CN110082365A - Steel roll rim quality testing machine people based on machine vision - Google Patents
Steel roll rim quality testing machine people based on machine vision Download PDFInfo
- Publication number
- CN110082365A CN110082365A CN201910275490.9A CN201910275490A CN110082365A CN 110082365 A CN110082365 A CN 110082365A CN 201910275490 A CN201910275490 A CN 201910275490A CN 110082365 A CN110082365 A CN 110082365A
- Authority
- CN
- China
- Prior art keywords
- unit
- image
- steel roll
- image processing
- quality testing
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
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
Landscapes
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a kind of steel roll rim quality testing machine people based on machine vision, relates generally to image acquisition units, image processing unit, PC unit, alert process unit.Image acquisition units are made of LED light source, CCD camera, mechanical arm, can quickly adjust different shooting angles according to different operating conditions.Direction adjusts the key connection for using electric machine main shaft and drive shaft, and drive shaft is connected realization with mechanical support arm.Video camera is connect with image processing unit, for acquiring product surface image information in real time, and acquired image information is sent to image processing unit, image processing unit receives the image information that video camera is sent and carries out compression processing, and will treated image information display over the display, for underproof product, alarm can be sounded an alarm.Finally, all data that will test are stored in storage unit, facilitate data management;PC unit includes storage unit, control unit and display unit.
Description
Technical field
The present invention relates to a kind of steel roll rim quality testing machine people, can detect steel roll rim flaw.
Background technique
With increasing for coil of strip output, the speed of artificial detection is far from reaching production requirement, artificial detection quality layers
It is secondary uneven, become and restricts an important factor for steel quality improves.Secondly because steel roll rim apparent surface illumination is insufficient, to inspection
The person of testing proposes more harsh requirement, and detection is harmful to direct labor's health under half-light environment for a long time.So automatic inspection in recent years
Survey technology has significant progress.
Patented technology such as CN208188046U devises product quality detection robot and system, includes one or six axis machines
People, a ultrasonic detection device, a terahertz detection device and a control device, detection system are based on ultrasonic wave and anti-
It penetrates to detect product quality.
In recent years, product testing technology its high efficiency based on machine vision, highly reliable and inexpensive advantage were flown
The development of speed.Machine Vision Inspecting System is applied to many industrial production lines, existing patented technology is such as
CN208341168U describes a kind of product surface detection system based on machine vision, which does product surface detection
It illustrates in detail, which detects primarily directed to product surface quality, and present invention is generally directed to steel roll rim quality testing,
It devises a kind of motor and directly drives steering, mobile detection robot easy for installation.
Summary of the invention:
It is an object of the invention to solve the problems, such as to detect quality of edges on steel plate automatic assembly line, efficiently, accurately
Identify edge defect.
A kind of steel roll rim quality testing machine people based on machine vision of the present invention, including image acquisition units, image
Processing unit, PC unit and alarm unit, described image acquisition unit include LED light source, CCD camera and mechanical arm, institute
State LED light source and the CCD camera be fixed in a screw connection manner on the mechanical arm, the mechanical arm include big support arm and
It is equipped with bradyseism beam among small support arm, the big support arm and the small support arm, the mechanical arm is directly driven by electric machine main shaft
It is dynamic, to shorten response time and simplied system structure;Described image acquisition unit and described image processing unit pass through I/O mouthfuls of companies
Transmission image data is connect, by using median filtering, the residual sound in removal image is made an uproar makes an uproar described image processing unit with light, and
Image dividing processing based on threshold value identifies edge fault with wavelet energy, specifically:
When carrying out the median filtering, using a template containing odd number point, by template center's pixel
Gray value template in the intermediate value of gray value of each point substitute, be equipped with one group of one-dimensional sequence f1, f2, f3..., fn, take institute
Stating template length is n, carries out median filtering, extracts n several f out from one-dimensional sequencei-v..., fi-1, fi, fi+1..., fi+v, wherein
fiCentered on point, v=(n-1)/2, v represents the unidirectional length of sequence, and this n number is sequentially rearranged for new number by size
Column, then take the numerical value y of the serial number central point of new ordered series of numbersiIt is exported as filtering;It is expressed with expression formula below:
yi=Med { fi-v..., fi-1, fi, fi+1..., fi+vI ∈ Z, v=(n-1)/2
Meanwhile the numerical value y of the serial number central point of the stylish ordered series of numbers of two dimension median filterijAre as follows:
yij=Med { fij,
Wherein f1, f2, f3..., fn, represent one-dimensional sequence: Med represents the median of the serial number central point of new ordered series of numbers,
When carrying out the image dividing processing based on threshold value, using preset gray value as threshold value t, image after segmentation
G (x, y) is indicated are as follows:
The energy of image after image processing unit is divided using the detection of wavelet energy method, when wavelet basis function is mutually orthogonal
When, the wavelet energy E under single scalejAre as follows:
J and k is positive integer in expression formula, and j expression is boundary scale, and k is the points of sampling, CjIt (k) is the k of vector
Layer decomposes;
Judged whether according to the height of wavelet energy defective, wherein defect point has high-energy information, and background dot has
Low energy information;The PC unit includes control unit, storage unit and display unit, and the alarm unit includes LED stroboscopic
Lamp and alarm bell, when the PC unit detects that steel roll rim wavelet energy reaches setting value, the alarm unit is given a warning.
