CN106153641A - A kind of object defect check system based on machine vision and method - Google Patents
A kind of object defect check system based on machine vision and method Download PDFInfo
- Publication number
- CN106153641A CN106153641A CN201610447672.6A CN201610447672A CN106153641A CN 106153641 A CN106153641 A CN 106153641A CN 201610447672 A CN201610447672 A CN 201610447672A CN 106153641 A CN106153641 A CN 106153641A
- Authority
- CN
- China
- Prior art keywords
- image
- detection
- detection object
- chain
- feeding
- 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
- G01N21/958—Inspecting transparent materials or objects, e.g. windscreens
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/952—Inspecting the exterior surface of cylindrical bodies or wires
Abstract
The invention discloses a kind of object defect check system based on machine vision and method, system includes feeding device, transporter, image acquiring device and control and image processing module;Feeding device docks and is installed in the entrance of transporter in frame, and feeding device is for being sent to transporter by thing to be detected;Described transporter is for delivering to detecting position by thing to be detected;The action of feeding device and transporter is controlled by described control and image processing module;Frame is provided with image acquiring device;Image acquiring device is for obtaining the image of the detected material being positioned at detecting position and by image transmitting to controlling and carrying out image procossing in image processing module to judge corresponding thing to be detected whether existing defects.This system structure is compact, feature richness, the number of drawbacks that at energy detection, article exist, and is skillfully constructed, it is easy to implement on high speed detection line.
Description
Technical field
The present invention relates to a kind of object defect check system based on machine vision and method.
Background technology
On industrialized production line, it is often necessary to article are carried out detection based on image, these article be transparent or
Opaque cylindrical shape, cylindrical objects, such as infusion bottle, ampoule bottle, mineral water bottle, car light glass bulb etc..
During detection, it is necessary first to article are sent to connecting gear by design feeding device in an orderly manner;
Then, need by connecting gear, article to be delivered to detecting position, at detecting position, owing to article are carried out comprehensive by needs
Detection, then need to control article and spin, on this basis, obtain article end and the image of sidepiece, be finally based on figure
As detecting.
The common open defect of article is the open defects such as gas line, dirty speckle, end capacity check, there is also size and is not inconsistent and wants
The defect asked.
And in prior art, there is not the solution of such a entirety, therefore, it is necessary to design is a kind of based on machine
The object defect check system of device vision and method.
Summary of the invention
The technical problem to be solved is to provide a kind of object defect check system based on machine vision and side
Method, this system structure is compact, feature richness, the number of drawbacks that at energy detection, article exist, and is skillfully constructed, it is easy in high speed detection
Implement on line.
The technical solution of invention is as follows:
A kind of object defect check system based on machine vision includes feeding device, transporter, image acquiring device
With control and image processing module;
Feeding device docks and is installed in the entrance of transporter in frame, and feeding device is for passing thing to be detected
Deliver to transporter;Described transporter is for delivering to detecting position by thing to be detected;
The action of feeding device and transporter is controlled by described control and image processing module;
Frame is provided with image acquiring device;Image acquiring device is for obtaining the image of the detected material being positioned at detecting position
And by image transmitting to controlling and image processing module carrying out, whether image procossing exists scarce with the thing to be detected judging correspondence
Fall into.Further, defect detecting system also includes drawing mechanism.Drawing mechanism docks with the outlet of transmitting device;If existing defects,
The rejecting module then controlling drawing mechanism rejects defect article, otherwise allows the article to be detected of not existing defects directly export.
Described transporter uses chain delivery mechanism, and chain delivery mechanism includes chain (62), chain driving wheel
(61), wheel shaft and transfer roller (63);Chain driving wheel is arranged in frame by wheel shaft;
Chain is 2, is provided with multiple described transfer roller, the setting direction of transfer roller and chain between 2 chains
Moving direction is vertical, can accommodate a thing to be detected to be sent between 2 adjacent transfer rollers, and chain driving wheel is at least two
Group;Often group chain driving wheel is 2;
Article 2, chain is opened and is located on described chain driving wheel, 2 chains of chain drive wheel drive thus drive transfer roller to follow
Ring moves to transmit thing to be detected.
Chain driving wheel is 4 groups, and 4 groups of chain driving wheel 2 chains of support and transfer roller form the transmission of the circulation that is square
Mechanism.
Being provided with 2 pieces of baffle plates being parallel to each other in frame, 2 pieces of baffle plates are vertical direction and arrange;The bearing of trend of baffle plate
Direction of transfer for thing to be detected;2 pieces of baffle plates are positioned at the top of chain delivery mechanism, and 2 pieces of baffle plates form one and are easy to be detected
The passage that thing forwards;Chain delivery mechanism is additionally provided with many press strips;Press strip is arranged on the top of the chain at top, is used for
The chain at constraint top keeps straight in the horizontal direction when transmitting.
The transporter of thing to be detected also includes the spin drive mechanism for driving thing to be detected to spin, and spin driving machine
Structure includes the roller (64) being arranged on each transfer roller and the tape drive mechanism being used for driving rolls;
Described tape drive mechanism include bearing (73), motor (71), driving belt (72), driving pulley (74), from
Movable belt pulley (76) and roller drive belt (75);
Bearing is arranged in frame, or bearing is a part for frame;
Motor is arranged in bearing or frame;
Driving pulley and driven pulley are arranged in frame or bearing by rotating shaft and bearing;
Motor drives driving pulley to rotate by driving belt;Open on driving pulley and driven pulley and be provided with roller driving skin
Band;Roller drives belt and roller contact, and when motor rotates, roller drives belt to drive roller to rotate realization by frictional force and treats
The spin of detectable substance.Spin drive mechanism is used to realize tubular or the spin of column article;So that above tubular or column
Photographic head obtains tubular or the image of column article sidepiece outer wall.
