CN110514664A - A kind of cheese Sha Gan detection and localization robot and method - Google Patents
A kind of cheese Sha Gan detection and localization robot and method Download PDFInfo
- Publication number
- CN110514664A CN110514664A CN201910767593.7A CN201910767593A CN110514664A CN 110514664 A CN110514664 A CN 110514664A CN 201910767593 A CN201910767593 A CN 201910767593A CN 110514664 A CN110514664 A CN 110514664A
- Authority
- CN
- China
- Prior art keywords
- axis
- yarn
- image
- coordinate
- yarn bar
- 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
-
- D—TEXTILES; PAPER
- D06—TREATMENT OF TEXTILES OR THE LIKE; LAUNDERING; FLEXIBLE MATERIALS NOT OTHERWISE PROVIDED FOR
- D06B—TREATING TEXTILE MATERIALS USING LIQUIDS, GASES OR VAPOURS
- D06B23/00—Component parts, details, or accessories of apparatus or machines, specially adapted for the treating of textile materials, not restricted to a particular kind of apparatus, provided for in groups D06B1/00 - D06B21/00
- D06B23/04—Carriers or supports for textile materials to be treated
- D06B23/047—Replacing or removing the core of the package
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01B—MEASURING LENGTH, THICKNESS OR SIMILAR LINEAR DIMENSIONS; MEASURING ANGLES; MEASURING AREAS; MEASURING IRREGULARITIES OF SURFACES OR CONTOURS
- G01B11/00—Measuring arrangements characterised by the use of optical techniques
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01C—MEASURING DISTANCES, LEVELS OR BEARINGS; SURVEYING; NAVIGATION; GYROSCOPIC INSTRUMENTS; PHOTOGRAMMETRY OR VIDEOGRAMMETRY
- G01C9/00—Measuring inclination, e.g. by clinometers, by levels
-
- 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
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Biochemistry (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Life Sciences & Earth Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Signal Processing (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Textile Engineering (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The present invention relates to a kind of cheese Sha Gan detection and localization robot and methods comprising kinematic system, kinematic system and vision system are all mounted on portal-framed structure, and are connect with control system;X-axis bracket, Y-axis bracket and the Z axis bracket and corresponding servo motor for constituting system of 3 axes use linear guide form, are provided with vision system on the side of Z axis frame bottom, vision system includes industrial camera and annular light source;Servo motor on Z axis bracket drives Z axis bracket to move up and down, while industrial camera and annular light source being driven to move up and down together;X-axis bracket and Y-axis bracket are mobile by respective Serve Motor Control vision system left and right horizontal, and vision system is driven to reach the coordinate position of specified yarn bar;Sarong turns round station installation on the ground, and the setting of A bracing strut is at sarong revolution station, using rotation supporting structure, drives carried sarong mobile for carrying standard sarong, and by the Serve Motor Control of A axis.
Description
Technical field
The present invention relates to a kind of field of textile equipment, especially with regard to a kind of cheese Sha Gan detection and localization robot and side
Method.
Background technique
Cheese intelligence dye-works is with the National Award for Science and Technology Progress first prize --- " the automatic dye of cheese digitlization
Color complete set technology and equipment " is technical foundation, by being digitized, information-based, intelligent General Promotion, realize from raw yarn to
The full-range digitlization of dyed yarn finished product and intelligent production;Yarn dyeing is created to match automatically from winder, scheduled production, logistics system
It closes, production task executes automatically, special purpose robot's substitution is artificial, dyes whole process parameter on-line checking and feedback, the quality of production
Controllably, the whole processes such as novel maintenance yarn dyeing intelligence manufacture new model.
A procedure in dyeing course is dress yarn, i.e., spindle is attached on the yarn bar in sarong.It is continuous with sarong
It uses, yarn bar will appear normal deviation skew, when crooked degree is excessive, will affect being automatically loaded for spindle.Yarn bar deviates
Distance is bigger, and dress yarn and the time for unloading yarn extend, and accuracy will reduce, and single sarong, which goes wrong, also will affect subsequent workshop work
The progress of work is also possible to will cause safety accident if serious.When crooked degree is more than a certain amount of, need to rectify yarn bar
Just.Currently, the correction work of yarn bar is also all by being accomplished manually, worker needs the data to each bar to measure record,
It is obtained further according to measurement data comparison and deviates more yarn bar bar number, be adjusted correspondingly, this original bearing calibration, surveyed
Amount process is slow, low efficiency, and dependence surveyor's subjective factor ingredient is excessive, and the accuracy rate of correction cannot be guaranteed.
Summary of the invention
In view of the above-mentioned problems, the object of the present invention is to provide a kind of cheese Sha Gan detection and localization robot and method,
Energy automatic measurement current yarn bar position judges whether current yarn bar needs to correct, and measurement is very fast, and high-efficient.
To achieve the above object, the present invention takes following technical scheme: a kind of cheese Sha Gan detection and localization robot,
Including portal-framed structure, kinematic system, vision system and control system;The kinematic system and vision system are all mounted on the door
In frame structure, and it is connect with the control system;Acquired image information is transmitted to the control by the vision system
System, the control system control the kinematic system according to the image information received and act;The kinematic system includes X-axis branch
Frame, Y-axis bracket, Z axis bracket and A bracing strut, and the servo motor being separately positioned on each bracing strut, each servo motor
It connect, is acted by each servo motor of control system control, and then drive corresponding bracing strut with the control system
It is moved;The X-axis bracket, Y-axis bracket and the Z axis bracket and corresponding servo motor for constituting system of 3 axes are led using straight line
Rail form, is provided with the vision system on the side of the Z axis frame bottom, the vision system include industrial camera and
The side of the Z axis pedestal lower end is arranged in annular light source, the industrial camera, is arranged positioned at the industrial camera lower part
There is the annular light source;Servo motor on the Z axis bracket drives the Z axis bracket to move up and down, while driving the work
Industry video camera and annular light source move up and down together;The X-axis bracket and Y-axis bracket pass through described in respective Serve Motor Control
Vision system left and right horizontal is mobile, and the vision system is driven to reach the coordinate position of specified yarn bar;Sarong turns round station installation
On the ground, the A bracing strut setting is at sarong revolution station, using rotation supporting structure, for carrying standard yarn
Cage, and drive carried sarong mobile by the Serve Motor Control of A axis.