Preferably, threshold value is set as a range [t1, t2], in the range [t1, t2] in gray value become 1 or 0,
In the range [t1, t2] outside become 0 or 1.
Preferably, the LED light source uses band light source, and the band light source passes through power supply line and the PC unit phase
Even, its switch and brightness are controlled by the PC unit;The CCD camera is connected by I/O mouthfuls with described image processing unit.
Preferably, it is equipped with sensor on the motor, PC unit described in the data access by acquisition, with real-time monitoring machine
Device people working condition and operating condition.
Preferably, the pedestal of the mechanical arm is equipped with threaded hole, to realize the fixation of described image acquisition unit.
Preferably, the bradyseism beam uses polyurethane bradyseism material, produces vibration noise shadow caused by shooting to weaken
It rings.
The invention has the benefit that
(1) machine vision technique is used, is replaced manually more accelerating to capture steel roll rim defect with industrial camera
Fast efficient detection steel roll rim quality, and can show faulty materials image, and identify flaw location;
(2) robot arm can not only be fixed in the production line, but also can fix independently of production line, among mechanical arm added with
On the one hand bradyseism beam can weaken the vibration bring noise of production line, on the other hand lightization mechanical arm quality;
(3) sensor access industrial PC is housed on motor, real-time monitoring robot working condition and whether can be needed
It repairs;
(4) light source uses LED matrix light source, increases the illumination quality in shooting area, reduces the influence that light is made an uproar in shooting process;
Video camera uses CCD camera, stability and high efficiency, and can adjust shooting speed according to the speed of speed of production.
Detailed description of the invention
Fig. 1 is robot system structure schematic diagram;
Fig. 2 is Image Acquisition structural schematic diagram;
Fig. 3 is the second electric-motor drive unit structural schematic diagram.
Appended drawing reference:
Pedestal 1;First motor 2;First motor rotation axis 3;Drive shaft clamping plate 4;Bearing block 5;Second motor 6;Big support
Arm 7;Big bradyseism beam 8;Third motor 9;4th motor 10;Small support arm 11;Small bradyseism beam 12;5th motor 13;LED light source
14;CCD camera 15;Drive shaft 16;Deep groove ball bearing 17;Electric machine main shaft 18;Keyway 19.
Specific embodiment
The preferred embodiments of the present invention will be described in detail with reference to the accompanying drawing, so that advantages and features of the invention energy
It is easier to be readily appreciated by one skilled in the art, so as to make a clearer definition of the protection scope of the present invention.
It is as shown in Figure 1 a kind of steel plate edge quality testing machine people's system structure diagram based on machine vision.This
Invention includes image acquisition units, image processing unit, power PC unit and alarm unit.Image acquisition units are by LED
Light source 14, the installation of CCD camera 15 are on the robotic arm.
Mechanical arm is directly driven by electric machine main shaft 18, to shorten the response time, increases system flexibility.All motors on
Equipped with sensor, monitoring data are passed to power PC, and the operation conditions of analysis motor and robot can be monitored in real time.
Small bradyseism beam 12, bradyseism are housed equipped with big bradyseism beam 8, the small support arm centre of mechanical arm among the big support arm of mechanical arm
Liang Jun is fastened by bolts on big support arm 6 and small support arm 11.
Big bradyseism beam 8 and small bradyseism beam 12 are all made of polyurethane bradyseism rubber, on the one hand polyurethane bradyseism rubber guarantees figure
Influence as reducing noise on image quality in collection process, on the other hand can reduce weight under the premise of keeping mechanical arm rigidity
Amount.