Feeding device includes feeding platform and feed mechanism;Described feed mechanism includes directly shake base, feeding passage, backgauge
Cylinder, rotating disk and pusher;Feeding passage is the passage sequentially advanced for single detection object (such as glass bulb);Feeding passage
It is arranged on and directly shakes on base, directly shake floor installation on feeding platform;Described backgauge cylinder is positioned at the port of export of feeding passage;Gear
Material cylinder is for the detection object stopped or let pass by described feeding passage;
Rotating disk and pusher are arranged in frame, and the edge of rotating disk is provided with the curved channel of circumference, curved channel
Arc be 90 ° relative to the central angle in the rotating disk center of circle;The entrance of curved channel docks with the port of export of feeding passage;Arc leads to
The exit in road is provided with pusher, and pusher is for shifting the detection object exported from curved channel onto detection equipment.
Described pusher is pusher cylinder.Feeding passage is two-in-one passage, has two pan feeding passages and one
Outlet, 2 pan feeding passages are all connected with outlet, and each pan feeding passage is equipped with stop cylinder.Pusher is additionally provided with
Rotary cylinder, for rotating to be positive sequence by the detection object of inverted sequence;Detection object orientation detection device it is provided with at backgauge cylinder.
Described feed mechanism is 2 sets, and the feeding passage often overlapping feed mechanism is arranged on feeding platform.Feeding device also include for
Control the PLC of feeding action and whether arrive the optical fiber detector in precalculated position for detection object;PLC is by controlling
Stopping that cylinder controls 2 pan feeding passages and in turn from the port of export of feeding passage, detection object exported curved channel, it is right to detect
As exporting in pusher after turntable rotation 90 ° in curved channel, finally Object Push will be detected by pusher
On detection equipment.
Image acquiring device includes detecting object connecting gear, light sources and phase unit;Control and in image processing module
There is control module;Control module is used for being turned on and off light sources and phase unit, and control module is additionally operable to control detection object
The operation of connecting gear;Detection object connecting gear for being sent to detection station by detection object;
Described phase unit includes being arranged on the top camera at detection object connecting gear top and being arranged on detection object
The sidepiece camera of connecting gear sidepiece;The length direction of detection object is vertical with the direction of transfer of connecting gear, and top camera is used
In the image obtaining detection object sidepiece;Sidepiece camera is for obtaining the image of detection object end;
Described light sources includes backlight bottom annular coaxial light source, LED side backlight and LED, annular coaxial light
Source is positioned at the front end of detection station for illuminating, and provides prospect light for sidepiece camera;LED side backlight is positioned at detection station
Rear end, provides bias light for sidepiece camera;Bottom LED, backlight is positioned at the bottom of detection station, provides backlight for top camera;
Under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection work
The detection object spin of position, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is dark
Figure;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
Frame also sets up Rhizoma Sparganii reflecting mirror, Rhizoma Sparganii reflecting mirror between sidepiece camera and annular coaxial light source, sidepiece phase
The optical axis of machine is vertical direction.Top camera uses bilateral telecentric lens, and the focal length that uses of sidepiece camera is the camera lens of 35mm, top
Portion's camera and sidepiece camera all use gigabit Ethernet CCD camera.
A kind of article defects detection method based on machine vision, uses aforesaid article defects based on machine vision to examine
Examining system transmits determinand, obtains image and carry out defects detection based on image procossing;
Comprise the following steps:
Step one: feeding and transfer step;
Determinand is sent on transporter by feeding device, then is sent to detecting position through transporter;
Step 2: image acquisition step;
Image acquiring device is after determinand arrives detecting position, by the side image of top camera shooting determinand, logical
Cross the end view picture of sidepiece motor shooting determinand;
Step 3: image detecting step;
Carry out image procossing for described side image and end view picture, treat to measure including side image detecting step
Survey the size of thing, also include whether plate defects detecting step exists crackle for the end face detecting determinand.
The control method of described feeding comprises the steps:
Step 1: detection object, after two pan feeding passages enter, is driven detection object to enter at 2 two by the base that directly shakes
Material passage advances;
Step 2:PLC controls 2 pan feeding passages and is exported from the port of export of feeding passage by detection object in turn;And in outlet
The direction of end detection detection object;
Step 3:PLC enters in curved channel by controlling backgauge cylinder control detection object;Detection object is at arc
Passage enters delivery device after turntable rotation 90 °;
Step 4: if current detection Obj State is inverted sequence, then PLC controls rotary cylinder by the detection object rotation of inverted sequence
Turn 180 ° and become output after positive sequence;If current detection Obj State is positive sequence, then rotary cylinder is failure to actuate, and detection object is direct
Output;
Step 5:PLC control propelling movement cylinder, by the detection Object Push of output to detection equipment, completes material process.
Described feed mechanism is 2 sets, and the feeding passage often overlapping feed mechanism is arranged on feeding platform, often overlaps feeder
Structure uses identical control strategy.