Further, the end of the Z axis pedestal lower end is provided with cheese automatic loading and unloading paw mechanism.
Further, the industrial camera uses Daheng's image Mercury series GigE digital camera;The yarn club head
Distance of camera lens with the industrial camera is 200~300mm;The industrial camera is parallel with the Z axis bracket to be installed, angle
Should be not more than by spending error by 1 °.
Further, the control system includes controller, X-axis driver, Y-axis driver, Z axis driver, A axis driver
And power supply, the controller is by the power supply power supply;The controller receives the vision system through industrial switch and is transmitted to
Image information, control instruction will be converted to after the Image Information Processing received, is transmitted separately to the X-axis through data/address bus
Driver, Y-axis driver, Z axis driver and A axis driver drive the servo motor of corresponding axis to act by each driver.
Further, the controller uses Siemens S7-1217 controller, and the data/address bus is total using ProfiNET
Line.
A kind of cheese yarn bar position finding and detection method based on above-mentioned robot comprising following steps: 1) by industry taken the photograph
After camera carries out Image Acquisition to each yarn bar, image information is transmitted to controller, image procossing is carried out by controller;2)
Controller carries out image procossing according to the image information received, and carries out vision positioning calibration to each yarn bar, will be industrial
The centering reference coordinate of each yarn bar of video camera is calibrated and is recorded;3) to the calibrated yarn bar for recording original centering coordinate into
Row vision-based detection: being issued the instruction of the detection new yarn bar of request detection by vision system, to the yarn of vision system write-in current request
Bar number;Whether real-time judge present bit replacement yarn-rod number is identical, the instruction for continuing request is issued when difference in position, in position
It is moved to when identical above the yarn bar of vision system request and returns to current yarn bar number to show in place;Vision system, which is read, works as
Preceding yarn bar number is detected when being equal to request number, and records detection coordinate as a result, completing yarn bar positioning vision-based detection.
Further, in the step 1), image procossing passes through the following steps are included: 1.1) progress industrial camera calibration
Standard round is directed at industrial camera center, calculates pixel value and standard circular diameter corresponding relationship;1.2) obtain includes yarn to be detected
The acquired original image of club head image;1.3) automatic threshold segmentation is carried out to acquired original image, and extracts connected domain, obtained
The connection area image of yarn club head to be detected;1.4) mass center extraction is carried out to yarn club head connected domain to be detected, obtaining mass center is
For the center position coordinates under yarn club head image coordinate system;1.5) coordinate system conversion is carried out, converts reality for image coordinate system
Coordinate under the coordinate system of border, by with home position coordinate pair ratio, obtain actual shifts size;1.6) defect estimation is carried out, it is right
Obtained connection area image carries out signature analysis, calculates and obtains area, shape feature, area, shape with standard yarn club head
Feature compares, and then judges that yarn club head with the presence or absence of defect, defect level is assessed by area ratio, provides this
The confidence level of yarn bar coordinate.
Further, in the step 1.5), coordinate system is converted the following steps are included: 1.5.1) use a known dimensions
Standard round is installed on yarn bar top, guarantees with yarn bar top in same level;1.5.2 it) is acquired using camera calibration method
Acquisition standard circular image carries out Hough transformation and obtains standard radius of circle Pixel Dimensions;1.5.3) with the standard round got half
Diameter Pixel Dimensions and actual size carry out ratio calculation, obtain calibration scale bar;It 1.5.4) will be under yarn club head image coordinate system
Center position coordinates be multiplied with scale bar, obtain final actual position coordinate.
Further, in the step 1.6), defect estimation is the following steps are included: 1.6.1) to the connection area image got
Area screening is carried out, the connected domain in yarn bar values extracts;1.6.2) analysis connected domain centroid position and figure
Whether the distance of inconocenter position judges the area of the connected domain in the pixel of 20, range image center in range;1.6.3)
If differed in 50 pixel coverages with default, judge that circularity, rectangular degree carry out analyzing defect degree;1.6.4) if with default
Difference judges whether distance-taxis differs in the connected domain center of second and third at 10 not in 50 pixel coverages
In pixel, expansion process is carried out if meeting, and with two connected domains of connection, then carries out step 1.6.2) processing.
Further, in the step 2), 2.1) localization method is the following steps are included: issue Location Request letter to vision system
Number and update identification serial number, carry out visual identity;2.2) industrial camera moves in advance after receiving positioning request signal
The primary standard of setting is directed at position, carries out identification of taking pictures, and the coordinate result of identification is fed back to controller, identifies serial number
Refresh together with location data, the recognition result fed back to is compared controller with the serial number of request;2.3) according to than
To modified result centering coordinate, and judge the deviation of the yarn bar coordinate of the location of current industrial video camera coordinate and identification
Whether presetting in allowed band, be to move to next yarn bar coordinate to be positioned to continue to position, otherwise industry is taken the photograph
Camera then moves to correction position, continues to calculate deviation, the reference coordinate of circulation change always, until deviation is in the permission model of setting
In enclosing;When feedback identifying result is consistent with the serial number of request, it will be considered that vision system has been completed positioning work sum number
According to processing, the X/Y axis deviation in data field is authentic and valid.
The invention adopts the above technical scheme, which has the following advantages: the present invention can be automatically on cheese sarong
Yarn club head carry out detection and localization, be compared with normal place, thus realize automatic measurement current yarn bar position, judgement work as
Whether preceding yarn bar needs to correct, and solves to measure using manpower and judges that bring waste of human resource and accuracy be not high
Problem;The plant automation degree is high simultaneously, greatly accelerates the paces of the intelligent modernization of entire cheese dyeing industry.
Detailed description of the invention
Fig. 1 is overall structure diagram of the invention;
Fig. 2 is control system hardware structural diagram of the invention;
Fig. 3 is localization method flow diagram of the invention;
Fig. 4 is detection method flow diagram of the invention.
Specific embodiment
The present invention is described in detail below with reference to the accompanying drawings and embodiments.
As shown in Figure 1, the present invention provides a kind of cheese Sha Gan detection and localization robot comprising portal-framed structure, movement
System, vision system and control system.Kinematic system and vision system are all mounted on portal-framed structure, and are connected with control system
It connects;Acquired image information is transmitted to control system by vision system, and control system is controlled according to the image information received
Kinematic system movement.Wherein, portal-framed structure includes four columns 1 and four crossbeams 2, and it is vertical that adjacent two are arranged in every crossbeam 2
Between 1 top of column, and it is located at the supporting beam 3 that crossbeam 2 is provided with reinforcement with 1 junction of column.
Kinematic system includes X-axis bracket 4, Y-axis bracket 5, Z axis bracket 6 and A bracing strut 7, and is separately positioned on each axis branch
Servo motor on frame, each servo motor are connect with control system, are controlled each servo motor by control system and are acted, and then drive
Corresponding bracing strut is moved to be moved.X-axis Y-axis uses conventional, prior art, drives linear guide with motor synchronous drive, Z axis is adopted
It is driven with the screw structure of the prior art.A axis is rotated by the engagement of existing traditional technology motor gear.
X-axis bracket 4, Y-axis bracket 5 and the Z axis bracket 6 and corresponding servo motor for constituting system of 3 axes use the prior art
In linear guide form, synchronize movement using respective servo motor driving pulley, details are not described herein, and it specific is tied
Structure and motion principle.It is provided with vision system on the side of 6 bottom of Z axis bracket, can effectively measure sarong yarn bar top end surface figure
As information, vision system includes industrial camera and annular light source, and industrial camera is arranged in the side of 6 lower end of Z axis bracket, adopts
It is fixed with angle bar screw, is arranged with annular light source positioned at industrial camera lower part, annular light source is fixed by screws in industry and takes the photograph
On camera;Servo motor on Z axis bracket 6 drives Z axis bracket 6 to move up and down, while driving industrial camera and annular light source
It moves up and down together, to adjust industrial camera at a distance from yarn club head.X-axis bracket 4 and Y-axis bracket 5 are watched by respective
It is mobile to take motor control vision system left and right horizontal, vision system can be driven to reach the coordinate position of specified yarn bar, to carry out
Detection and localization operation is imaged.Sarong turn round station installation on the ground, A bracing strut 7 setting sarong revolution station at, using return
Turn support construction, drives carried sarong mobile for carrying standard sarong, and by the Serve Motor Control of A axis.
In above-described embodiment, the end of 6 lower end of Z axis bracket is provided with cheese automatic loading and unloading paw mechanism.
In above-described embodiment, it is a industry that industrial camera, which uses Daheng's image Mercury series GigE digital camera,
Camera, 8mm focal length.The distance of camera lens of yarn club head and industrial camera is 200~300mm;Industrial camera and Z axis bracket 6
Parallel installation, to guarantee that industrial camera visual field coordinate system is overlapped with Z axis kinetic coordinate system, angular error should be not more than 1 °.
In above-described embodiment, annular light source uses annular LED adjustable brightness light source, for illuminating yarn club head, virtualization
Background.
As shown in Fig. 2, control system includes controller, X-axis driver, Y-axis driver, Z axis driver, A axis driver
And power supply, controller is by power supply power supply.Controller receives the image information that vision system is transmitted to through industrial switch, will receive
To Image Information Processing after be converted to control instruction, be transmitted separately to X-axis driver, Y-axis driver, Z axis through data/address bus
Driver and A axis driver drive the servo motor of corresponding axis to act by each driver.
In above-described embodiment, control system further includes graphic control panel, and graphic control panel is through industrial switch and controller
Carry out information exchange;External command information is inputted into controller by the graphic control panel.
In above-described embodiment, controller uses Siemens S7-1217 controller, and the PLC of Siemens S7-1200 series is controlled
The tasks such as the achievable simple logic control of device, higher-order logic control, HMI and network communication.Data/address bus is total using ProfiNET
Line realizes the real-time control to servo motor.
In above-described embodiment, is all connected, adopted by Industrial Ethernet between controller and vision system and graphic control panel
Man-machine information interaction and communication are realized with S7 communications protocol.