Big bradyseism beam 8 uses the rectangular parallelepiped structure wide with two big support arms, has threaded hole on big bradyseism beam 8, with big branch
Brace threaded hole is corresponding, is fastened on big support arm with bolt arrangement.
Small 12 structure of bradyseism beam is similar to big bradyseism beam, but in order to increase forearm rigidity, small 12 top of bradyseism beam is cambered
And slightly higher than small support arm portion, length are slightly shorter than small support arm length, reserve rotation space for the 5th motor 13.Small bradyseism beam
12 with small support arm also by being bolted.
LED light source 14 uses the uniform band light source of luminous intensity, and luminous intensity can uniformly be reduced in image acquisition process
The distortion and distortion of image guarantee that Image Acquisition is accurate.
Further, alarm unit is made of LED stroboscopic lamp with alarm bell, and LED stroboscopic lamp flashes when discovery defect is alarmed, alert
Bell ring rises, and only after defect is found by manual confirmation, operator can cancel LED warning light and police by the end PC key
Bell.
Power PC can be divided into display unit, storage unit and control unit, the function one that display unit is mainly realized
Aspect is to show the processed image of video processing board-card in display screen;It on the other hand is by the biography on the parts such as motor
Sensor shows the operating status of robot entirety;The detected defect information of storage unit storage;Control unit control is entire
The operating of robot system, including adjusting mechanical arm angular adjustment, frequency of taking pictures of CCD camera 15 etc..
Further, display unit is high-precision liquid crystal display, and storage unit is power PC internal memory, control
Unit is Keyemce KV-7000 series of PLC control system.
Image processing unit mainly includes video processing board-card, and positioning, denoising and the entropy energy curve of image are all by scheming
As processing card is completed.Image acquisition units, which are connected with image processing unit by I/O, transmits data.
Treated image is through I/O port transmission to power PC display, and power PC display is for showing detection steel
The defect of plate quality of edges.
Specific work process of the present invention is as follows:
The installation of mechanical arm pedestal 1 is in the production line or independently installed.Power PC adjusts camera angle alignment production
Line.When mechanical arm needs to adjust shooting angle, only need to regulate and control motor by power PC makes its rotation.Such as Fig. 3 electric motor units
Shown, stepper motor main shaft 18 is connect with drive shaft 16 by keyway 19, and power PC controls motor and rotates, and 18 turns of electric machine main shaft
Dynamic, so that drive shaft 16 be driven to rotate, drive shaft 16 is connect with big support arm side plate 7, and big support arm 7 is mounted on deep groove ball bearing
On 17, the rotation of drive shaft 16 drives support arm 7 to rotate, and deep groove ball bearing 17 reduces rotary resistance, and drives big support plate
Rotation.
Further, first motor 2, third motor 9, the 4th motor 10, the 5th motor 13 rotating manner described in second
Motor is similar, and the rotation of first motor 2 drives the rotation of the first single machine rotation axis 3, realizes that the mechanical arm in the rotation of horizontal plane, drives
Moving axis clamping plate 4 connects rotation axis 3 and big support arm 7, and bearing block 5 installs rolling bearing.
After adjusting to best angle, CCD camera takes pictures to product of production line continuity, and the photo of shooting is by I/O
Mouth is directly transmitted into image processing unit, directly cannot generally have ideal effect by the image that industrial camera is absorbed
Fruit, can be due to noise, and the reasons such as illumination so that picture quality is lower, therefore is generally filtered or goes to the image directly absorbed
The pretreatment made an uproar etc..Pretreatment can reduce interference etc. caused by the noise jamming and light source problem of original image, so that image
Useless information is removed, useful information is retained, is subsequent image segmentation, feature identification provides good basis.
Image preprocessing should eliminate noise jamming, again the various details that cannot ignore in maintenance image, at image
It manages unit and uses median filtering.For median filtering under the conditions of a certain, it can not only make image become more smooth, but also can
So that image detail is apparent from.
The groundwork step of median filtering is: during template is mobile in figure, center corresponds to single picture
Vegetarian refreshments;Read all grey scale pixel values in template;The gray value of reading is sequentially sorted by size;Find out its intermediate value;By this
One intermediate value is assigned to the pixel of template center, as treated gray value.And because median filtering is not simply to ask equal
Value, obscuring for generating are less.
It further, will be in template by the principle of median filtering it is found that it can only use a template containing odd number point
The intermediate value of the gray value of each point substitutes in the gray value of imago vegetarian refreshments template.Assuming that the template that template is 3 × 3, template
In value be respectively 21,24,34,15,35,23,30,43,17 so the intermediate values be 23.