In image acquisition step, under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection work
The detection object spin of position, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is dark
Figure;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
Image detecting step includes:
(1) plate defects detecting step is: for described detection object end face image, uses vertical integral projection method
Realization detection object crack detection;End view drawing picture transfers gray level image to;
Step a1: location white annulus in detection object end face image;
Step a2: be a rectangular image by white circular ring transformation;The image of the n*m size of rectangular image;Wherein n and m
It is respectively the wide number of pixels corresponding with height;
Step a3: described rectangular image is carried out binary conversion treatment, obtains bianry image;
Step a4: this bianry image is carried out vertical integral projection, obtains vertical integral projection image;
Step a5: if vertical integral projection image has the projection exceeding predetermined threshold value, then illustrate this detection object end face
There is crackle in the detection object that image is corresponding;
The described vertical integral projection in step a4 refers to the white pixel point in every string pixel of bianry image
Cumulative, obtain value f (i), i=1,2 ..., n;Then the beneath individual pixel of f (i) of the i-th row is black picture element, remaining of these row
Pixel is white pixel.
(2) side image detecting step is used for measuring detection object geometrical characteristic, described detection object geometric properties
Value includes detecting the length of object, external diameter and wall thickness;Geometrical characteristic is compared with the eigenvalue of standard with detection detection object
The most qualified;
Measurement specifically comprises the following steps that
Step b1: selected processing region;
Step b2: search marginal point in selected processing region, until marginal point quantity reaches threshold value or region is swept
Retouch complete;
Step b3: multiple marginal points are fitted to edge line with least square method;
Step b4: calculate the edge line pixel distance to corresponding edge line, then pixel distance is converted to reality
Distance, thus obtain the length of transparent object, external diameter and wall thickness.
Described geometry transparent object geometrical characteristic also includes transparent object neck gradient, and transparent object neck gradient is fixed
Justice is the horizontal range between two end points of cervical region, and concrete step is as follows:
Step a: selected processing region chooses neck area under transparent object upper neck region and transparent object;
Step b: obtain the edge flex point of neck area under transparent object upper neck region and transparent object;
Upper neck region and lower neck area are respectively divided into many sub regions:
(1) Edge Feature Points is searched in each subregion in transparent object upper neck region, when transparent object is close
Block towards time right, to take edge, the characteristic point transparent object upper neck region flex point of train value maximum;
(2) under transparent object, Edge Feature Points in each subregion in neck area, is searched, when transparent object is close
Block towards time right, neck area edge flex point under the characteristic point transparent object maximum to take train value;
(3) pixel distance of neck area edge flex point is finally calculated under edge, upper neck region flex point and transparent object again
Be converted to actual range and i.e. obtain transparent object neck gradient.
Described thing to be detected is tubular or column article.
Thing to be detected is the detection object for being arranged in car light.
In addition to detection object, the present invention can be used for other tubular or column article, such as test tube, the storage blood of blood count
The body etc. of liquid, it is also possible to be cylindrical or similar cylindrical article, such as bullet shell etc..
Beneficial effect:
The invention provides a whole set of solution of article, in particular for tubular or the column article of detection object class;
The system of the present invention uses actively feed mechanism that article are sent to connecting gear, then is sent by article by connecting gear
To detecting position, at detecting position, coordinate spin drive mechanism, top camera obtain multiple side image of article, thus spend thing
The side of product carries out omnibearing detection (including the detections such as size);Meanwhile, the image of the end that sidepiece camera obtains article (wraps
Include bright figure and dark figure), the crack defect that whether exists of detection object being detected, if there is defective product, then going out
Being rejected by defect article in material mechanism, rejecting mechanism is existing mature technology.
Feeding device uses the mechanical mode of movement of vibration feeding, has high reliability and the advantage such as speed is adjustable.Detection
Object feeding device is by stopping that cylinder controls the most current of detection object, and the Polling object of 2 passages is let pass,
Be provided with detection device detection detection object whether to put in place and be whether positive sequence, use rotating disk will detection object half-twist, use rotation
Rotaring cylinder transfers reverse detection object to positive sequence, and whole flow process is automatically performed under the control of PLC, and automaticity is high, easily
In control, concordance is good, provides solid technical foundation for follow-up detection operation.Feeding device and based on PLC control
Detection object feeding control method can ensure tubular or column article (i.e. detecting object, such as glass bulb etc.) be sent to swimmingly inspection
On measurement equipment, efficiency is high, and concordance is good, and structure is ingenious and compact, can meet the rigors on high speed detection line.
The mode that actuating device uses sprocket wheel to combine is fed forward tubular or column article, in order to tubular or column article
(such as glass bulb) does not cause damage and realizes comprehensive continuous detecting;The transporter of the present invention uses a main chain mechanism
Combine mode for (i.e. chain delivery mechanism) and a secondary belt mechanism (i.e. spin drive mechanism), and three phase electric machine controls main chain
Structure travels forward, and secondary follower belt is used for controlling to detect object and is specifying detection station high speed rotation to obtain multiple figures
Picture, thus ensure omnibearing detection.Chain induction system use two set chains add a collection of transfer roller and roller composition, tubular or
Column article (i.e. detecting object, such as glass bulb etc.) lie low inside the groove between two adjacent rollers, between adjacent rollers
Gap can ensure that when taking pictures from top, on whole tubular or column article (i.e. detecting object, such as glass bulb etc.) length direction all
Can all photograph the region of one fixed width, and roller can be freely rotatable, drives above it under the effect of drive mechanism
Tubular or column article (i.e. detecting object, such as glass bulb etc.) uniform rotation one enclose, use and take pictures for 12 times, often rotate 30 ° take pictures
Once, obtain image sequence, thus the face of cylinder photographing whole tubular or column article (i.e. detecting object, such as glass bulb etc.) is real
Existing 360 ° of comprehensive detections.Briquetting ensures that roller is in sustained height all the time, it is ensured that camera operating distance is stable.Baffle plate is used for cylinder
Guide on shape or column article (i.e. detecting object, such as glass bulb etc.) transmission, it is ensured that the operating distance of side-view camera.