Based on above-mentioned robot, the present invention also provides a kind of cheese yarn bar position finding and detection methods comprising following steps:
1) Image Acquisition is carried out to each yarn bar by industrial camera, image information is transmitted to controller, by controlling
Device carries out image procossing;
Image procossing the following steps are included:
1.1) industrial camera calibration is carried out, industrial camera center is directed at by standard round, calculates pixel value and standard
Circular diameter corresponding relationship;
1.2) the acquired original image comprising yarn club head image to be detected is obtained;
1.3) automatic threshold segmentation is carried out to acquired original image, and extracts connected domain, obtain yarn club head to be detected
It is connected to area image;
Wherein, it is screened by characteristic conditions such as area, circularity, rectangular degrees, obtains the connected domain for meeting the condition of presetting
Connected domain as yarn club head to be detected;Wherein circularity is the area of connected domain and the ratio of its minimum circumscribed circle area, square
Shape degree is the area of connected domain and the ratio of minimum circumscribed rectangle area.
1.4) mass center extraction is carried out to yarn club head connected domain to be detected, obtaining mass center is yarn club head image coordinate system
Under center position coordinates;
Wherein, mass center extraction is carried out to yarn club head connected domain to be detected, can directly carries out connected domain analysis and extracts matter
The heart can also be extracted by Hough transformation in the acquired original image in step 1.2) or the connection area image in step 1.3) and be justified
Profile seeks the center of circle of circle contour;
1.5) carry out coordinate system conversion, the coordinate converted image coordinate system under actual coordinates, by with raw bits
Coordinate pair ratio is set, actual shifts size is obtained.
1.6) defect estimation is carried out, signature analysis is carried out to obtained connection area image, calculates and obtains the spies such as area, shape
Sign, compares with features such as area, the shapes of standard yarn club head, and then judges yarn club head with the presence or absence of defect, to scarce
The degree of falling into is assessed by area ratio, provides the confidence level of the yarn bar coordinate.
Above-mentioned steps 1.3) in, obtain the connection area image of yarn club head to be detected using image segmentation algorithm, including with
Lower step:
1.3.1 filtering method (being not limited to mean filter)) is used to acquired original image, then degree of comparing stretches;
1.3.2 adaptive threshold fuzziness method) is used to pretreated image, obtains the figure with head candidate region
Picture;Wherein, adaptive threshold fuzziness uses but is not limited to OTSU Da-Jin algorithm, can also be using other methods such as Two-peak methods;
1.3.3 connected domain extraction) is carried out to the candidate region image of acquisition, and calculates each candidate regions in set of candidate regions
The size in domain, location information;
Above-mentioned steps 1.5) in, coordinate system conversion the following steps are included:
1.5.1 yarn bar top) is installed on using the standard round of a known dimensions, is guaranteed with yarn bar top in same level
In face;
1.5.2 acquisition standard circular image) is acquired using camera calibration method, Hough transformation is carried out and obtains standard radius of circle picture
Plain size;
1.5.3 ratio calculation) is carried out with the standard radius of circle Pixel Dimensions and actual size that get, obtains marked ratio
Example ruler;
1.5.4 the center position coordinates in step 1.4) are multiplied with scale bar), obtain final actual position coordinate.
Above-mentioned steps 1.6) in, defect estimation the following steps are included:
1.6.1 area screening) is carried out to the connection area image got, the connected domain in yarn bar values carries out
It extracts;
1.6.2) analysis connected domain centroid position judges 20, range image center pixel at a distance from image center location
Whether the area of interior connected domain is in range;
1.6.3) if differed in 50 pixel coverages with default, judge that circularity, rectangular degree carry out analyzing defect degree,
Middle circularity is the area of connected domain and the ratio of its minimum circumscribed circle area, and rectangular degree is the area and minimum external square of connected domain
The ratio of shape area.
1.6.4) if differed not in 50 pixel coverages with default, judge that distance-taxis is connected to second with third
Whether domain center differs in 10 pixels, and expansion process is carried out if meeting, and with two connected domains of connection, then carries out
Step 1.6.2) processing.
2) controller carries out image procossing according to the image information received, and carries out vision positioning school to each yarn bar
The centering reference coordinate for each yarn bar that industrial camera takes is calibrated and is recorded by standard;
As shown in figure 3, localization method the following steps are included:
2.1) positioning request signal is issued to industrial camera and update identification serial number, carry out visual identity;
2.2) industrial camera moves to preset primary standard alignment position after receiving positioning request signal, into
Capable identification of taking pictures, and the coordinate result of identification is fed back into controller, identification serial number refreshes together with location data, controller
The recognition result fed back to is compared with the serial number of request;
2.3) according to comparison result amendment centering coordinate, (centering coordinate is the position for being directed at industrial camera picture central point
The reference coordinate that coordinate, i.e. this coordinate can be compared as the error of final yarn bar), and judge position locating for current industrial video camera
Whether the deviation for setting the yarn bar coordinate of coordinate and identification (is set as presetting in allowed band in the present embodiment
It 0.1mm), is to move to next yarn bar coordinate to be positioned to continue to position, otherwise industrial camera then moves to amendment position
It sets, continues to calculate deviation, the reference coordinate of circulation change always, until deviation is in the allowed band of setting;
When feedback identifying result is consistent with the serial number of request, it will be considered that vision system have been completed positioning work and
Data processing, the X/Y axis deviation in data field is authentic and valid.
3) vision-based detection is carried out to the calibrated yarn bar for recording original centering coordinate: as shown in figure 4, being issued by vision system
The instruction for detecting the new yarn bar of request detection, to the yarn bar number of vision system write-in current request;Real-time judge current location yarn
Whether bar number is identical as the yarn bar number of request detection, issues the instruction for continuing request when difference in position, identical in position
When move to above the yarn bar of vision system request and to return to current yarn bar number in place to show.Vision system reads current yarn
Bar number is detected when being equal to request number, and records detection coordinate as a result, completing yarn bar positioning vision-based detection.