Further, it is equipped with one group of one-dimensional sequence f1, f2, f3..., fn, taking template length is n, median filtering is carried out, from
Extract n several f in one-dimensional sequence outi-v..., fi-1, fi, fi+1..., fi+v, wherein fiCentered on point, v=(n-1)/2, and by this n
Number sequentially rearranges by size, then takes the numerical value of serial number central point as filtering output;With mathematic(al) representation below
Expression:
yi=Med { fi-v..., fi-1, fi, fi+1..., fi+vI ∈ Z, v=(n-1)/2
Meanwhile two dimension median filter can indicate are as follows:
yij=Med { fij}
Image processing unit of the present invention is to carry out image segmentation to the image after pretreatment.Image segmentation is by a width figure
As being divided into some regions for not overlapping, having its characteristic meaning.The purpose of image segmentation is that the part of defect will be present from overall diagram
It splits as in, individually calculates.Image processing unit of the present invention uses the image segmentation based on threshold value, and Threshold segmentation utilizes figure
The gray feature difference of target point and background dot as in, thinking of the image is the combination for having two class regions of different gray scales, benefit
Corresponding bianry image is generated so that each pixel in figure is able to divide into target point or background dot with a threshold value;
Further, it is assumed that image finds a gray value as threshold value t, then after segmentation by certain method
Image may be expressed as:
Further, threshold value can also be set as a range [t1,t2], as long as gray value within this range becomes 1 or 0,
It is other then on the contrary, become 0 or 1.
Further, the energy of image after image processing unit is divided using the detection of wavelet energy method.Work as wavelet basis function
When being mutually orthogonal, the wavelet transformation conservation of energy can then define the wavelet energy under single scale
Further, j and k is positive integer in the derivation of energy formula formula, and j expression is boundary scale, and k is sampling
Points.
Further, the total energy of signal is
Further, defect point is the information of high-energy, and background dot then belongs to the information of low energy.The information of high-energy
It is a relative concept with low energy information, the affiliated information of background dot is low energy information.In other words, according to the height of wavelet energy
It is low judge whether it is defective, wherein defect point have high-energy information, background dot have low energy information.
Further, alarm unit sounds an alarm, and when manually pressing stop button in control unit, warning device stops report
Alert, to product defects data, computer saves defective data, facilitates data management.
Finally, it should be noted that above-described each embodiment is merely to illustrate technical solution of the present invention, rather than it is limited
System;Although the present invention is described in detail referring to the foregoing embodiments, those skilled in the art should understand that: its
It can still modify to technical solution documented by previous embodiment, or part of or all technical features are carried out
Equivalent replacement;And these modifications or substitutions, technical solution of various embodiments of the present invention that it does not separate the essence of the corresponding technical solution
Range.
Claims (6)
1. a kind of steel roll rim quality testing machine people based on machine vision, including image acquisition units, image processing unit,
PC unit and alarm unit, it is characterised in that:
Described image acquisition unit includes LED light source, CCD camera and mechanical arm, the LED light source and the CCD camera
Be fixed on the mechanical arm in a screw connection manner, the mechanical arm includes big support arm and small support arm, the big support arm with
It is equipped with bradyseism beam among the small support arm, the mechanical arm is directly driven by the main shaft of motor, to shorten the response time;
Described image acquisition unit connect transmission image data, described image processing by I/O mouthfuls with described image processing unit
Unit is by using median filtering, and the residual sound in removal image is made an uproar makes an uproar with light, and the image dividing processing based on threshold value, uses
Wavelet energy identifies edge fault, specifically:
When carrying out the median filtering, using a template containing odd number point, by the ash of template center's pixel
The intermediate value of the gray value of each point substitutes in angle value template, is equipped with one group of one-dimensional sequence f1,f2,f3,…,fn, take the mould
Plate length is n, carries out median filtering, extracts n several f out from one-dimensional sequencei-v,…,fi-1,fi,fi+1,…,fi+v, wherein fiFor
Central point, v=(n-1)/2, v represents the unidirectional length of sequence, and this n number is sequentially rearranged for new ordered series of numbers by size, so
The numerical value y of the serial number central point of new ordered series of numbers is taken afterwardsiIt is exported as filtering;It is expressed with expression formula below:
yi=Med { fi-v..., fi-1, fi, fi+1..., fi+vI ∈ Z, v=(n-1)/2
Meanwhile the numerical value y of the serial number central point of the stylish ordered series of numbers of two dimension median filterijAre as follows:
yij=Med { fij,
Wherein f1,f2,f3,…,fn, represent one-dimensional sequence: Med represents the median of the serial number central point of new ordered series of numbers,
When carrying out the image dividing processing based on threshold value, using preset gray value as threshold value t, image g (x, y) table after segmentation
It is shown as:
The energy of image after image processing unit is divided using the detection of wavelet energy method, it is single when wavelet basis function is mutually orthogonal
Wavelet energy E under one scalejAre as follows:
J and k is positive integer in expression formula, and j expression is boundary scale, and k is the points of sampling, CjIt (k) is the k layer of vector point
Solution;
Judged whether according to the height of wavelet energy it is defective, wherein defect point have high-energy information, background dot have low energy
Measure information;
The PC unit includes control unit, storage unit and display unit, and the alarm unit includes LED stroboscopic lamp and police
Bell, when the PC unit detects that steel roll rim wavelet energy reaches setting value, the alarm unit is given a warning.
