CN108387580A - A kind of bearing defect detection device - Google Patents
A kind of bearing defect detection device Download PDFInfo
- Publication number
- CN108387580A CN108387580A CN201810130619.2A CN201810130619A CN108387580A CN 108387580 A CN108387580 A CN 108387580A CN 201810130619 A CN201810130619 A CN 201810130619A CN 108387580 A CN108387580 A CN 108387580A
- Authority
- CN
- China
- Prior art keywords
- steel ball
- defect
- bearing
- image
- processing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- 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
Abstract
The invention discloses a kind of bearing defect detection devices, including bearing steel ball conveying mechanism, separation system, video camera, defect recognition host and remote terminal；The bearing steel ball conveying mechanism is used to be responsible for the transport of steel ball to be detected, and steel ball is transported to separation system；The separation system is used to carry out preliminary classification to steel ball, and steel ball classification is transported to immediately below video camera：The video camera setting captures steel ball image, and steel ball image is exported and gives defect recognition host in real time in the top of steel ball to be measured；The defect recognition host judges whether steel ball defective and defect classification, and defects detection result is exported to remote terminal by carrying out analyzing processing to steel ball image.The present invention can detect the defect condition of steel ball in bearing automatically, and testing result is accurate, and detection efficiency is high, avoids influence of the artificial detection uncertain factor to testing result.
Description
Technical field
The present invention relates to Bearing testing technical fields, and in particular to a kind of bearing defect detection device.
Background technology
Bearing is for determining rotary shaft and other parts relative motion position, playing the parts of support or guiding role.
The major function of bearing is support rotary shaft or other movable bodies, guides rotation or moving movement and bears by axis or parts on shaft
The load being transmitted to.Steel ball plays critically important effect within the bearing in bearing, and defect, which will once occur, in steel ball to be seriously affected
The service life of bearing.The detection for steel ball in bearing is usually artificial visual inspection at present, and labor intensity is big, and efficiency is low,
Testing result is unreliable.
Invention content
In view of the above-mentioned problems, the present invention provides a kind of bearing defect detection device, carried to solve background section above
To the problem of.
The purpose of the present invention is realized using following technical scheme：
A kind of bearing defect detection device, including bearing steel ball conveying mechanism, separation system, video camera and defect recognition master
Machine；Bearing steel ball conveying mechanism is used to be responsible for the transport of steel ball to be detected, steel ball is transported to separation system, separation system is used for
Preliminary classification is carried out to steel ball, and steel ball classification is transported to immediately below video camera；Video camera setting is upper steel ball to be detected
Side captures steel ball image, and exports and defect recognition host, defect recognition host is given to handle steel ball image in real time, judges
Steel ball exports defects detection result with the presence or absence of defect and the classification of defect.
Preferably, which further includes wireless data transfer module and remote terminal, and wireless data passes
Defeated module is used to obtained defects detection result radioing to remote terminal.
Preferably, video camera selects CCD camera.
Beneficial effects of the present invention are：Bearing defect detection device proposed by the present invention can detect steel in bearing automatically
The defect condition of ball, testing result is accurate, and detection efficiency is high, avoids shadow of the artificial detection uncertain factor to testing result
It rings.
Description of the drawings
Using attached drawing, the invention will be further described, but the embodiment in attached drawing does not constitute any limit to the present invention
System, for those of ordinary skill in the art, without creative efforts, can also obtain according to the following drawings
Other attached drawings.
Fig. 1 is bearing defect structure of the detecting device schematic diagram provided in an embodiment of the present invention；
Fig. 2 is the frame construction drawing of defect recognition host in the embodiment of the present invention.
Reference numeral：Bearing steel ball conveying mechanism 1；Separation system 2；Video camera 3；Defect recognition host 4；Wireless data passes
Defeated module 5；Remote terminal 6；Pre-processing module 40；Characteristic extracting module 41；Defect characteristic identification module 42；Enhancement unit
401；Denoising unit 402；Cutting unit 403.