In above steps, dyeing and finishing workshop sarong, yarn bar are all made of standardized production, there is 120 yarns on each sarong
Bar performs label in order, and yarn bar top plan circular diameter 7mm has diameter 5mm threaded hole, and material is stainless steel reflective surface.Yarn
Bar bottom end is used to be auxiliarily fixed on the disk of cheese to be provided with wavy annulus, and diffusing reflection can occur for light irradiation, causes
Background interference, and it is reflective than more serious.By adjusting annular light source brightness, solves the problems, such as background reflection interference, make background blurring
Dark treatment, top yarn bar brightness reflection enhancement.
In above steps, vision system needs once to be demarcated before the use as a kind of measurer.Vision camera
It cannot be guaranteed completely the same with the last time when needing to reinstall for some reason, each sarong yarn bar overhead height also has certain
Deviation needs to demarcate pixel/distance scale bar before each is used.
To sum up, the present invention controls ambient brightness, the figure acquired in real time with high resolution industrial video camera using annular light source
As being chief source of information, yarn bar is positioned and analyzed, realizes the online automatic detection of resultant yarn bar positioning.Due to knowing in image
It is other the result is that pixel deviations value on X/Y axis, calibration process will convert thereof into actual deviation value, mounting height difference phase
The actual deviation value that same pixel deviations value represents is different, but every camera, each sarong cannot be completely secured and pacify every time
The difference in height of camera and sarong is constant when dress.Therefore a changeless scale is needed to carry out the scale bar of uncalibrated image.It uses
For one standard round as scale, it is concordant with yarn bar that the standard round of diameter 50mm is inserted in top surface at the top of yarn bar.Pass through examination criteria circle
Diameter (pixel size) in the camera demarcates pixel/distance scale bar size, vision positioning calibration and detection journey
The data measured in sequence use this scale bar to convert.
Because respectively there are the coordinate system of its own in industrial camera and robot, so there are two sets of coordinate systems, and Two coordinate system
Between there are deviations.Before vision positioning calibration, Two coordinate system need to be calibrated.The method of coordinate origin calibration is more multiple
It is miscellaneous, and do not need to know the position of origin in experimentation.Therefore, when positioning calibration, the deviation of current yarn bar is provided
Value is stored in controller, and makes comparing calculation with robot coordinate system's (i.e. coordinate system of xyz axis), enable changing coordinates be (0,
0), as a reference value, i.e., new origin.When vision-based detection, this reference value coordinate is called, exports the detected value of result i.e.
For the deviation of yarn bar top central coordinate of circle.
The various embodiments described above are merely to illustrate the present invention, and structure and size, setting position and the step of each component are all can be with
It is varied, based on the technical solution of the present invention, all principles according to the present invention change individual part and step
Into and equivalents, should not exclude except protection scope of the present invention.
Claims (10)
1. a kind of cheese Sha Gan detection and localization robot, it is characterised in that: including portal-framed structure, kinematic system, vision system
And control system;The kinematic system and vision system are all mounted on the portal-framed structure, and are connect with the control system;
Acquired image information is transmitted to the control system by the vision system, and the control system is according to the image received
Information controls the kinematic system movement;
The kinematic system includes X-axis bracket, Y-axis bracket, Z axis bracket and A bracing strut, and is separately positioned on each bracing strut
Servo motor, each servo motor connect with the control system, by each servo electricity of control system control
Motor-driven work, and then corresponding bracing strut is driven to be moved;
The X-axis bracket, Y-axis bracket and the Z axis bracket and corresponding servo motor for constituting system of 3 axes use linear guide shape
Formula is provided with the vision system on the side of the Z axis frame bottom, and the vision system includes industrial camera and annular
The side of the Z axis pedestal lower end is arranged in light source, the industrial camera, is arranged positioned at the industrial camera lower part
State annular light source;Servo motor on the Z axis bracket drives the Z axis bracket to move up and down, while the industry being driven to take the photograph
Camera and annular light source move up and down together;The X-axis bracket and Y-axis bracket pass through vision described in respective Serve Motor Control
System left and right horizontal is mobile, and the vision system is driven to reach the coordinate position of specified yarn bar;Sarong revolution station is mounted on ground
On face, the A bracing strut setting is at sarong revolution station, using rotation supporting structure, for carrying standard sarong, and
Drive carried sarong mobile by the Serve Motor Control of A axis.
2. robot as described in claim 1, it is characterised in that: it is automatic that the end of the Z axis pedestal lower end is provided with cheese
Load and unload paw mechanism.
3. robot as described in claim 1, it is characterised in that: the industrial camera uses Daheng's image Mercury series GigE
Digital camera;The distance of camera lens of the yarn club head and the industrial camera is 200~300mm;The industrial camera
Parallel with the Z axis bracket to install, angular error should be not more than 1 °.
4. robot as described in claim 1, it is characterised in that: the control system includes controller, X-axis driver, Y-axis drive
Dynamic device, Z axis driver, A axis driver and power supply, the controller is by the power supply power supply;The controller is exchanged through industry
Machine receives the image information that the vision system is transmitted to, and will be converted to control instruction after the Image Information Processing received, warp
Data/address bus is transmitted separately to the X-axis driver, Y-axis driver, Z axis driver and A axis driver, is driven by each driver
The servo motor movement of corresponding axis.
5. robot as claimed in claim 4, it is characterised in that: the controller uses Siemens S7-1217 controller, described
Data/address bus uses ProfiNET bus.
6. a kind of cheese yarn bar position finding and detection method based on the robot as described in any one of claim 1 to 5, feature exist
In the following steps are included:
1) after carrying out Image Acquisition to each yarn bar by industrial camera, image information is transmitted to controller, by controller
Carry out image procossing;
2) controller carries out image procossing according to the image information received, and carries out vision positioning calibration to each yarn bar,
The centering reference coordinate of each yarn bar of industrial camera is calibrated and recorded;
3) carry out vision-based detection to the calibrated yarn bar for recording original centering coordinate: it is new to issue detection request detection by vision system
The instruction of yarn bar, to the yarn bar number of vision system write-in current request;Whether real-time judge present bit replacement yarn-rod number is identical,
The instruction for continuing request is issued when difference in position, moved to when position is identical above the yarn bar of vision system request and is returned
Current yarn bar number is in place to show;Vision system reads when current yarn bar number is equal to request number and is detected, and records
Coordinate is detected as a result, completing yarn bar positions vision-based detection.
7. method as claimed in claim 6, it is characterised in that: in the step 1), image procossing the following steps are included:
1.1) industrial camera calibration is carried out, industrial camera center is directed at by standard round, pixel value is calculated and standard round is straight
Diameter corresponding relationship;
1.2) the acquired original image comprising yarn club head image to be detected is obtained;
1.3) automatic threshold segmentation is carried out to acquired original image, and extracts connected domain, obtain the connection of yarn club head to be detected
Area image;
1.4) mass center extraction is carried out to yarn club head connected domain to be detected, obtaining mass center is under yarn club head image coordinate system
Center position coordinates;
1.5) carry out coordinate system conversion, coordinate image coordinate system converted under actual coordinates, by sitting with home position
Mark comparison, obtains actual shifts size;
1.6) defect estimation is carried out, signature analysis is carried out to obtained connection area image, calculates and obtains area, shape feature, with
Area, the shape feature of standard yarn club head compare, and then judge that yarn club head with the presence or absence of defect, leads to defect level
It crosses area ratio to be assessed, provides the confidence level of the yarn bar coordinate.
8. method as claimed in claim 7, it is characterised in that: in the step 1.5), coordinate system conversion the following steps are included:
1.5.1 yarn bar top) is installed on using the standard round of a known dimensions, is guaranteed with yarn bar top in same level
It is interior;
1.5.2 acquisition standard circular image) is acquired using camera calibration method, Hough transformation is carried out and obtains standard radius of circle pixel ruler
It is very little;
1.5.3 ratio calculation) is carried out with the standard radius of circle Pixel Dimensions and actual size that get, obtains calibration ratio
Ruler;
1.5.4 the center position coordinates under yarn club head image coordinate system are multiplied with scale bar), final physical location is obtained and sits
Mark.
9. method as claimed in claim 7, it is characterised in that: in the step 1.6), defect estimation the following steps are included:
1.6.1 area screening) is carried out to the connection area image got, the connected domain in yarn bar values is mentioned
It takes;
1.6.2) analysis connected domain centroid position judges in the pixel of 20, range image center at a distance from image center location
Whether the area of connected domain is in range;
1.6.3) if differed in 50 pixel coverages with default, judge that circularity, rectangular degree carry out analyzing defect degree;
1.6.4) if with it is default differ in 50 pixel coverages, do not judge distance-taxis second and third connected domain in
Whether heart position differs in 10 pixels, and expansion process is carried out if meeting, and with two connected domains of connection, then carries out step
1.6.2 it) handles.
10. such as any one of claim 6 to 9 the method, it is characterised in that: in the step 2), localization method includes following
Step:
2.1) positioning request signal is issued to vision system and update identification serial number, carry out visual identity;
2.2) industrial camera moves to preset primary standard alignment position after receiving positioning request signal, is clapped
Controller is fed back to according to identification, and by the coordinate result of identification, identification serial number refreshes together with location data, and controller will be anti-
The recognition result being fed back to is compared with the serial number of request;
2.3) centering coordinate is corrected according to comparison result, and judges the yarn of the location of current industrial video camera coordinate and identification
Whether the deviation of bar coordinate is presetting in allowed band, is to move to next yarn bar coordinate to be positioned to continue to determine
Position, on the contrary industrial camera then moves to correction position, continues to calculate deviation, the reference coordinate of circulation change always, until deviation
In the allowed band of setting;
When feedback identifying result is consistent with the serial number of request, it will be considered that vision system has been completed positioning work and data