2. the steel roll rim quality testing machine people according to claim 1 based on machine vision, it is characterised in that: threshold
Value is set as a range [t1,t2], in the range [t1,t2] in gray value become 1 or 0, in the range [t1,t2] outside
Become 0 or 1.
3. the steel roll rim quality testing machine people according to claim 2 based on machine vision, it is characterised in that: described
LED light source uses band light source, and the band light source is connected by power supply line with the PC unit, controls it by the PC unit
Switch and brightness;The CCD camera is connected by I/O mouthfuls with described image processing unit.
4. the steel roll rim quality testing machine people according to claim 3 based on machine vision, it is characterised in that: described
Sensor is equipped on motor, PC unit described in the data access by acquisition, with real-time monitoring robot working condition and operation
Situation.
5. the steel roll rim quality testing machine people according to claim 4 based on machine vision, it is characterised in that: described
The pedestal of mechanical arm is equipped with threaded hole, to realize the fixation of described image acquisition unit.
6. the steel roll rim quality testing machine people according to claim 5 based on machine vision, it is characterised in that: described
Bradyseism beam uses polyurethane bradyseism material, produces vibration influence of noise caused by shooting to weaken.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910275490.9A CN110082365B (en) | 2019-04-08 | 2019-04-08 | Robot for detecting quality of edge of steel coil based on machine vision |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910275490.9A CN110082365B (en) | 2019-04-08 | 2019-04-08 | Robot for detecting quality of edge of steel coil based on machine vision |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110082365A true CN110082365A (en) | 2019-08-02 |
CN110082365B CN110082365B (en) | 2020-07-03 |
Family
ID=67414414
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910275490.9A Active CN110082365B (en) | 2019-04-08 | 2019-04-08 | Robot for detecting quality of edge of steel coil based on machine vision |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110082365B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113465511A (en) * | 2021-06-19 | 2021-10-01 | 精锐视觉智能科技(上海)有限公司 | Online measurement and omnibearing end surface defect online detection method for steel coil size |
TWI814254B (en) * | 2021-07-01 | 2023-09-01 | 日商東芝三菱電機產業系統股份有限公司 | Data collection device |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103207185A (en) * | 2012-01-11 | 2013-07-17 | 宝山钢铁股份有限公司 | Steel coil end portion quality detection system and method thereof |
CN104535586A (en) * | 2014-11-18 | 2015-04-22 | 华东理工大学 | Strip steel edge defect detection identification method |
US20150185147A1 (en) * | 2012-08-30 | 2015-07-02 | Jason Miller | Apparatus for determining gauge profile for flat rolled material with laser-based lap counter |
CN104777224A (en) * | 2015-04-28 | 2015-07-15 | 华南理工大学 | Defect detecting method for junction surface of metal alloy |
CN207256220U (en) * | 2017-05-10 | 2018-04-20 | 岳佩瑜 | A kind of omnidirectional of substation crusing robot |
CN108698774A (en) * | 2016-02-25 | 2018-10-23 | 康宁股份有限公司 | The method and apparatus that edge surface inspection is carried out to movable glass coiled material |
CN109186912A (en) * | 2018-06-29 | 2019-01-11 | 浙江工业大学 | Mechanical arm damping test device based on camera |
CN109540918A (en) * | 2018-11-28 | 2019-03-29 | 鞍钢集团自动化有限公司 | A kind of hot rolled coil edge fault detection device and method |
CN109564169A (en) * | 2016-07-06 | 2019-04-02 | 蒂阿马公司 | Determine the inspection method, check device and inspection line of the crimping on the inward flange of ring surface |
-
2019
- 2019-04-08 CN CN201910275490.9A patent/CN110082365B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN103207185A (en) * | 2012-01-11 | 2013-07-17 | 宝山钢铁股份有限公司 | Steel coil end portion quality detection system and method thereof |