Specific implementation mode
The invention will be further described with the following Examples.
A kind of bearing defect detection device, including bearing steel ball conveying mechanism 1, separation system 2, video camera 3 and defect are known
Other host 4；Bearing steel ball conveying mechanism 1 is used to be responsible for the transport of steel ball to be detected, and steel ball is transported to separation system 2, sorts
System 2 is used to carry out preliminary classification to steel ball, and steel ball classification is transported to immediately below video camera 3；Video camera 3 is arranged to be checked
Survey the top of steel ball, capture steel ball image in real time, and export to defect recognition host 4, defect recognition host 4 to steel ball image into
Row processing judges steel ball with the presence or absence of defect and the classification of defect, and exports defects detection result.
Preferably, which further includes wireless data transfer module 5 and remote terminal 6, wireless data
Transmission module 5 is used to the defects detection result that defect recognition host 4 obtains radioing to remote terminal 6.
Preferably, video camera 3 selects CCD camera.
Preferably, defect recognition host includes pre-processing module 40, characteristic extracting module 41 and defect characteristic identification mould
Block 42；Pre-processing module 40 is used to carry out pre-processing to the steel ball image of crawl；Characteristic extracting module 41 was used for from early period
Steel ball characteristic parameter is extracted in treated steel ball image；The steel ball characteristic parameter that defect characteristic identification module 42 is used to obtain
It is compared, judges steel ball with the presence or absence of defect and lacks with the steel ball defect characteristic parameter to prestore in defect recognition host
Sunken classification, and export defect recognition result.
Preferably, pre-processing module 40 includes enhancement unit 401, denoising unit 402 and cutting unit 403；Enhancing is single
Member 401 to the steel ball image of crawl for carrying out enhancing processing；Denoising unit 402 is for removing in enhanced steel ball image
Random noise；Cutting unit 403 is to be split processing to the steel ball image after denoising using FCM algorithms.
Preferably, enhancing processing is carried out to the steel ball image of crawl, specially：
(1) gray processing processing is carried out to the steel ball image of crawl；
(2) scanning window that size is M × M is set, using the scanning window to gray processing treated steel ball figure
As being scanned, and using the gray value of each pixel in following enhancing formula calculating scanning window, wherein enhancing is public
Formula is：
In formula, Y (m, n) is the gray value at pixel (m, n), Y in scanning windownew(m, n) is enhancing in scanning window
The gray value of treated pixel (m, n), YminFor the minimum gradation value that pixel (m, n) is expert in scanning window, Ymax
For the maximum gradation value that pixel (m, n) is expert in scanning window；A, b, c are regulatory factor；
(3) all pixels point in the image of traversal after gray processing, the set that all pixels point is constituted are to enhance
Treated steel ball image.
Advantageous effect：Enhancing processing is carried out to steel ball image using above formula, which can be according to each in steel ball image
The gray value of pixel it is adaptive carry out enhancing operation, which can not only improve the light and shade pair of enhanced steel ball image
Than degree, and the contrast of enhanced steel ball picture edge characteristic is also improved, is conducive to subsequently to steel ball defect characteristic
Extraction, and then improve bearing in steel ball defects detection accuracy rate.
Preferably, the random noise in enhanced steel ball image is removed, specially：Using following denoising formula to enhancing
Treated steel ball image carries out denoising, wherein the denoising formula is：
In formula, F (m, n) is the gray value after denoising at pixel (m, n), Ynew(m, n) is after enhanced processing
Image in gray value at pixel (m, n),For the shade of gray value of pixel (m, n) in horizontal direction,For the shade of gray value of pixel (m, n) in vertical direction, δ is the preset parameter of setting, Ω be with pixel (m,
N) centered on, the rectangular area that size is 3 × 3, (m+i, n+j) indicates the pixel point coordinates in the Ω of region, wherein i ∈ [-
1,1], [- 1,1] j ∈, w (m, n) are Filtering Template function, and α 1 and α 2 are weight factor.