It handles, the X/Y axis deviation in data field is authentic and valid.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910767593.7A CN110514664B (en) | 2019-08-20 | 2019-08-20 | Cheese yarn rod positioning and detecting robot and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910767593.7A CN110514664B (en) | 2019-08-20 | 2019-08-20 | Cheese yarn rod positioning and detecting robot and method |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110514664A true CN110514664A (en) | 2019-11-29 |
CN110514664B CN110514664B (en) | 2022-08-12 |
Family
ID=68626657
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910767593.7A Active CN110514664B (en) | 2019-08-20 | 2019-08-20 | Cheese yarn rod positioning and detecting robot and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110514664B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110961289A (en) * | 2019-12-09 | 2020-04-07 | 国网智能科技股份有限公司 | Transformer substation insulator anti-pollution flashover coating spraying tool and spraying method |
CN110992358A (en) * | 2019-12-18 | 2020-04-10 | 北京机科国创轻量化科学研究院有限公司 | Method and device for positioning yarn rods of yarn cage, storage medium and processor |
CN117864871A (en) * | 2024-03-05 | 2024-04-12 | 北自所(北京)科技发展股份有限公司 | Automatic loading and positioning system of glass fiber doffer and operation method |
Citations (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030081215A1 (en) * | 2001-01-09 | 2003-05-01 | Ajay Kumar | Defect detection system for quality assurance using automated visual inspection |
US20080289125A1 (en) * | 2006-08-24 | 2008-11-27 | Mario Scatizzi | Device and Method for Carrying out Optical Readings on Textile Materials Submitted to Dyeing |
CN102706274A (en) * | 2012-04-25 | 2012-10-03 | 复旦大学 | System for accurately positioning mechanical part by machine vision in industrially-structured scene |
CN103197599A (en) * | 2013-03-25 | 2013-07-10 | 东华大学 | System and method for numerical control (NC) workbench error self correction based on machine vision |
CN103406905A (en) * | 2013-08-20 | 2013-11-27 | 西北工业大学 | Robot system with visual servo and detection functions |
CN103869186A (en) * | 2014-03-05 | 2014-06-18 | 随州泰华电子科技有限公司 | Tester for quartz crystal wafer |
US20140184786A1 (en) * | 2013-01-02 | 2014-07-03 | The Boeing Company | Systems and Methods for Stand-Off Inspection of Aircraft Structures |
CN104609175A (en) * | 2014-12-05 | 2015-05-13 | 机械科学研究总院先进制造技术研究中心 | Gripping device and method for high-level sarong to load and unload yarn rolls |
WO2015120734A1 (en) * | 2014-02-17 | 2015-08-20 | 华南理工大学 | Special testing device and method for correcting welding track based on machine vision |
CN105180905A (en) * | 2015-07-23 | 2015-12-23 | 陕西科技大学 | Double-camera visual positioning system and method |
CN206187342U (en) * | 2016-11-24 | 2017-05-24 | 西南科技大学 | Novel four rotor motor fixing structure |
CN107014818A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of laser carving defective vision detecting system and method |
CN107102009A (en) * | 2017-05-04 | 2017-08-29 | 武汉理工大学 | A kind of method of the cylinder spool quality testing based on machine vision |
CN107742289A (en) * | 2017-10-15 | 2018-02-27 | 哈尔滨理工大学 | One kind is based on machine vision revolving body workpieces detection method |
CN207123496U (en) * | 2017-06-22 | 2018-03-20 | 惠州市德赛电池有限公司 | A kind of battery plus-negative plate identification equipment |
CN108476282A (en) * | 2016-01-22 | 2018-08-31 | 富士胶片株式会社 | Radiography assisting device and photography householder method |
CN207806014U (en) * | 2017-11-17 | 2018-09-04 | 青岛杰瑞工控技术有限公司 | Spool sorting system based on machine vision |
CN108802040A (en) * | 2017-05-04 | 2018-11-13 | 南京市特种设备安全监督检验研究院 | A kind of unmanned plane device and detection method for crane surface defects detection |
CN109059810A (en) * | 2018-07-24 | 2018-12-21 | 天津大学 | Concretion abrasive abrasive surface landforms detection method and device |
CN109085179A (en) * | 2018-09-17 | 2018-12-25 | 周口师范学院 | A kind of board surface flaw detection device and detection method |
CN109108607A (en) * | 2018-10-24 | 2019-01-01 | 泰安康平纳机械有限公司 | Yarn bar positioning device |
US10207421B1 (en) * | 2016-09-26 | 2019-02-19 | Wein Holding LLC | Automated multi-headed saw and method for lumber |
CN109507192A (en) * | 2018-11-02 | 2019-03-22 | 江苏理工学院 | A kind of magnetic core detection method of surface flaw based on machine vision |
CN109520420A (en) * | 2018-12-21 | 2019-03-26 | 中国航空工业集团公司北京航空精密机械研究所 | A kind of space coordinate at rotation of rotary table center determines method |
CN209009859U (en) * | 2018-05-23 | 2019-06-21 | 广州赫伽力智能科技有限公司 | A kind of outer diameter crawl band u-turn yarn grabbing fixture |
-
2019
- 2019-08-20 CN CN201910767593.7A patent/CN110514664B/en active Active
Patent Citations (25)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20030081215A1 (en) * | 2001-01-09 | 2003-05-01 | Ajay Kumar | Defect detection system for quality assurance using automated visual inspection |
US20080289125A1 (en) * | 2006-08-24 | 2008-11-27 | Mario Scatizzi | Device and Method for Carrying out Optical Readings on Textile Materials Submitted to Dyeing |
CN102706274A (en) * | 2012-04-25 | 2012-10-03 | 复旦大学 | System for accurately positioning mechanical part by machine vision in industrially-structured scene |
US20140184786A1 (en) * | 2013-01-02 | 2014-07-03 | The Boeing Company | Systems and Methods for Stand-Off Inspection of Aircraft Structures |
CN103197599A (en) * | 2013-03-25 | 2013-07-10 | 东华大学 | System and method for numerical control (NC) workbench error self correction based on machine vision |
CN103406905A (en) * | 2013-08-20 | 2013-11-27 | 西北工业大学 | Robot system with visual servo and detection functions |
WO2015120734A1 (en) * | 2014-02-17 | 2015-08-20 | 华南理工大学 | Special testing device and method for correcting welding track based on machine vision |
CN103869186A (en) * | 2014-03-05 | 2014-06-18 | 随州泰华电子科技有限公司 | Tester for quartz crystal wafer |
CN104609175A (en) * | 2014-12-05 | 2015-05-13 | 机械科学研究总院先进制造技术研究中心 | Gripping device and method for high-level sarong to load and unload yarn rolls |
CN105180905A (en) * | 2015-07-23 | 2015-12-23 | 陕西科技大学 | Double-camera visual positioning system and method |
CN108476282A (en) * | 2016-01-22 | 2018-08-31 | 富士胶片株式会社 | Radiography assisting device and photography householder method |
US10207421B1 (en) * | 2016-09-26 | 2019-02-19 | Wein Holding LLC | Automated multi-headed saw and method for lumber |
CN206187342U (en) * | 2016-11-24 | 2017-05-24 | 西南科技大学 | Novel four rotor motor fixing structure |
CN107014818A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of laser carving defective vision detecting system and method |
CN107102009A (en) * | 2017-05-04 | 2017-08-29 | 武汉理工大学 | A kind of method of the cylinder spool quality testing based on machine vision |
CN108802040A (en) * | 2017-05-04 | 2018-11-13 | 南京市特种设备安全监督检验研究院 | A kind of unmanned plane device and detection method for crane surface defects detection |
CN207123496U (en) * | 2017-06-22 | 2018-03-20 | 惠州市德赛电池有限公司 | A kind of battery plus-negative plate identification equipment |
CN107742289A (en) * | 2017-10-15 | 2018-02-27 | 哈尔滨理工大学 | One kind is based on machine vision revolving body workpieces detection method |
CN207806014U (en) * | 2017-11-17 | 2018-09-04 | 青岛杰瑞工控技术有限公司 | Spool sorting system based on machine vision |
CN209009859U (en) * | 2018-05-23 | 2019-06-21 | 广州赫伽力智能科技有限公司 | A kind of outer diameter crawl band u-turn yarn grabbing fixture |
CN109059810A (en) * | 2018-07-24 | 2018-12-21 | 天津大学 | Concretion abrasive abrasive surface landforms detection method and device |
CN109085179A (en) * | 2018-09-17 | 2018-12-25 | 周口师范学院 | A kind of board surface flaw detection device and detection method |
CN109108607A (en) * | 2018-10-24 | 2019-01-01 | 泰安康平纳机械有限公司 | Yarn bar positioning device |
CN109507192A (en) * | 2018-11-02 | 2019-03-22 | 江苏理工学院 | A kind of magnetic core detection method of surface flaw based on machine vision |
CN109520420A (en) * | 2018-12-21 | 2019-03-26 | 中国航空工业集团公司北京航空精密机械研究所 | A kind of space coordinate at rotation of rotary table center determines method |
Non-Patent Citations (3)
Title |
---|
张建祥 等: "筒子纱数字化自动染色车间的装备研究与工艺应用", 《染整技术》 * |
来佳伟 等: "基于单目视觉的机械臂目标定位系统设计", 《光学技术》 * |
陈军: "基于机器视觉的汽车油泵支撑杆尺寸实时检测系统", 《中国优秀硕士学位论文全文数据库工程科技Ⅱ辑》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN110961289A (en) * | 2019-12-09 | 2020-04-07 | 国网智能科技股份有限公司 | Transformer substation insulator anti-pollution flashover coating spraying tool and spraying method |
CN110961289B (en) * | 2019-12-09 | 2021-06-29 | 国网智能科技股份有限公司 | Transformer substation insulator anti-pollution flashover coating spraying tool and spraying method |
CN110992358A (en) * | 2019-12-18 | 2020-04-10 | 北京机科国创轻量化科学研究院有限公司 | Method and device for positioning yarn rods of yarn cage, storage medium and processor |
CN110992358B (en) * | 2019-12-18 | 2023-10-20 | 北京机科国创轻量化科学研究院有限公司 | Method and device for positioning yarn rod of yarn cage, storage medium and processor |
CN117864871A (en) * | 2024-03-05 | 2024-04-12 | 北自所(北京)科技发展股份有限公司 | Automatic loading and positioning system of glass fiber doffer and operation method |
Also Published As
Publication number | Publication date |
---|---|
CN110514664B (en) | 2022-08-12 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN110514664A (en) | A kind of cheese Sha Gan detection and localization robot and method | |
CN110501342B (en) | Cheese yarn rod positioning visual detection method | |
CN105606629B (en) | A kind of eyeglass planar defect automatic detection device and method | |
CN104634791A (en) | Omnidirectional and visual on-line detection system of surface defects of strip-shaped workpiece | |
CN101456159A (en) | Spark identification tool-setting method and abrasive machining automatic system | |
CN103698094B (en) | The Work piece sealing pick-up unit of multistation | |
CN107063640B (en) | A kind of backrest-type twin-lens light axis consistency test fixture and its test method | |
CN103116351A (en) | Spinning defective cloth detection camera and detection system thereof | |
CN108534994A (en) | A kind of homogeneity test device and method of heavy caliber uniform source of light | |
CN106645193A (en) | On-line detection apparatus | |
CN105373756B (en) | A kind of laser two-dimensional code recognition device | |
CN105806323B (en) | Graticule automatic checkout system and method | |
CN111223067A (en) | Automatic alignment method for processing circular hole of bridge anchorage | |
CN213543477U (en) | Power battery pole piece coating uniformity online metering test system | |
CN105043254A (en) | Method and device for automatically detecting clothes label print quality | |
CN108426892A (en) | A kind of machine vision device of detection household plate surface defect | |
CN212320647U (en) | Rotary scanning equipment for circular object | |
CN208752021U (en) | A kind of machine vision device detecting household plate surface defect | |
CN105486691A (en) | Device and method for detecting surface quality of steel plates and automatically positioning illumination angle | |
CN106736850A (en) | The angular detent mechanism and localization method of a kind of fiber laser arrays | |
CN103499586B (en) | A kind of cillin bottle head detection device for lamp inspection machine | |
CN208012837U (en) | A kind of homogeneity test device of heavy caliber uniform source of light | |
CN217385266U (en) | Stitch detection device for special-shaped component | |
CN114392940B (en) | Pin detection method and device for special-shaped component | |
CN106477855B (en) | A kind of control method based on pid algorithm intelligence bottle-making machine |
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 |