure obtained step S5 to detect respectively, executes step such as
Under:
Step S61 calculates the average brightness value of end region figure；
Step S62 carries out Threshold segmentation according to average brightness value, if current grayvalue f (i, j) < mean (i, j)-T,
Current point is determined as defect point, wherein mean (i, j) is the gray average at coordinate (i, j), and T is preset constant offset
Amount；
Step S63 filters out too small defect point using morphologic filtering method, obtains defect candidate region.
Step S7, to the defect candidate region, using connected domain analysis method, and according to defective locations and shape information
Obtain final defect area.
Those skilled in the art should be able to recognize that, side described in conjunction with the examples disclosed in the embodiments of the present disclosure
Method step, can be realized with electronic hardware, computer software, or a combination of the two, in order to clearly demonstrate electronic hardware and
The interchangeability of software generally describes each exemplary composition and step according to function in the above description.These
Function is executed actually with electronic hardware or software mode, specific application and design constraint depending on technical solution.
Those skilled in the art can use different methods to achieve the described function each specific application, but this reality
Now it should not be considered as beyond the scope of the present invention.
The software that the step of method or algorithm of the embodiments described herein description can be executed with hardware, processor
The combination of module or the two is implemented.Software module can be placed in random access memory (RAM), memory, read-only memory
(ROM), electrically programmable ROM, electrically erasable ROM, register, hard disk, moveable magnetic disc, CD-ROM or technical field
In any other form of storage medium well known to interior.
Term " first ", " second " etc. are to be used to distinguish similar objects, rather than be used to describe or indicate specific suitable
Sequence or precedence.
Term " includes " or any other like term are intended to cover non-exclusive inclusion, so that including a system
Process, method, article or equipment/device of column element not only includes those elements, but also including being not explicitly listed
Other elements, or further include the intrinsic element of these process, method, article or equipment/devices.
So far, it has been combined preferred embodiment shown in the drawings and describes technical solution of the present invention, still, this field
Technical staff is it is easily understood that protection scope of the present invention is expressly not limited to these specific embodiments.Without departing from this
Under the premise of the principle of invention, those skilled in the art can make equivalent change or replacement to the relevant technologies feature, these
Technical solution after change or replacement will fall within the scope of protection of the present invention.
Claims (10)
1. a kind of bearing finished products plate defects detection method of view-based access control model, which comprises the following steps:
Step S1 obtains the acquired original image comprising bearing face image to be detected；
Step S2 obtains the first workpiece image of bearing face to be detected based on the acquired original image；
Step S3 carries out edge extracting to the first workpiece image, and obtains to extracted marginal point with the fitting algorithm of circle
The accurate positioning of bearing component obtains the second workpiece image of bearing face to be detected；
Step S4 is based on the second workpiece image, the annular workpieces figure of bearing face to be detected is expanded into rectangular piece figure；
The rectangular piece figure is divided into multiple end region figures using the method for one direction projection localization by step S5；
Step S6 carries out binarization segmentation by adaptive gray threshold, obtains defect according to the different gamma characteristics in each region
Candidate region；
Step S7 obtains the defect candidate region using connected domain analysis method, and according to defective locations and shape information
Final defect area.
2. the bearing finished products plate defects detection method of view-based access control model according to claim 1, which is characterized in that step S2
The middle method used obtains the first workpiece image of bearing face to be detected for image segmentation algorithm, comprising the following steps:
Step S21 obtains first using adaptive threshold fuzziness and connected component analysis method to the acquired original image
Candidate artifacts regional ensemble；
Step S22, based on the size in each candidate artifacts region, location information in the first candidate artifacts regional ensemble to candidate
Workpiece area is screened and is merged, and the second candidate artifacts regional ensemble is obtained；
Step S23 determines the first workpiece image based on the second candidate artifacts regional ensemble.
3. the bearing finished products plate defects detection method of view-based access control model according to claim 2, which is characterized in that its feature
It is, " obtains the first candidate artifacts regional ensemble " in step S21, method are as follows:
Step S211 carries out down-sampling to the acquired original image；
Step S212, the grey level histogram of statistical picture, and threshold value required for segmented image is acquired using ISODATA algorithm
t2a1；
Step S213 calculates the peak value t of the grey level histogram of image2a2, and further calculate the threshold value t for image segmentation2a3,
t2a3=A t2a1-Bt2a2, wherein A, B are predetermined coefficient；
Step S214 is based on t2a3Image segmentation is carried out, the first candidate artifacts regional ensemble is obtained.
4. the bearing finished products plate defects detection method of view-based access control model according to claim 2, which is characterized in that its feature
It is, " candidate artifacts region is screened and merged " in step S22, comprising: from the first candidate artifacts regional ensemble
Middle screening size meets the candidate artifacts region of given threshold, and will further have weight in the candidate artifacts region after screening
The candidate artifacts region of folded part merges, and generates the second candidate artifacts regional ensemble.
5. the bearing finished products plate defects detection method of view-based access control model according to claim 1, which is characterized in that step S3
Described in the obtaining step of second workpiece image include:
Step S31 uses edge extraction techniques to the first workpiece image, obtains its outermost edge under polar coordinate system and most interior
Marginal information obtains the marginal information of the inside and outside circle of bearing to be detected；
Step S32 is obtained to be detected based on the marginal information of the inside and outside circle of bearing to be detected with least square circle approximating method
The accurate positioning of the inside and outside circle of bearing, and then obtain the second workpiece image.
6. the bearing finished products plate defects detection method of view-based access control model according to claim 5, which is characterized in that step
" accurate positioning for obtaining the inside and outside circle of bearing to be detected ", method in S32 are as follows:
Step S321 treats the edge sample point of the detection inside and outside circle of bearing based on the marginal information of the inside and outside circle of bearing to be detected
Do down-sampling；Initializing all edge sample points is interior point, and carries out Least Square Circle fitting based on this；
Step S322 judges the edge sample point of all inside and outside circles of bearing to be detected: if edge sample point and fitting
Circle distance be less than C, then determine edge sample point be interior point, be otherwise judged as exterior point；C is given threshold；
Step S323 carries out Least Square Circle fitting using obtained interior point set after judging in step S322 again, obtain to
Detect the accurate positioning of the inside and outside circle of bearing.
7. the bearing finished products plate defects detection method of view-based access control model according to claim 1, which is characterized in that step S4
In " the annular workpieces figure of bearing face to be detected is expanded into rectangular piece figure ", method are as follows:
Step S41 obtains the center of bearing face to be detected based on the accurate positioning of the inside and outside circle of bearing to be detected；
Step S42 obtains coordinate conversion parameter based on the rectangular image dimension information to be stretched；
The annular workpieces figure of bearing face to be detected is expanded into rectangular piece according to the coordinate conversion parameter by step S43
Figure.
8. the bearing finished products plate defects detection method of view-based access control model according to claim 1, which is characterized in that step S5
Described in one direction projection localization method, comprising:
Rectangular piece figure is projected to short-axis direction, calculates mean value by step S51, and the intensity profile for obtaining workpiece height direction is bent
Line；
Step S52 carries out Threshold segmentation according to gradient transformation to the intensity profile curve, obtains the boundary in each region of workpiece
Line.
9. the bearing finished products plate defects detection method of view-based access control model according to claim 1, which is characterized in that step S6
In " obtain defect candidate region ", method are as follows: each described end region figure that step S5 is obtained, execute following step
It is rapid:
Step S61 calculates the average brightness value of end region figure；
Step S62 carries out Threshold segmentation according to average brightness value, if current grayvalue f (i, j) < mean (i, j)-T, will work as
Preceding point is determined as defect point, and wherein mean (i, j) is the gray average at coordinate (i, j), and T is preset constant offset amount；
Step S63 filters out too small defect point using morphologic filtering method, obtains defect candidate region.
10. the bearing finished products plate defects detection method of -9 described in any item view-based access control models, feature exist according to claim 1
In multiple end regions described in step S5 include: inner ring end region, outer ring end region, sealing cover area and chamfered region
Domain.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810764930.2A CN109187581A (en) | 2018-07-12 | 2018-07-12 | The bearing finished products plate defects detection method of view-based access control model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810764930.2A CN109187581A (en) | 2018-07-12 | 2018-07-12 | The bearing finished products plate defects detection method of view-based access control model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109187581A true CN109187581A (en) | 2019-01-11 |
Family
ID=64935977
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810764930.2A Pending CN109187581A (en) | 2018-07-12 | 2018-07-12 | The bearing finished products plate defects detection method of view-based access control model |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109187581A (en) |
Cited By (21)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109975307A (en) * | 2019-03-11 | 2019-07-05 | 中国科学院上海技术物理研究所 | Bearing surface defect detection system and detection method based on statistics projection training |
| CN110012288A (en) * | 2019-04-16 | 2019-07-12 | 昆山丘钛微电子科技有限公司 | A kind of vision positioning method and device of camera module camera lens |
| CN110288566A (en) * | 2019-05-23 | 2019-09-27 | 北京中科晶上科技股份有限公司 | A kind of target defect extracting method |
| CN110501342A (en) * | 2019-08-20 | 2019-11-26 | 北京信息科技大学 | A kind of cheese yarn bar positioning visible detection method |
| CN110766684A (en) * | 2019-10-30 | 2020-02-07 | 江南大学 | Stator surface defect detection system and detection method based on machine vision |
| CN110927171A (en) * | 2019-12-09 | 2020-03-27 | 中国科学院沈阳自动化研究所 | Bearing roller chamfer surface defect detection method based on machine vision |
| CN111047574A (en) * | 2019-12-12 | 2020-04-21 | 韦士肯(厦门)智能科技有限公司 | Visual inspection defect identification method for cylindrical workpiece |
| CN111862023A (en) * | 2020-07-14 | 2020-10-30 | 苏州华启智能科技有限公司 | Pantograph structure abnormity detection method based on computer vision |
| CN112434694A (en) * | 2020-11-20 | 2021-03-02 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying damage fault of outer ring of front cover of rolling bearing |
| CN112730459A (en) * | 2021-02-05 | 2021-04-30 | 北京嘉恒中自图像技术有限公司 | Bearing inner ring end surface defect detection method based on multi-angle light source shooting image |
| CN112781452A (en) * | 2021-03-25 | 2021-05-11 | 湘潭大学 | Bullet primer top appearance defect detection method |
| CN112858332A (en) * | 2021-02-22 | 2021-05-28 | 菲特(珠海横琴)智能科技有限公司 | Synchronizer gear hub surface defect detection method, system and terminal based on machine vision |
| CN112862832A (en) * | 2020-12-31 | 2021-05-28 | 重庆盛泰光电有限公司 | Dirt detection method based on concentric circle segmentation positioning |
| CN112991284A (en) * | 2021-03-05 | 2021-06-18 | 佛山科学技术学院 | Temperature controller guide frame defect detection method and system |
| CN113139943A (en) * | 2021-04-22 | 2021-07-20 | 苏州华兴源创科技股份有限公司 | Method and system for detecting appearance defects of open circular ring workpiece and computer storage medium |
| CN113686876A (en) * | 2021-08-24 | 2021-11-23 | 华南农业大学 | Poultry egg crack detection method and device |
| CN114544173A (en) * | 2022-01-21 | 2022-05-27 | 慧三维智能科技(苏州)有限公司 | Bearing defect detection equipment |
| CN114841971A (en) * | 2022-05-10 | 2022-08-02 | 广州大学 | Method, system and device for detecting pit defects of metal block |
| CN114882044A (en) * | 2022-07-12 | 2022-08-09 | 山东汇通工业制造有限公司 | Metal pipe surface quality detection method |
| CN115063431A (en) * | 2022-08-19 | 2022-09-16 | 山东远盾网络技术股份有限公司 | Automobile part quality tracing method based on image processing |
| CN116342589A (en) * | 2023-05-23 | 2023-06-27 | 之江实验室 | Cross-field scratch defect continuity detection method and system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101699217A (en) * | 2009-11-03 | 2010-04-28 | 武汉大学 | Method used for detecting concentric circle of industrial part |
| CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
| CN106204614A (en) * | 2016-07-21 | 2016-12-07 | 湘潭大学 | A kind of workpiece appearance defects detection method based on machine vision |
| CN106404793A (en) * | 2016-09-06 | 2017-02-15 | 中国科学院自动化研究所 | Method for detecting defects of bearing sealing element based on vision |