Advantageous effect：Using above-mentioned smoothing formula, treated that steel ball image is smoothed to enhanced, the formula
The smoothing processing that selectivity can be carried out to steel ball image can be according to the gray scale of pixel in the Ω of region in filtering
Value automatically updates Filtering Template function, is smoothed in noise region, then without smoothly locating in non-noise region
Reason, this algorithm enable to the fog-level of steel ball image to be preferably minimized, improve the clarity of steel ball image, be conducive to improve
The subsequently accuracy of the cutting operation to steel ball image and steel ball image deflects feature extraction.
Preferably, described that processing is split to the steel ball image after denoising using FCM algorithms, specially：
(1) setting clusters number c, FUZZY WEIGHTED index d and iteration ends threshold value λ, and primary iteration number r=0 is set,
Wherein 2≤c≤n, d >=2, n are the sample numbers of Cluster space；
(2) initialization cluster centre matrix V (0)=[vk], wherein k=1,2 ..., c, vkFor k-th of cluster centre；
(3) subordinated-degree matrix U is calculated using following formula according to cluster centre matrix(r)=[ukp], wherein k=1,2 ..., c；
P=1,2 ..., N, N be denoising after steel ball image in pixel number：
In formula, xpFor the gray value of p-th of pixel in the steel ball image after denoising, ukpIndicate that p-th of pixel belongs to
K-th class is subordinate to angle value, dqpIndicate p-th of pixel and cluster centre vqBetween Euclidean distance, dkqIndicate cluster centre
vkWith cluster centre vqBetween Euclidean distance, and meet k ≠ q, q=1,2 ..., c, vkIndicate k-th of cluster centre；
All pixels point is traversed, the matrix for being subordinate to angle value composition of all pixels point is subordinated-degree matrix U(r)；
(4) forK updates cluster centre matrix V according to the following formula(r)For V(r+1)：
In formula, vkIndicate k-th of cluster centre, xpIndicate the gray value of p-th of pixel in the steel ball image after denoising,
It indicates centered on p-th of pixel, a size is the average gray value of 3 × 3 image blocks；dqpIndicate p-th pixel with
Cluster centre vqBetween Euclidean distance, dkqIndicate cluster centre vkWith cluster centre vqBetween Euclidean distance, and meet k ≠
Q, q=1,2 ..., c；ukpIndicate that p-th of pixel belonged to k-th class is subordinate to angle value；
(5) according to obtained V(r+1)With the formula in step (3), update subordinated-degree matrix U(r+1)；
(6) compare V(r)And V(r+1)If | | V(r+1)-V(r)| | ＜ λ then stop iteration, wherein λ is iteration ends threshold
Value；Otherwise, step (4) is gone to, iteration is until convergence；After iteration ends, according to cluster centre matrix is obtained, using maximum
Degree of membership de-fuzzy algorithm converts the result of fuzzy classification to determining classification, to realize to the steel ball image after denoising
Cutting operation.
Advantageous effect：Operation is split to the steel ball image after denoising using above-mentioned formula, which fully considers
Restriction relation between pixel and cluster centre, while the gray average for being also contemplated for central pixel point is subordinate to central pixel point
The influence of angle value, the influence which can be with this restriction relation of precise expression to central pixel point, and then can obtain can
Characterize defect area in steel ball image.
Finally it should be noted that the above embodiments are merely illustrative of the technical solutions of the present invention, rather than the present invention is protected
The limitation of range is protected, although being explained in detail to the present invention with reference to preferred embodiment, those skilled in the art answer
Work as understanding, technical scheme of the present invention can be modified or replaced equivalently, without departing from the reality of technical solution of the present invention
Matter and range.
Claims (6)
1. a kind of bearing defect detection device, which is characterized in that including bearing steel ball conveying mechanism, separation system, video camera and
Defect recognition host；The bearing steel ball conveying mechanism is used to be responsible for the transport of steel ball to be detected, and steel ball is transported to sorting system
System, the separation system is used to carry out preliminary classification to steel ball, and steel ball classification is transported to immediately below video camera；The camera shooting
Machine is arranged in the top of steel ball to be detected, captures steel ball image in real time, and exports to the defect recognition host, and the defect is known
Other host handles steel ball image, judges steel ball with the presence or absence of defect and the classification of defect, and export defects detection knot
Fruit.