The end face of article and side can be taken pictures by illumination and image acquiring device, obtain the image that detection needs, for
Follow-up detection operation provides the gray level image that concordance is good clearly.
In end surface measurement method, using the realization detection object crack detection of vertical integral projection method, the method is very
Ingenious, amount of calculation is few, is highly suitable to be applied in high-speed production lines.
In sum, the object defect check system compact conformation based on machine vision of invention, perfect in shape and function, can ensure
Tubular or column article (i.e. detecting object, such as glass bulb etc.) be sent to swimmingly image detection station, and can realize tubular or
The spin of column article (i.e. detecting object, such as glass bulb etc.) drives, and can obtain end and side image, can process image
Thus detect the various defects that article exist, the rigors on high speed detection line can be met.
Accompanying drawing explanation
The structural representation of Tu1Shi chain delivery mechanism;
Fig. 2 is the structural representation of spin drive mechanism;
Fig. 3 is feeding device top view;
Fig. 4 is feeding device axonometric chart;
Fig. 5 is feeding flow chart;
Fig. 6 is image acquiring device schematic diagram;
Fig. 7 is the general flow chart of the present invention;
Fig. 8 is detection object size schematic diagram;
Fig. 9 is detection object side image artwork;
Figure 10 is detection object torticollis degree instrumentation plan;
Figure 11 is the parameter schematic diagrams such as detection object torticollis degree;
Figure 12 is the geometric parameter schematic diagrams such as detection object length.
Figure 13 be zero defect white annulus schematic diagram (figure a is white annulus artwork, and figure b is the image after scale transformation,
Figure c is binary image, and figure d is vertical integral image);
Figure 14 is that (figure a is defect white annulus artwork, and figure is after b is scale transformation for the schematic diagram of the white annulus having crackle
Image, figure c is binary image, figure d be vertical integral image).
Figure 15 is the population structure schematic diagram (front view) of the object defect check system based on machine vision of the present invention;
Figure 16 is the population structure schematic diagram (top view) of the object defect check system based on machine vision of the present invention;
Label declaration: 1-reason material cylinder A, 2-backgauge cylinder A, 3-reason material cylinder B, 4-backgauge cylinder B, 5-backgauge cylinder C,
6-rotating disk, 7-pusher cylinder, 8-rotary cylinder, in figure, the position of arrow indication is feeding mouth.61-chain driving wheel, 62-chain
Bar, 63-transfer roller, 64-roller, 65-detects object, 66-press strip, 67-baffle plate;71-motor, 72-driving belt, 73-bearing,
74-driving pulley, 75-roller drives belt, 76-driven pulley;80-frame, 81-feeding device, 82-transporter, 83-schemes
As acquisition device.
Detailed description of the invention
Below with reference to the drawings and specific embodiments, the present invention is described in further details:
Embodiment 1: such as Fig. 1-16, a kind of object defect check system based on machine vision includes that frame 80, feeding fill
Put 81, transporter 82, image acquiring device 93 and control and image processing module;
Feeding device docks and is installed in the entrance of transporter in frame, and feeding device is for passing thing to be detected
Deliver to transporter;Described transporter is for delivering to detecting position by thing to be detected;
The action of feeding device and transporter is controlled by described control and image processing module;
Frame is provided with image acquiring device;Image acquiring device is for obtaining the image of the detected material being positioned at detecting position
And by image transmitting to controlling and image processing module carrying out, whether image procossing exists scarce with the thing to be detected judging correspondence
Fall into.Further, defect detecting system also includes drawing mechanism.Drawing mechanism docks with the outlet of transmitting device;If existing defects,
The rejecting module then controlling drawing mechanism rejects defect article, otherwise allows the article to be detected of not existing defects directly export.
Described transporter uses chain delivery mechanism, and chain delivery mechanism includes chain 62, chain driving wheel 61, wheel
Axle and transfer roller 63;Chain driving wheel is arranged in frame by wheel shaft;
Chain is 2, is provided with multiple described transfer roller, the setting direction of transfer roller and chain between 2 chains
Moving direction is vertical, can accommodate a thing to be detected to be sent between 2 adjacent transfer rollers, and chain driving wheel is at least two
Group;Often group chain driving wheel is 2;
Article 2, chain is opened and is located on described chain driving wheel, 2 chains of chain drive wheel drive thus drive transfer roller to follow
Ring moves to transmit thing to be detected.
Chain driving wheel is 4 groups, and 4 groups of chain driving wheel 2 chains of support and transfer roller form the transmission of the circulation that is square
Mechanism.
Being provided with 2 pieces of baffle plates being parallel to each other in frame, 2 pieces of baffle plates are vertical direction and arrange;The bearing of trend of baffle plate
Direction of transfer for thing to be detected;2 pieces of baffle plates are positioned at the top of chain delivery mechanism, and 2 pieces of baffle plates form one and are easy to be detected
The passage that thing forwards;Chain delivery mechanism is additionally provided with many press strips;Press strip is arranged on the top of the chain at top, is used for
The chain at constraint top keeps straight in the horizontal direction when transmitting.