| CN106767425A (en) * | 2016-11-07 | 2017-05-31 | 无锡浩远视觉科技有限公司 | A kind of vision measuring method of bearing snap spring gap |
| CN107884413A (en) * | 2017-10-24 | 2018-04-06 | 华东交通大学 | A kind of device and detection method of automatic detection bearing roller missing |
-
2018
- 2018-07-12 CN CN201810764930.2A patent/CN109187581A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101699217A (en) * | 2009-11-03 | 2010-04-28 | 武汉大学 | Method used for detecting concentric circle of industrial part |
| CN103499590A (en) * | 2013-10-17 | 2014-01-08 | 福州大学 | Method and system for detecting and screening end defects in circular parts |
| CN106204614A (en) * | 2016-07-21 | 2016-12-07 | 湘潭大学 | A kind of workpiece appearance defects detection method based on machine vision |
| CN106404793A (en) * | 2016-09-06 | 2017-02-15 | 中国科学院自动化研究所 | Method for detecting defects of bearing sealing element based on vision |
| CN106767425A (en) * | 2016-11-07 | 2017-05-31 | 无锡浩远视觉科技有限公司 | A kind of vision measuring method of bearing snap spring gap |
| CN107884413A (en) * | 2017-10-24 | 2018-04-06 | 华东交通大学 | A kind of device and detection method of automatic detection bearing roller missing |
Non-Patent Citations (3)
| Title |
|---|
| SHEN HAO ET AL.: "Bearing defect inspection based on machine vision", 《MEASUREMENT》 * |
| 王恒迪等: "轴承端面缺陷的视觉检测方法", 《轴承》 * |
| 郝勇等: "基于机器视觉的深沟球轴承滚珠遗漏检测", 《激光与光电子进展》 * |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109975307A (en) * | 2019-03-11 | 2019-07-05 | 中国科学院上海技术物理研究所 | Bearing surface defect detection system and detection method based on statistics projection training |
| CN110012288B (en) * | 2019-04-16 | 2021-04-13 | 昆山丘钛微电子科技有限公司 | Visual positioning method and device for camera module lens |
| CN110012288A (en) * | 2019-04-16 | 2019-07-12 | 昆山丘钛微电子科技有限公司 | A kind of vision positioning method and device of camera module camera lens |
| CN110288566B (en) * | 2019-05-23 | 2021-12-07 | 北京中科晶上科技股份有限公司 | Target defect extraction method |
| CN110288566A (en) * | 2019-05-23 | 2019-09-27 | 北京中科晶上科技股份有限公司 | A kind of target defect extracting method |
| CN110501342A (en) * | 2019-08-20 | 2019-11-26 | 北京信息科技大学 | A kind of cheese yarn bar positioning visible detection method |
| CN110501342B (en) * | 2019-08-20 | 2022-06-07 | 北京信息科技大学 | Cheese yarn rod positioning visual detection method |
| CN110766684B (en) * | 2019-10-30 | 2022-03-18 | 江南大学 | Stator surface defect detection system and detection method based on machine vision |
| CN110766684A (en) * | 2019-10-30 | 2020-02-07 | 江南大学 | Stator surface defect detection system and detection method based on machine vision |
| CN110927171A (en) * | 2019-12-09 | 2020-03-27 | 中国科学院沈阳自动化研究所 | Bearing roller chamfer surface defect detection method based on machine vision |
| CN111047574A (en) * | 2019-12-12 | 2020-04-21 | 韦士肯(厦门)智能科技有限公司 | Visual inspection defect identification method for cylindrical workpiece |
| CN111047574B (en) * | 2019-12-12 | 2024-06-18 | 苏州韦士肯检测科技有限公司 | Visual inspection defect identification method for cylindrical workpiece |
| CN111862023A (en) * | 2020-07-14 | 2020-10-30 | 苏州华启智能科技有限公司 | Pantograph structure abnormity detection method based on computer vision |
| CN112434694A (en) * | 2020-11-20 | 2021-03-02 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying damage fault of outer ring of front cover of rolling bearing |
| CN112434694B (en) * | 2020-11-20 | 2021-07-16 | 哈尔滨市科佳通用机电股份有限公司 | Method and system for identifying damage fault of outer ring of front cover of rolling bearing |
| CN112862832A (en) * | 2020-12-31 | 2021-05-28 | 重庆盛泰光电有限公司 | Dirt detection method based on concentric circle segmentation positioning |
| CN112730459A (en) * | 2021-02-05 | 2021-04-30 | 北京嘉恒中自图像技术有限公司 | Bearing inner ring end surface defect detection method based on multi-angle light source shooting image |
| CN112730459B (en) * | 2021-02-05 | 2023-01-24 | 北京嘉恒中自图像技术有限公司 | Bearing inner ring end surface defect detection method based on multi-angle light source shooting image |
| CN112858332A (en) * | 2021-02-22 | 2021-05-28 | 菲特(珠海横琴)智能科技有限公司 | Synchronizer gear hub surface defect detection method, system and terminal based on machine vision |
| CN112991284B (en) * | 2021-03-05 | 2022-11-01 | 佛山科学技术学院 | Temperature controller guide frame defect detection method and system |
| CN112991284A (en) * | 2021-03-05 | 2021-06-18 | 佛山科学技术学院 | Temperature controller guide frame defect detection method and system |
| CN112781452A (en) * | 2021-03-25 | 2021-05-11 | 湘潭大学 | Bullet primer top appearance defect detection method |
| CN112781452B (en) * | 2021-03-25 | 2022-10-18 | 湘潭大学 | Bullet primer top appearance defect detection method |
| CN113139943B (en) * | 2021-04-22 | 2022-08-12 | 苏州华兴源创科技股份有限公司 | Method and system for detecting appearance defects of open circular ring workpiece and computer storage medium |
| WO2022222467A1 (en) * | 2021-04-22 | 2022-10-27 | 苏州华兴源创科技股份有限公司 | Open circular ring workpiece appearance defect detection method and system, and computer storage medium |
| CN113139943A (en) * | 2021-04-22 | 2021-07-20 | 苏州华兴源创科技股份有限公司 | Method and system for detecting appearance defects of open circular ring workpiece and computer storage medium |
| CN113686876A (en) * | 2021-08-24 | 2021-11-23 | 华南农业大学 | Poultry egg crack detection method and device |
| CN113686876B (en) * | 2021-08-24 | 2022-10-04 | 华南农业大学 | Poultry egg crack detection method and device |
| CN114544173A (en) * | 2022-01-21 | 2022-05-27 | 慧三维智能科技(苏州)有限公司 | Bearing defect detection equipment |
| CN114544173B (en) * | 2022-01-21 | 2022-09-30 | 慧三维智能科技(苏州)有限公司 | Bearing defect detection equipment |
| CN114841971A (en) * | 2022-05-10 | 2022-08-02 | 广州大学 | Method, system and device for detecting pit defects of metal block |
| CN114882044B (en) * | 2022-07-12 | 2022-09-06 | 山东汇通工业制造有限公司 | Metal pipe surface quality detection method |
| CN114882044A (en) * | 2022-07-12 | 2022-08-09 | 山东汇通工业制造有限公司 | Metal pipe surface quality detection method |
| CN115063431A (en) * | 2022-08-19 | 2022-09-16 | 山东远盾网络技术股份有限公司 | Automobile part quality tracing method based on image processing |
| CN115063431B (en) * | 2022-08-19 | 2022-11-11 | 山东远盾网络技术股份有限公司 | Automobile part quality tracing method based on image processing |
| CN116342589A (en) * | 2023-05-23 | 2023-06-27 | 之江实验室 | Cross-field scratch defect continuity detection method and system |
| CN116342589B (en) * | 2023-05-23 | 2023-08-22 | 之江实验室 | Cross-field scratch defect continuity detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109187581A (en) | The bearing finished products plate defects detection method of view-based access control model | |
| CN111310558B (en) | Intelligent pavement disease extraction method based on deep learning and image processing method | |
| CN108921176B (en) | Pointer instrument positioning and identifying method based on machine vision | |
| CN110389127B (en) | System and method for identifying metal ceramic parts and detecting surface defects | |
| EP1430446B1 (en) | Image processing method for appearance inspection | |
| CN108596880A (en) | Weld defect feature extraction based on image procossing and welding quality analysis method | |
| CN109242870A (en) | A kind of sea horizon detection method divided based on image with textural characteristics | |
| CN107341790B (en) | A kind of image processing method of environment cleanliness detection | |
| CN109859160B (en) | Aluminum magnesium alloy casting internal defect image identification method based on machine vision | |
| CN109087286A (en) | A kind of detection method and application based on Computer Image Processing and pattern-recognition | |
| CN107610104A (en) | Crack detecting method at a kind of elevator compensation chain R based on machine vision | |
| CN109682839A (en) | A kind of metal arc Surface Flaw online test method | |
| CN107610085A (en) | A kind of welding point defect detecting system based on computer vision | |
| CN108921004A (en) | Safety cap wears recognition methods, electronic equipment, storage medium and system | |
| CN106780526A (en) | A kind of ferrite wafer alligatoring recognition methods | |
| CN108186051A (en) | A kind of image processing method and processing system of the automatic measurement fetus Double Tops electrical path length from ultrasonoscopy | |
| CN112862744B (en) | Intelligent detection method for internal defects of capacitor based on ultrasonic image | |
| CN108537787A (en) | A kind of quality judging method of facial image | |
| CN112734761A (en) | Industrial product image boundary contour extraction method | |
| CN109544513A (en) | A kind of steel pipe end surface defect extraction knowledge method for distinguishing | |
| CN113588656A (en) | Novel bolt appearance size detection method | |
| CN117237747B (en) | Hardware defect classification and identification method based on artificial intelligence | |
| Junxiong et al. | Feature extraction of jujube fruit wrinkle based on the watershed segmentation | |
| CN116740449A (en) | Shaving form detection method and system based on AI (advanced technology attachment) computer vision technology | |
| CN111210419A (en) | Micro magnetic tile surface defect detection method based on human visual characteristics |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190111 |
|
| WD01 | Invention patent application deemed withdrawn after publication |