2. bearing defect detection device according to claim 1, which is characterized in that further include wireless data transfer module and
Remote terminal, the defects detection result that the wireless data transfer module is used to obtain in the defect recognition host are wirelessly transferred
To the remote terminal.
3. bearing defect detection device according to claim 2, which is characterized in that the video camera selects CCD camera.
4. bearing defect detection device according to claim 3, which is characterized in that the defect recognition host includes early period
Processing module, characteristic extracting module and defect characteristic identification module；The pre-processing module is used for the steel ball image to crawl
Carry out pre-processing；The characteristic extracting module is used to extract steel ball characteristic parameter from treated early period steel ball image；Institute
State steel ball defect of the defect characteristic identification module for will prestore in obtained steel ball characteristic parameter and the defect recognition host
Characteristic parameter is compared, and judges steel ball with the presence or absence of defect and the classification of defect, and export defect recognition result.
5. bearing defect detection device according to claim 4, which is characterized in that the pre-processing module includes enhancing
Unit, denoising unit and cutting unit；
The enhancement unit to the steel ball image of crawl for carrying out enhancing processing；
The denoising unit is used to remove the random noise in enhanced steel ball image；
The cutting unit is to be split processing to the steel ball image after denoising using FCM algorithms.
6. bearing defect detection device according to claim 5, which is characterized in that the steel ball image of described pair of crawl carries out
Enhancing is handled, specially：
(1) gray processing processing is carried out to the steel ball image of crawl；
(2) scanning window that size is M × M is set, using the scanning window to gray processing treated steel ball image into
Row scanning, and using the gray value of each pixel in following enhancing formula calculating scanning window, wherein enhancing formula is：
In formula, Y (m, n) is the gray value at pixel (m, n), Y in scanning windownew(m, n) is enhancing processing in scanning window
The gray value of pixel (m, n) afterwards, YminFor the minimum gradation value that pixel (m, n) is expert in scanning window, YmaxTo sweep
Retouch the maximum gradation value that pixel (m, n) is expert in window；A, b, c are regulatory factor；
(3) all pixels point in the image that traversal is handled through gray processing, the set that all pixels point is constituted are after enhancing is handled
Steel ball image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810130619.2A CN108387580A (en) | 2018-02-08 | 2018-02-08 | A kind of bearing defect detection device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810130619.2A CN108387580A (en) | 2018-02-08 | 2018-02-08 | A kind of bearing defect detection device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108387580A true CN108387580A (en) | 2018-08-10 |
Family
ID=63074745
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810130619.2A Pending CN108387580A (en) | 2018-02-08 | 2018-02-08 | A kind of bearing defect detection device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108387580A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN115239738A (en) * | 2022-09-26 | 2022-10-25 | 南通鑫生派智能科技有限公司 | Intelligent detection method for automobile part configuration |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04127010A (en) * | 1989-12-12 | 1992-04-28 | Galai Lab Ltd | Inspection system of the ball of a bearing and its method of inspection |
| US20060091333A1 (en) * | 2002-06-21 | 2006-05-04 | Cochran Don W | Patterned illumination method and apparatus for machine vision systems |
| CN202903780U (en) * | 2012-10-30 | 2013-04-24 | 敖荣庆 | Nondestructive inspection sorting device for rolling bearing steel ball |
| CN103286081A (en) * | 2013-05-07 | 2013-09-11 | 浙江工业大学 | Monocular multi-perspective machine vision-based online automatic sorting device for steel ball surface defect |
| CN106914428A (en) * | 2017-01-16 | 2017-07-04 | 哈尔滨理工大学 | A kind of New Algorithm of the steel ball surface defect Differential Detection based on machine vision |
| CN107492093A (en) * | 2017-07-20 | 2017-12-19 | 浙江大学 | Bearing method for detecting abnormality based on image procossing |
| CN107543826A (en) * | 2016-06-29 | 2018-01-05 | 缪建良 | A kind of bearing defect detection means |
-
2018
- 2018-02-08 CN CN201810130619.2A patent/CN108387580A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH04127010A (en) * | 1989-12-12 | 1992-04-28 | Galai Lab Ltd | Inspection system of the ball of a bearing and its method of inspection |
| US20060091333A1 (en) * | 2002-06-21 | 2006-05-04 | Cochran Don W | Patterned illumination method and apparatus for machine vision systems |
| CN202903780U (en) * | 2012-10-30 | 2013-04-24 | 敖荣庆 | Nondestructive inspection sorting device for rolling bearing steel ball |
| CN103286081A (en) * | 2013-05-07 | 2013-09-11 | 浙江工业大学 | Monocular multi-perspective machine vision-based online automatic sorting device for steel ball surface defect |
| CN107543826A (en) * | 2016-06-29 | 2018-01-05 | 缪建良 | A kind of bearing defect detection means |
| CN106914428A (en) * | 2017-01-16 | 2017-07-04 | 哈尔滨理工大学 | A kind of New Algorithm of the steel ball surface defect Differential Detection based on machine vision |
| CN107492093A (en) * | 2017-07-20 | 2017-12-19 | 浙江大学 | Bearing method for detecting abnormality based on image procossing |
Non-Patent Citations (1)
| Title |
|---|
| 杨东林等: "轴承钢球表面缺陷的快速检测方法", 《兵工学报》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| 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 |
| CN115239738A (en) * | 2022-09-26 | 2022-10-25 | 南通鑫生派智能科技有限公司 | Intelligent detection method for automobile part configuration |
| CN115239738B (en) * | 2022-09-26 | 2022-12-09 | 南通鑫生派智能科技有限公司 | Intelligent detection method for automobile part configuration |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106910186B (en) | Bridge crack detection and positioning method based on CNN deep learning | |
| CN104483326B (en) | High-voltage line defects of insulator detection method and system based on depth belief network | |
| CN111814711A (en) | Image feature fast matching method and system applied to mine machine vision | |
| CN107203990A (en) | A kind of labeling damage testing method based on template matches and image quality measure | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN101059425A (en) | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis | |
| CN111292321B (en) | Transmission line insulator defect image identification method | |
| CN109191421A (en) | Cylindricality lithium battery periphery pit visible detection method | |
| CN111062938B (en) | Plate expansion plug detection system and method based on machine learning | |
| CN109146859A (en) | A kind of pavement crack detection system based on machine vision | |
| CN113706566B (en) | Edge detection-based perfuming and spraying performance detection method | |
| CN110781913B (en) | Zipper cloth belt defect detection method | |
| CN109345524A (en) | A kind of bearing open defect detection system of view-based access control model | |
| CN112149543A (en) | Building raise dust identification system and method based on computer vision | |
| CN107240112A (en) | Individual X Angular Point Extracting Methods under a kind of complex scene | |
| WO2023231262A1 (en) | Hoisting wire rope tension measurement method based on visual vibration frequency identification | |
| CN114549441A (en) | Sucker defect detection method based on image processing | |
| CN111667475A (en) | Machine vision-based Chinese date grading detection method | |
| CN110554052A (en) | artificial board surface defect detection method and system | |
| CN108387580A (en) | A kind of bearing defect detection device | |
| CN112634262A (en) | Writing quality evaluation method based on Internet | |
| CN111665199A (en) | Wire and cable color detection and identification method based on machine vision | |
| CN115841633A (en) | Power tower and power line associated correction power tower and power line detection method | |
| CN108305270A (en) | A kind of storage grain worm number system and method based on mobile phone photograph | |
| CN112184619A (en) | Metal part surface defect detection method based on deep learning |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180810 |
|
| RJ01 | Rejection of invention patent application after publication |