The transporter of thing to be detected also includes the spin drive mechanism for driving thing to be detected to spin, and spin driving machine
Structure includes the roller (64) being arranged on each transfer roller and the tape drive mechanism being used for driving rolls;
Described tape drive mechanism includes bearing 73, motor 71, driving belt 72, driving pulley 74, driven pulley 76
Belt 75 is driven with roller;
Bearing is arranged in frame, or bearing is a part for frame;
Motor is arranged in bearing or frame;
Driving pulley and driven pulley are arranged in frame or bearing by rotating shaft and bearing;
Motor drives driving pulley to rotate by driving belt;Open on driving pulley and driven pulley and be provided with roller driving skin
Band;Roller drives belt and roller contact, and when motor rotates, roller drives belt to drive roller to rotate realization by frictional force and treats
The spin of detectable substance.
Feeding device includes feeding platform and feed mechanism;Described feed mechanism includes directly shake base, feeding passage, backgauge
Cylinder, rotating disk and pusher;Feeding passage is the passage sequentially advanced for single detection object;Feeding passage is arranged on directly
Shake on base, directly shake floor installation on feeding platform;Described backgauge cylinder is positioned at the port of export of feeding passage;Backgauge cylinder is used
In the detection object stopped or let pass by described feeding passage;
Rotating disk and pusher are arranged in frame, and the edge of rotating disk is provided with the curved channel of circumference, curved channel
Arc be 90 ° relative to the central angle in the rotating disk center of circle;The entrance of curved channel docks with the port of export of feeding passage;Arc leads to
The exit in road is provided with pusher, and pusher is for shifting the detection object exported from curved channel onto detection equipment.
Described pusher is pusher cylinder.Feeding passage is two-in-one passage, has two pan feeding passages and one
Outlet, 2 pan feeding passages are all connected with outlet, and each pan feeding passage is equipped with stop cylinder.Pusher is additionally provided with
Rotary cylinder, for rotating to be positive sequence by the detection object of inverted sequence;Detection object orientation detection device it is provided with at backgauge cylinder.
Described feed mechanism is 2 sets, and the feeding passage often overlapping feed mechanism is arranged on feeding platform.Feeding device also include for
Control the PLC of feeding action and whether arrive the optical fiber detector in precalculated position for detection object;PLC is by controlling
Stopping that cylinder controls 2 pan feeding passages and in turn from the port of export of feeding passage, detection object exported curved channel, it is right to detect
As exporting in pusher after turntable rotation 90 ° in curved channel, finally Object Push will be detected by pusher
On detection equipment.
Image acquiring device includes detecting object connecting gear, light sources and phase unit;Control and in image processing module
There is control module;Control module is used for being turned on and off light sources and phase unit, and control module is additionally operable to control detection object
The operation of connecting gear;Detection object connecting gear for being sent to detection station by detection object;
Described phase unit includes being arranged on the top camera at detection object connecting gear top and being arranged on detection object
The sidepiece camera of connecting gear sidepiece;The length direction of detection object is vertical with the direction of transfer of connecting gear, and top camera is used
In the image obtaining detection object sidepiece;Sidepiece camera is for obtaining the image of detection object end;
Described light sources includes backlight bottom annular coaxial light source, LED side backlight and LED, annular coaxial light
Source is positioned at the front end of detection station for illuminating, and provides prospect light for sidepiece camera;LED side backlight is positioned at detection station
Rear end, provides bias light for sidepiece camera;Bottom LED, backlight is positioned at the bottom of detection station, provides backlight for top camera;
Under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection work
The detection object spin of position, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is dark
Figure;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
Frame also sets up Rhizoma Sparganii reflecting mirror, Rhizoma Sparganii reflecting mirror between sidepiece camera and annular coaxial light source, sidepiece phase
The optical axis of machine is vertical direction.Top camera uses bilateral telecentric lens, and the focal length that uses of sidepiece camera is the camera lens of 35mm, top
Portion's camera and sidepiece camera all use gigabit Ethernet CCD camera.
A kind of article defects detection method based on machine vision, uses aforesaid article defects based on machine vision to examine
Examining system transmits determinand, obtains image and carry out defects detection based on image procossing;
Comprise the following steps:
Step one: feeding and transfer step;
Determinand is sent on transporter by feeding device, then is sent to detecting position through transporter;
Step 2: image acquisition step;
Image acquiring device is after determinand arrives detecting position, by the side image of top camera shooting determinand, logical
Cross the end view picture of sidepiece motor shooting determinand;
Step 3: image detecting step;
Carry out image procossing for described side image and end view picture, treat to measure including side image detecting step
Survey the size of thing, also include whether plate defects detecting step exists crackle for the end face detecting determinand.
The control method of described feeding comprises the steps:
Step 1: detection object, after two pan feeding passages enter, is driven detection object to enter at 2 two by the base that directly shakes
Material passage advances;
Step 2:PLC controls 2 pan feeding passages and is exported from the port of export of feeding passage by detection object in turn;And in outlet
The direction of end detection detection object;
Step 3:PLC enters in curved channel by controlling backgauge cylinder control detection object;Detection object is at arc
Passage enters delivery device after turntable rotation 90 °;
Step 4: if current detection Obj State is inverted sequence, then PLC controls rotary cylinder by the detection object rotation of inverted sequence
Turn 180 ° and become output after positive sequence;If current detection Obj State is positive sequence, then rotary cylinder is failure to actuate, and detection object is direct
Output;
Step 5:PLC control propelling movement cylinder, by the detection Object Push of output to detection equipment, completes material process.
Described feed mechanism is 2 sets, and the feeding passage often overlapping feed mechanism is arranged on feeding platform, often overlaps feeder
Structure uses identical control strategy.
Detection object feeding device is automatic charging device, uses the pattern of vibration feeding, and vibrator can control feeding
Speed, the detection object of detection object machine continuously enters detecting system from 4 feeding mouths simultaneously.Directly shake base by the detection of pan feeding
Object is fed forward, by reason material cylinder A1, backgauge cylinder A2, reason material cylinder B3, alternately the following of backgauge tetra-cylinders of cylinder B4
Gyration is made that the pan feeding of four passages detects object alternately interruption and is carried forward, and in backgauge cylinder C5 position to detection object
Direction detect.Detection object is by entering rotary cylinder 8 after the unified half-twist of rotating disk 6, according to detection object orientation
After the direction of detection object is rotated into unanimously by information automatically, pusher cylinder 7 detection object is promoted to enter sprocket wheel conveying knot
Structure, completes feeding work.
Reason material cylinder and the effect of backgauge cylinder: be provided to from disordered state, detection object is become order state, and
Alternately it is interrupted and carries forward, be unlikely to occur that detection object gets stuck or rambling motion causes feeding to break down.
The effect of rotary cylinder: the detection object of inverted order is become positive sequence state.
The effect of pusher cylinder: (sprocket wheel transfer structure belongs to detection equipment from pushing sprocket wheel transfer structure to by detection object
A part) on.
Automatic charging device uses the pattern of vibration feeding, and the base that directly shakes is driven by vibrator, on vibrator can control
Material speed.The detection object of detection object machine continuously enters detecting system from 4 feeding mouths simultaneously.Directly shake base by the inspection of pan feeding
Survey object is fed forward, and feeding device is additionally provided with reason material cylinder;Material tidying mechanism reason material is by photoelectric sensor, various cylinder
Control with PLC program, thus realize orderly feeding;
By the alternate cycles action of backgauge cylinder, the pan feeding of four passages is detected object alternately interruption to carry forward, and
At backgauge cylinder location, the direction of detection object is detected.Detection object enters after unifying half-twist by rotating disk
Rotary cylinder, after being automatically rotated into unanimously in the direction of detection object according to detection object orientation information, is pushed away by pusher cylinder
Dynamic detection object enters sprocket wheel structure for conveying, completes feeding work.
In order to realize orderly feeding, through debugging repeatedly and experiment, PLC controls cylinder operation program flow chart such as figure
Shown in 4.The sequence of movement that cooperates between cylinder is as follows.
Cylinder original state: stop cylinders retract, the cylinder that puts in place is closed, and blanking cylinders retract, rotary cylinder does not rotates;
Regular event flow process: detection object is advanced from the vibration of rear charge channel, stops that cylinder relies on and puts detection in place (based on arriving
Level sensor, such as photoelectric sensor) action, put in place, Exocarpium Citri Rubrum colored lights bright stop cylinder stretches out, and lamp goes out, and retracts, and another one is led to
Road action too, two passage blowings in turn.
Abnormal operation state: after cylinder is let pass, later detecting position is not detected by 1.5S, then this later detecting position
Cylinder automatically let pass detection object.
In image acquisition step, under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection work
The detection object spin of position, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is dark
Figure;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
Image detecting step includes:
(1) plate defects detecting step is: for described detection object end face image, uses vertical integral projection method
Realization detection object crack detection;End view drawing picture transfers gray level image to;
Step a1: location white annulus in detection object end face image;
Step a2: be a rectangular image by white circular ring transformation;The image of the n*m size of rectangular image;Wherein n and m
It is respectively the wide number of pixels corresponding with height;
Step a3: described rectangular image is carried out binary conversion treatment, obtains bianry image;
Step a4: this bianry image is carried out vertical integral projection, obtains vertical integral projection image;
Step a5: if vertical integral projection image has the projection exceeding predetermined threshold value, then illustrate this detection object end face
There is crackle in the detection object that image is corresponding;
The described vertical integral projection in step a4 refers to the white pixel point in every string pixel of bianry image
Cumulative, obtain value f (i), i=1,2 ..., n;Then the beneath individual pixel of f (i) of the i-th row is black picture element, remaining of these row
Pixel is white pixel.Vertical integral projection basic ideas are to take each pixel f of image (x, y) value put, then to x quadrature
Getting upright projection, floor projection of quadraturing y to obtain, its core is exactly: add up the white pixel of every string, if certain string is white
Color pixel exceedes threshold value, then existing defects.
(2) side image detecting step is used for measuring detection object geometrical characteristic, described detection object geometric properties
Value includes detecting the length of object, external diameter and wall thickness;Geometrical characteristic is compared with the eigenvalue of standard with detection detection object
The most qualified;
Measurement specifically comprises the following steps that
Step b1: selected processing region;
Step b2: search marginal point in selected processing region, until marginal point quantity reaches threshold value or region is swept
Retouch complete;
Step b3: multiple marginal points are fitted to edge line with least square method;
Step b4: calculate the edge line pixel distance to corresponding edge line, then pixel distance is converted to reality
Distance, thus obtain the length of transparent object, external diameter and wall thickness.