US20150185147A1 (en) * | 2012-08-30 | 2015-07-02 | Jason Miller | Apparatus for determining gauge profile for flat rolled material with laser-based lap counter |
CN104535586A (en) * | 2014-11-18 | 2015-04-22 | 华东理工大学 | Strip steel edge defect detection identification method |
CN104777224A (en) * | 2015-04-28 | 2015-07-15 | 华南理工大学 | Defect detecting method for junction surface of metal alloy |
CN108698774A (en) * | 2016-02-25 | 2018-10-23 | 康宁股份有限公司 | The method and apparatus that edge surface inspection is carried out to movable glass coiled material |
CN109564169A (en) * | 2016-07-06 | 2019-04-02 | 蒂阿马公司 | Determine the inspection method, check device and inspection line of the crimping on the inward flange of ring surface |
CN207256220U (en) * | 2017-05-10 | 2018-04-20 | 岳佩瑜 | A kind of omnidirectional of substation crusing robot |
CN109186912A (en) * | 2018-06-29 | 2019-01-11 | 浙江工业大学 | Mechanical arm damping test device based on camera |
CN109540918A (en) * | 2018-11-28 | 2019-03-29 | 鞍钢集团自动化有限公司 | A kind of hot rolled coil edge fault detection device and method |
Non-Patent Citations (2)
Title |
---|
俞翔等: "基于局部纹理特征的钢卷侧面缺陷检测方法研究", 《计算机工程与设计》 * |
杨志强: "铝板热轧中间坯边部和头尾缺陷机器视觉检测", 《万方硕士论文库》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113465511A (en) * | 2021-06-19 | 2021-10-01 | 精锐视觉智能科技(上海)有限公司 | Online measurement and omnibearing end surface defect online detection method for steel coil size |
CN113465511B (en) * | 2021-06-19 | 2022-12-13 | 精锐视觉智能科技(上海)有限公司 | Steel coil size online measurement and omnibearing end surface defect online detection method |
TWI814254B (en) * | 2021-07-01 | 2023-09-01 | 日商東芝三菱電機產業系統股份有限公司 | Data collection device |
Also Published As
Publication number | Publication date |
---|---|
CN110082365B (en) | 2020-07-03 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106525873A (en) | Machine vision based full-automatic rotation printed product defect detection device | |
CN103149215B (en) | A kind of toughened glass insulator defect inspection method and device | |
CN202002894U (en) | Quick online paper flaw detecting system based on machine vision | |
CN110082365A (en) | Steel roll rim quality testing machine people based on machine vision | |
CN210038861U (en) | Online detection device for laser drilling of cigarette filter tip | |
CN107443428A (en) | A kind of band visual identity flapping articulation manipulator and visual identity method | |
CN101055256A (en) | Defect inspection apparatus | |
CN109030498A (en) | The vision inspection apparatus and method of a kind of blade knife mark and surface defect | |
CN107132238A (en) | A kind of image collecting device for bearing roller periphery | |
CN217505690U (en) | Film product detection device based on machine vision | |
CN210834708U (en) | Screen circuit detection equipment | |
CN113606437A (en) | All-weather inspection device for computer room | |
CN206609797U (en) | Exploded perspective vision testing machine | |
CN115420757A (en) | Power cable insulation defect detection device and use method | |
CN102183578B (en) | Wheel set fluorescent magnetic powder flaw detector | |
CN208255087U (en) | Roll dressing surface cleaning and Machine Vision Inspecting System | |
CN206146851U (en) | Intellectual detection system industrial part surface quality's device | |
CN108020474A (en) | A kind of quality detecting system for fatigue test specimen | |
CN109813726B (en) | Device for circuit board information integration | |
CN206095970U (en) | Planar defect specular reflection detection device in ABS ring gear | |
CN114119583A (en) | Industrial visual inspection system, method, network model selection method and warp knitting machine | |
CN109632815A (en) | Surface defect on-line detecting system and method | |
CN208547978U (en) | A kind of adhesive label two dimensional code automatic checkout equipment | |
CN111780674A (en) | Visual detection system based on image processing and machine learning | |
CN215931692U (en) | Fabric flaw detection equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |