CN106053485A - Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects - Google Patents
Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects Download PDFInfo
- Publication number
- CN106053485A CN106053485A CN201610615717.6A CN201610615717A CN106053485A CN 106053485 A CN106053485 A CN 106053485A CN 201610615717 A CN201610615717 A CN 201610615717A CN 106053485 A CN106053485 A CN 106053485A
- Authority
- CN
- China
- Prior art keywords
- steel ball
- image
- inspection
- detection
- machine vision
- 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
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/951—Balls
Abstract
The invention discloses a machine vision-based novel method of intelligent circular inspection of steel ball surface defects. Compared with the traditional manual detection, mechanical detection and ray detection, the novel method has relatively high detection precision and reliability. The intelligent circular inspection algorithm is characterized by using the center of a circle of a steel ball in a captured image as a center, outwardly conducting circling inspection in turn at an interval of several pixels in a circulating way within fifty milliseconds, counting and calculating the gray value of each pixel, taking a difference between the maximum value and the minimum value of the gray value and comparing with a threshold value, thereby judging whether the quality of the steel ball is good or bad. By combining the high efficiency and repeatability of a computer with a digital image processing technology by using opencv, and fusing the intelligent circular inspection algorithm for visual inspection, the six balls presented in a visual image are sequentially treated; the test confirms that the detection efficiency and precision can be greatly improved, and the application prospect in industrial production is wide.
Description
Technical field
The present invention relates to field of precision measurement, a kind of steel ball surface defect intelligence based on machine vision annular
Follow the New Algorithm of inspection.
Background technology
Along with making rapid progress of science and technology, rolling bearing is widely used in machine-building, transportation, Aero-Space
And the field such as instrument and meter, and steel ball is as the core component of bearing, and the quality of its quality is for the performance of bearing and safety
Play key effect.Traditional method for piano plate mainly has: manual detection, mechanical detection and ray detection.Wherein,
Manual detection labor intensity is big, and efficiency is low, of poor quality, and detection reliability depends on subjective factors more；Mechanical detection is usual
Being contact measurement, detection process needs steel ball to be measured is carried out position adjustment, thus inefficient；Its advantage of ray detection is
Image resolution ratio can be improved, but structure is complicated, cost high；And by the high-efficiency and precision reliability of computer and Digital Image Processing
Technology combines and carries out vision-based detection, has the features such as full-automation, noncontact, high accuracy, detection efficiency can be greatly improved, tool
Have wide application prospects, be widely used in developed country at present, but application at home is still in developmental stage.
Full-fledged along with mechanical vision inspection technology, its applied research in piano plate field is increasingly by state
The attention of inside and outside scholar, becomes the mainstream research direction in current piano plate field.Along with the development of machine vision technique,
It constantly displays at the superiority of field of non destructive testing, for instance, it is possible to realize non-cpntact measurement；Simple to operate, detection essence
Degree height；Low etc. to detection environmental requirement.And in piano plate field, application visible sensation method carries out piano plate can be directly
Obtain steel ball surface defect information, breach the restriction of contact measurement and point by point scanning, therefore suffer from the favor of researcher.
According to machinery industry standard " rolling bearing steel ball surface defect atlas and assessment method ", steel ball surface defect divides
For speckle, pit, scratch, draw bar and pit etc..The face crack that produces during Mechanical Product's Machining, cut, stain, recessed
The defective effects such as hole, hole, burr are to product serviceability, integrity, even safety.Comprehensive piano plate instrument both at home and abroad
Development, applying most is optical inspection method and detection method of eddy.Owing to optical inspection method has certain limitation, now
Application gradually decreases, and detection method of eddy is the most ripe through development.In field of precision measurement, for instrument and meter
Certainty of measurement require more and more higher, but general sensor is both for single measured parameter and serves instrumentation or device,
The measurement of individual event parameter can only be realized, and in terms of precision and reliability, still have several drawbacks part.
Summary of the invention
Goal of the invention: the problem existed for traditional steel ball surface defect detection method, overcomes the deficiency of prior art,
The pixel gray value presented according to the defective region of steel ball surface is abnormal, utilizes computer vision storehouse, efficient by computer
Property and repeatable combine with digital image processing techniques, merge intelligence annular and follow and examine algorithm and carry out vision-based detection, to vision
Six steel balls presented in image carry out high speed processing, have the features such as high accuracy, high reliability, high efficiency, can accurately sentence
The quality problems of disconnected steel ball.Result sampling statistics after testing, accuracy rate is up to 98%, and false drop rate 3% is in domestic piano plate
Advanced technology level.
Technical scheme: in order to solve the deficiencies in the prior art, a kind of steel ball table based on machine vision of the present invention
Planar defect intelligence annular follows the New Algorithm of inspection, and it is as follows that it processes step:
Step 1, creates the rectangular histogram of gray level image in VS2010 software, creates histogram image according to rectangular histogram, and
Image is carried out histogram equalization process, thus expands the rank of foreground and background gray scale, to reach to strengthen image local pair
Purpose than degree；
Step 2, utilizes the medium filtering improved to eliminate noise.Medium filtering is to be set to by the gray value of each pixel
The intermediate value of all pixel gray values in this some neighborhood window.And the median filter method improved is to containing noisy figure
Centered by it, all choose (M × M) region as upper every bit, in this region, search the intermediate value of gray value, and in district
Every bit in territory all calculates its weights centered by this intermediate value, when certain gray value put is closer to intermediate value, and its weights are the biggest,
Otherwise weights are the least.The assignment giving noise spot in this way is the least, and when cumulative summation, the value of noise spot can be neglected
Slightly disregard, thus can filter a part of noise spot；
Step 3, carries out binary conversion treatment to image, the gray value of the pixel on image is set to 0 or 255, the most just
It is that whole image is presented obvious black and white effect；
Step 4, calls computer vision storehouse, detects the circular edge of steel ball；
Step 5, finds the center of circle of steel ball, calculates the pixel number at distance steel ball edge, the steel ball center of circle；
Step 6, uses the steel ball image that CCD1 imageing sensor collects, and utilizes intelligence annular to follow inspection algorithm and examines
Surveying, it is contemplated that due to illumination reason, the gray value in the annular region of the center of circle to outer ten pixel is relatively big, and not at detection model
Enclose, thus in each border circular areas, centered by the steel ball center of circle, with ten pixels as initial value, rear five pictures of increase every time
Vegetarian refreshments is that radius draws loop truss, until its radius processes more than stopping during steel ball radius.Add up the pixel that each circle sampling arrives
Point, asks for maximum gradation value HDmaxWith minimum gradation value HDminDifference and compare with threshold values f, if more than threshold values f, i.e. HDmax-
HDmin> f, then it is designated as defective steel ball；
Step 7, the image using CCD2 imageing sensor again to collect does step 6 same treatment, if quilt in step 6
It is designated as defective steel ball and is still denoted as defective steel ball in second time detection, then this steel ball is labeled as final defective steel
Ball；
Step 8, drives relevant actuator separately to operate it for final underproof steel ball motor, and without matter
The steel ball of amount problem separately processes, thus has reached to carry out steel ball quality inspection the purpose classified；
As present invention further optimization, in step 6, in the expansion dish of industrial computer, steel ball is at the band of actuator
Do irregular movement under Dong, in 50 milliseconds, circulate picture loop truss can guarantee that the circular arc of detection spreads all over steel ball surface.
As present invention further optimization, read in the algorithm of step 6 is captured by two CCD camera
Image, the resolution of CCD is 1280*1024.
Provide the benefit that: a kind of steel ball surface defect intelligence annular based on machine vision of the present invention follows the new of inspection
Type algorithm, strongly fragrant prior art is compared, and has the advantage that
1, can be effectively improved the precision of steel ball quality inspection, have the highest reliability, experiment proves that, accuracy rate is up to
98%, false drop rate 3%, it is in domestic piano plate advanced technology level.
2, this detection algorithm extends among the research of the quality inspection to other industrial products, for flexible Application digital picture
Treatment technology, has good reference；
3, steel ball plays key as the core component of bearing, the quality of its quality for performance and the safety of bearing
Effect.Use this algorithm can steel ball quality quality effectively be classified, for extending the service life of industrial control equipment, improve
Industrial production efficiency and promotion socialist modernization have positive role.
Accompanying drawing explanation
Fig. 1 intelligence annular follows inspection algorithm flow chart；
Fig. 2 algorithm process mark schematic diagram；
Fig. 3 expansion dish and relevant device scheme of installation；
Fig. 4 is the steel ball image after side figure equalization, the medium filtering of improvement, binary conversion treatment；
Fig. 5 is the steel ball image after the center of circle of steel ball and profile being detected；
Fig. 6 is the steel ball image after intelligence annular follows inspection algorithm process.
Detailed description of the invention
The invention will be further described with example below in conjunction with the accompanying drawings.
As shown in drawings, after by histogram equalization, the medium filtering of improvement and binary conversion treatment, can be clearly
Observe that right 1 steel ball surface exists open defect.Algorithm accurately finds the center of circle of steel ball and the profile of steel ball very much, in terms of this comes
The pixel number of the center of circle distance steel ball profile of steel ball in nomogram picture.After using intelligence annular to follow inspection algorithm, steel ball surface lacks
Fall into and can be the most accurately marked, be hereafter classified under the driving of electric machine actuating mechanism.
Embodiment
A kind of steel ball surface defect intelligence annular based on machine vision of the present invention follows the place of the New Algorithm of inspection
Reason step is as follows:
Step 1, creates the rectangular histogram of gray level image, creates histogram image according to rectangular histogram, and image is carried out Nogata
Figure equalization processing, thus expand the rank of foreground and background gray scale, to reach to strengthen the purpose of Image Warping；
Step 2, utilizes the medium filtering improved to eliminate noise；
Step 3, carries out binary conversion treatment to image, the gray value of the pixel on image is set to 0 or 255, the most just
It is that whole image is presented obvious black and white effect；
Step 4, calls computer vision storehouse, detects the circular edge of steel ball；
Step 5, finds the center of circle of steel ball, calculates the pixel number at distance steel ball edge, the steel ball center of circle；
Step 6, uses the steel ball image that CCD1 imageing sensor collects, and utilizes intelligence annular to follow inspection algorithm and examines
Surveying, it is contemplated that due to illumination reason, the gray value in the annular region of the center of circle to outer ten pixel is relatively big, and not at detection model
Enclose, thus in each border circular areas, centered by the steel ball center of circle, with ten pixels as initial value, rear five pictures of increase every time
Vegetarian refreshments is that radius draws loop truss, until its radius processes more than stopping during steel ball radius.Add up the pixel that each circle sampling arrives
Point, asks for maximum gradation value HDmaxWith minimum gradation value HDminDifference and compare with threshold values f, if more than threshold values f, i.e. HDmax-
HDmin> f, then it is designated as defective steel ball；
Step 7, the image using CCD2 imageing sensor again to collect does step 6 same treatment, if quilt in step 6
It is designated as defective steel ball and is still denoted as defective steel ball in second time detection, then this steel ball is labeled as final defective steel
Ball；
Step 8, drives relevant actuator separately to operate it for final underproof steel ball motor, and without matter
The steel ball of amount problem separately processes, thus has reached to carry out steel ball quality inspection the purpose classified；
The medium filtering of improvement described above, all chooses one to containing the every bit on noisy image centered by it
(M × M) region, searches the intermediate value of gray value in this region, and the every bit in region all calculates centered by this intermediate value
Its weights, when certain gray value put is closer to intermediate value, and its weights are the biggest, otherwise weights are the least.Made an uproar in this way
The assignment of sound point is the least, and when cumulative summation, the value of noise spot is negligible, and thus can filter a part of noise spot,
Thus reached to suppress noise and the good result of protection image detail.Two CCD camera MVC1000SAM_GE30ST2 are with complete
The speed becoming full resolution (1280*1024) is 30fps, 8bit sampling resolution.Two cameras can synchronize continuous acquisition, comprehensive
Acquisition monitoring area image, and use a gigabit network cable transmission, multipotency transmits the image of two cameras, bigger joint simultaneously
Save acquisition controlling equipment, extend the distance of transmission.
Above-described embodiment, only for technology design and the feature of the explanation present invention, its objective is to allow and is familiar with the skill of this technical field
Art personnel will appreciate that present disclosure and implement according to this, can not limit the scope of the invention with this.All bases
Equivalents that spirit of the invention is made or modification, all should contain within protection scope of the present invention.
Claims (6)
1. a steel ball surface defect intelligence annular based on machine vision follows the New Algorithm of inspection, it is characterised in that: include with
Lower step:
Step one: create the rectangular histogram of gray level image in VS2010 software, creates histogram image according to rectangular histogram, and to figure
As carrying out histogram equalization process；
Step 2: utilize the medium filtering improved to eliminate noise；
Step 3: image is carried out binary conversion treatment, is set to 0 or 255 by the gray value of the pixel on image；
Step 4: call computer vision storehouse, detects the circular edge of steel ball；
Step 5, finds the center of circle of steel ball, calculates the pixel number at distance steel ball edge, the steel ball center of circle；
Step 6: collected steel ball image by CCD1 imageing sensor for the first time, is utilized intelligence annular to follow inspection algorithm and carries out
Detection；
Step 7: the image again collected by CCD2 imageing sensor does step 6 same treatment, if being remembered in step 6
In second time detection, still it is denoted as defective steel ball for defective steel ball, then this steel ball is labeled as final defective steel ball；
Step 8: drive relevant actuator separately to operate it for final underproof steel ball motor, and without quality
The steel ball of problem separately processes, thus has reached to carry out steel ball quality inspection the purpose classified.
Steel ball surface defect intelligence annular based on machine vision the most according to claim 1 follows the New Algorithm of inspection, its
It is characterised by: carrying out image, gray balance, improvement medium filtering, image segmentation, binaryzation and contour detecting are a series of
After pretreatment, find six steel ball centers of circle in image, and calculate the steel ball center of circle pixel number to profile the most successively.
A kind of steel ball surface defect intelligence annular based on machine vision the most according to claim 1 follows the novel calculation of inspection
Method, it is characterised in that: in step 6, when detection, centered by the steel ball center of circle, with ten pixels as initial value, the most every time
Increasing by five pixels is that radius circulates picture loop truss in 50 milliseconds, until when its radius is more than steel ball radius at stopping
Reason.
Steel ball surface defect intelligence annular based on machine vision the most according to claim 1 follows the New Algorithm of inspection, its
It is characterised by: in step 6, when detection, adds up each annular and follow the pixel that inspection circle sampling arrives, by maximum gradation value
HDmaxWith minimum gradation value HDminDo difference and compare with threshold values f, if more than threshold values f, i.e. HDmax-HDmin＞ f, then be marked as not
Qualified steel ball.
Steel ball surface defect intelligence annular based on machine vision the most according to claim 1 follows the New Algorithm of inspection, its
Being characterised by: in step 6, in the expansion dish of industrial computer, steel ball does irregular movement under the drive of actuator, 50
In millisecond, circulation picture loop truss can guarantee that the circular arc of detection spreads all over steel ball surface.
Steel ball surface defect intelligence annular based on machine vision the most according to claim 1 follows the New Algorithm of inspection, its
It is characterised by: read in the algorithm of step 6 is the image captured by two ccd image sensor cameras, ccd image
The resolution of sensor is 1280*1024.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610615717.6A CN106053485A (en) | 2016-08-01 | 2016-08-01 | Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610615717.6A CN106053485A (en) | 2016-08-01 | 2016-08-01 | Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106053485A true CN106053485A (en) | 2016-10-26 |
Family
ID=57195949
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610615717.6A Pending CN106053485A (en) | 2016-08-01 | 2016-08-01 | Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106053485A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108279239A (en) * | 2017-12-22 | 2018-07-13 | 中核北方核燃料元件有限公司 | A kind of automatic appearance delection device of spheric fuel element |
| CN111815575A (en) * | 2020-06-19 | 2020-10-23 | 浙江工业大学 | Bearing steel ball part detection method based on machine vision |
| CN111862064A (en) * | 2020-07-28 | 2020-10-30 | 桂林电子科技大学 | Silver wire surface flaw identification method based on deep learning |
| CN113311440A (en) * | 2020-02-26 | 2021-08-27 | 苏州佳世达电通有限公司 | Method for improving ultrasonic image quality and detection system thereof |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101858734A (en) * | 2010-05-19 | 2010-10-13 | 山东明佳包装检测科技有限公司 | Method and device for detecting PET bottleneck quality |
| CN102914545A (en) * | 2012-11-15 | 2013-02-06 | 潍坊学院 | Gear defect detection method and system based on computer vision |
| CN103235939A (en) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | Datum point positioning method based on machine vision |
| 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 |
| CN103606167A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Outer bottle cap profile determining method for defect detection |
| CN103605470A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Circle positioning tool for detecting defect image of bottle cap |
| CN104280406A (en) * | 2014-09-16 | 2015-01-14 | 中国科学院广州能源研究所 | Machine vision system for detecting surface defects of copper part |
| CN104535588A (en) * | 2014-12-24 | 2015-04-22 | 河海大学常州校区 | Egg freshness nondestructive testing system based on Android system and egg freshness nondestructive testing method |
| CN105466953A (en) * | 2015-12-28 | 2016-04-06 | 天津大学 | Steel ball surface defect detecting method based on reorganization of steel ball surface reflection pattern integrity |
| CN105740868A (en) * | 2016-01-26 | 2016-07-06 | 北京工商大学 | Circular operator based image edge extraction method and apparatus |
-
2016
- 2016-08-01 CN CN201610615717.6A patent/CN106053485A/en active Pending
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101858734A (en) * | 2010-05-19 | 2010-10-13 | 山东明佳包装检测科技有限公司 | Method and device for detecting PET bottleneck quality |
| CN102914545A (en) * | 2012-11-15 | 2013-02-06 | 潍坊学院 | Gear defect detection method and system based on computer vision |
| 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 |
| CN103235939A (en) * | 2013-05-08 | 2013-08-07 | 哈尔滨工业大学 | Datum point positioning method based on machine vision |
| CN103606167A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Outer bottle cap profile determining method for defect detection |
| CN103605470A (en) * | 2013-12-04 | 2014-02-26 | 天津普达软件技术有限公司 | Circle positioning tool for detecting defect image of bottle cap |
| CN104280406A (en) * | 2014-09-16 | 2015-01-14 | 中国科学院广州能源研究所 | Machine vision system for detecting surface defects of copper part |
| CN104535588A (en) * | 2014-12-24 | 2015-04-22 | 河海大学常州校区 | Egg freshness nondestructive testing system based on Android system and egg freshness nondestructive testing method |
| CN105466953A (en) * | 2015-12-28 | 2016-04-06 | 天津大学 | Steel ball surface defect detecting method based on reorganization of steel ball surface reflection pattern integrity |
| CN105740868A (en) * | 2016-01-26 | 2016-07-06 | 北京工商大学 | Circular operator based image edge extraction method and apparatus |
Non-Patent Citations (4)
| Title |
|---|
| 丰艳 等: "利用图像像素灰度值变化速度的相似性进行图像分割", 《计算机应用与软件》 * |
| 刘青 等: "基于机器视觉的钢球表面缺陷检测和分类", 《轴承》 * |
| 张燕 等: "基于圆心定位的瓶口三圆周快速缺陷检测算法", 《计算机技术与发展》 * |
| 耿亮: "基于表面特征的钢球分拣技术的研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108279239A (en) * | 2017-12-22 | 2018-07-13 | 中核北方核燃料元件有限公司 | A kind of automatic appearance delection device of spheric fuel element |
| CN113311440A (en) * | 2020-02-26 | 2021-08-27 | 苏州佳世达电通有限公司 | Method for improving ultrasonic image quality and detection system thereof |
| CN111815575A (en) * | 2020-06-19 | 2020-10-23 | 浙江工业大学 | Bearing steel ball part detection method based on machine vision |
| CN111815575B (en) * | 2020-06-19 | 2024-03-26 | 浙江工业大学 | Bearing steel ball part detection method based on machine vision |
| CN111862064A (en) * | 2020-07-28 | 2020-10-30 | 桂林电子科技大学 | Silver wire surface flaw identification method based on deep learning |
| CN111862064B (en) * | 2020-07-28 | 2022-08-09 | 桂林电子科技大学 | Silver wire surface flaw identification method based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102636490B (en) | Method for detecting surface defects of dustproof cover of bearing based on machine vision | |
| CN106053485A (en) | Machine vision-based novel algorithm of intelligent circular inspection of steel ball surface defects | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN110687119A (en) | Defect detection method, system and device and computer readable storage medium | |
| WO2015055060A1 (en) | Online detecting method for continuous casting slab surface quality | |
| CN106802215A (en) | A kind of device for detecting water leakage of water pipe and detection method | |
| CN104574389A (en) | Battery piece chromatism selection control method based on color machine vision | |
| CN210604434U (en) | Online detection device for typical defects of curved glass of mobile phone | |
| CN101221134A (en) | Method and device for detecting tiny bearing surface defect by computer vision technology | |
| CN113325001A (en) | Automatic distinguishing and detecting equipment and method for surface appearance flaws of copper foil | |
| CN112001917A (en) | Machine vision-based geometric tolerance detection method for circular perforated part | |
| CN102175692A (en) | System and method for detecting defects of fabric gray cloth quickly | |
| CN102735688A (en) | Defect detection method | |
| CN102673979A (en) | Method and device for judging deviation of conveying belt | |
| CN110096980A (en) | Character machining identifying system | |
| CN201141839Y (en) | Device for detecting tiny bearing surface defect by computer vision technology | |
| CN110132982A (en) | A kind of automatic optical detection method and equipment of high sensitivity | |
| CN116539620A (en) | On-machine detection method for surface defects of cutter | |
| CN208042989U (en) | A kind of large-scale sheet metal works almost T-stable automatic detection device | |
| CN110021012B (en) | Mobile phone lens window glass defect detection method based on machine vision technology | |
| Chen et al. | Research on detection algorithm of lithium battery surface defects based on embedded machine vision | |
| TW419634B (en) | Automatic detection system and method using bar code positioning | |
| CN105067629B (en) | A kind of chromatic circle resistance open defect and resistance value integrated testing method | |
| CN113808104A (en) | Block-based metal surface defect detection method and system | |
| Song et al. | An improved bearing defect detection algorithm based on Yolo |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20161026 |
|
| RJ01 | Rejection of invention patent application after publication |