Described geometry transparent object geometrical characteristic also includes transparent object neck gradient, and transparent object neck gradient is fixed
Justice is the horizontal range between two end points of cervical region, and concrete step is as follows:
Step a: selected processing region chooses neck area under transparent object upper neck region and transparent object;
Step b: obtain the edge flex point of neck area under transparent object upper neck region and transparent object;
Upper neck region and lower neck area are respectively divided into many sub regions:
(1) Edge Feature Points is searched in each subregion in transparent object upper neck region, when transparent object is close
Block towards time right, to take edge, the characteristic point transparent object upper neck region flex point of train value maximum;
(2) under transparent object, Edge Feature Points in each subregion in neck area, is searched, when transparent object is close
Block towards time right, neck area edge flex point under the characteristic point transparent object maximum to take train value;
(3) pixel distance of neck area edge flex point is finally calculated under edge, upper neck region flex point and transparent object again
Be converted to actual range and i.e. obtain transparent object neck gradient.
In some instantiation, the detection object length recorded is 1218, and detection object external diameter is 410, detects object
Wall thickness is 60, and all of measurement result is all referring to the number of pixels under image coordinate.First the upper and lower neck regions of detection and localization object
Territory, then becomes 10 zonules by cervical region on detection object with the region segmentation of lower cervical region, searches feature in each zonule
Point, upper cervical region finds 5 feature point pairs, and remaining 5 are not found, and takes the train value maximum some edge flex point as upper cervical region,
In like manner, lower cervical region finds 7 feature point pairs altogether, takes the maximum point of train value equally as lower cervical region flex point, finally calculates two and turn
I.e. to detect object neck gradient be 142 to the vertical dimension of point, then compare with standard value and judge whether defect.
Claims (10)
1. an object defect check system based on machine vision, it is characterised in that include feeding device, transporter, figure
As acquisition device and control and image processing module;
Feeding device docks and is installed in the entrance of transporter in frame, and feeding device is for being sent to thing to be detected
Transporter;Described transporter is for delivering to detecting position by thing to be detected;
The action of feeding device and transporter is controlled by described control and image processing module;
Frame is provided with image acquiring device;Image acquiring device is for obtaining the image of the detected material being positioned at detecting position and inciting somebody to action
Image transmitting is to controlling and carrying out image procossing in image processing module to judge corresponding thing to be detected whether existing defects.
Object defect check system based on machine vision the most according to claim 1, it is characterised in that described transmission
Device uses chain delivery mechanism, and chain delivery mechanism includes chain (62), chain driving wheel (61), wheel shaft and transfer roller
(63);Chain driving wheel is arranged in frame by wheel shaft;
Chain is 2, is provided with multiple described transfer roller, the setting direction of transfer roller and the movement of chain between 2 chains
Direction is vertical, can accommodate a thing to be detected to be sent between 2 adjacent transfer rollers, and chain driving wheel is at least two groups;
Often group chain driving wheel is 2;
Article 2, chain is opened and is located on described chain driving wheel, 2 chains of chain drive wheel drive thus drive transfer roller circulation move
Dynamic to transmit thing to be detected.
Object defect check system based on machine vision the most according to claim 2, it is characterised in that chain driving wheel
Being 4 groups, 4 groups of chain driving wheel 2 chains of support and transfer roller form the connecting gear of the circulation that is square.
Object defect check system based on machine vision the most according to claim 2, it is characterised in that arrange in frame
2 pieces of baffle plates being parallel to each other, 2 pieces of baffle plates are had to be vertical direction and arrange;The bearing of trend of baffle plate is the sender of thing to be detected
To;2 pieces of baffle plates are positioned at the top of chain delivery mechanism, and 2 pieces of baffle plates form one and are easy to the passage that thing to be detected forwards;Chain
Formula connecting gear is additionally provided with many press strips;Press strip is arranged on the top of the chain at top, is passing for retraining the chain at top
Keep straight when sending in the horizontal direction.
The transporter of thing to be detected the most according to claim 2, it is characterised in that the transporter of thing to be detected also wraps
Including the spin drive mechanism for driving thing to be detected to spin, spin drive mechanism includes the rolling being arranged on each transfer roller
Wheel (64) and the tape drive mechanism for driving rolls;
Described tape drive mechanism includes bearing (73), motor (71), driving belt (72), driving pulley (74), driven belt
Wheel (76) and roller drive belt (75);
Bearing is arranged in frame, or bearing is a part for frame;
Motor is arranged in bearing or frame;
Driving pulley and driven pulley are arranged in frame or bearing by rotating shaft and bearing;
Motor drives driving pulley to rotate by driving belt;Open on driving pulley and driven pulley and be provided with roller driving belt;
Roller driving belt and roller contact, when motor rotates, roller drives belt to drive roller rotation to realize to be checked by frictional force
Survey the spin of thing.
Object defect check system based on machine vision the most according to claim 1, it is characterised in that
Feeding device includes feeding platform and feed mechanism;Described feed mechanism includes directly shake base, feeding passage, backgauge gas
Cylinder, rotating disk and pusher;Feeding passage is the passage sequentially advanced for single detection object;Feeding passage is arranged on and directly shakes
On base, directly shake floor installation on feeding platform;Described backgauge cylinder is positioned at the port of export of feeding passage;Backgauge cylinder is used for
The detection object stopped or let pass by described feeding passage;
Rotating disk and pusher are arranged in frame, and the edge of rotating disk is provided with the curved channel of circumference, the arc of curved channel
Line segment is 90 ° relative to the central angle in the rotating disk center of circle;The entrance of curved channel docks with the port of export of feeding passage;Curved channel
Exit is provided with pusher, and pusher is for shifting the detection object exported from curved channel onto detection equipment.
Object defect check system based on machine vision the most according to claim 1, it is characterised in that Image Acquisition fills
Put and include detecting object connecting gear, light sources and phase unit;Control and image processing module has control module;Control mould
Block is used for being turned on and off light sources and phase unit, and control module is additionally operable to control the operation of detection object connecting gear;Detection
Object connecting gear for being sent to detection station by detection object;
Described phase unit includes the top camera being arranged on detection object connecting gear top and is arranged on detection object transmission
The sidepiece camera of mechanism's sidepiece;The length direction of detection object is vertical with the direction of transfer of connecting gear, and top camera is used for obtaining
Take the image of detection object sidepiece;Sidepiece camera is for obtaining the image of detection object end;
Described light sources includes backlight bottom annular coaxial light source, LED side backlight and LED, and annular coaxial light source is used
In illuminating the front end being positioned at detection station, provide prospect light for sidepiece camera;After LED side backlight is positioned at detection station
End, provides bias light for sidepiece camera;Bottom LED, backlight is positioned at the bottom of detection station, provides backlight for top camera;
Under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection station
Detection object spin, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is secretly schemed;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
8. an article defects detection method based on machine vision, it is characterised in that use described in any one of claim 1-7
Object defect check system based on machine vision transmit determinand, obtain image and carry out defect inspection based on image procossing
Survey;
Comprise the following steps:
Step one: feeding and transfer step;
Determinand is sent on transporter by feeding device, then is sent to detecting position through transporter;
Step 2: image acquisition step;
Image acquiring device, after determinand arrives detecting position, by the side image of top camera shooting determinand, passes through side
The end view picture of portion's motor shooting determinand;
Step 3: image detecting step;
Image procossing is carried out, including side image detecting step to measure determinand for described side image and end view picture
Size, also include whether plate defects detecting step exists crackle for the end face detecting determinand.
Article defects detection method based on machine vision the most according to claim 1, it is characterised in that described feeding
Control method comprises the steps:
Step 1: detection object, after two pan feeding passages enter, is driven detection object to lead at 2 two pan feedings by the base that directly shakes
Road advances;
Step 2:PLC controls 2 pan feeding passages and is exported from the port of export of feeding passage by detection object in turn;And examine at the port of export
Survey the direction of detection object;
Step 3:PLC enters in curved channel by controlling backgauge cylinder control detection object;Detection object is at curved channel
In along entering delivery device after turntable rotation 90 °;
Step 4: if current detection Obj State is inverted sequence, then PLC controls rotary cylinder by the detection object rotation of inverted sequence
180 ° become output after positive sequence;If current detection Obj State is positive sequence, then rotary cylinder is failure to actuate, and detection object is the most defeated
Go out;
Step 5:PLC control propelling movement cylinder, by the detection Object Push of output to detection equipment, completes material process.
In image acquisition step, under the control of control module:
(1), when bottom only LED, backlight is opened, control module is driven by spin drive mechanism and is positioned at detection station
Detection object spin, top camera obtains multiple side image of this detection object;
(2) when only LED side backlight is opened, the end face that sidepiece camera acquisition detection station detects object is secretly schemed;
(3) when only annular coaxial light source is opened, sidepiece camera obtains the bright figure of end face detecting object on detection station.
Article defects detection method based on machine vision the most according to claim 9, it is characterised in that
(1) plate defects detecting step is: for described detection object end face image, use the reality of vertical integral projection method
Now detection object crack detection;End view drawing picture transfers gray level image to;
Step a1: location white annulus in detection object end face image;
Step a2: be a rectangular image by white circular ring transformation;The image of the n*m size of rectangular image;Wherein n and m is respectively
For the number of pixels that width is corresponding with height;
Step a3: described rectangular image is carried out binary conversion treatment, obtains bianry image;
Step a4: this bianry image is carried out vertical integral projection, obtains vertical integral projection image;
Step a5: if vertical integral projection image has the projection exceeding predetermined threshold value, then illustrate this detection object end face image
There is crackle in corresponding detection object;
The described vertical integral projection in step a4 refers to add up the white pixel point in every string pixel of bianry image,
Obtain value f (i), i=1,2 ..., n;Then the beneath individual pixel of f (i) of the i-th row is black picture element, and the rest of pixels of these row is
White pixel.
(2) side image detecting step is used for measuring detection object geometrical characteristic, described detection object geometrical characteristic bag
Include the detection length of object, external diameter and wall thickness;Geometrical characteristic is compared so that whether detection detects object with the eigenvalue of standard
Qualified;
Measurement specifically comprises the following steps that
Step b1: selected processing region;
Step b2: search marginal point in selected processing region, until marginal point quantity reaches threshold value or sector scanning is complete
Finish;
Step b3: multiple marginal points are fitted to edge line with least square method;
Step b4: calculate the edge line pixel distance to corresponding edge line, then pixel distance is converted to actual range,
Thus obtain the length of transparent object, external diameter and wall thickness.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610447672.6A CN106153641A (en) | 2016-06-20 | 2016-06-20 | A kind of object defect check system based on machine vision and method |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610447672.6A CN106153641A (en) | 2016-06-20 | 2016-06-20 | A kind of object defect check system based on machine vision and method |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106153641A true CN106153641A (en) | 2016-11-23 |
Family
ID=57353500
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610447672.6A Pending CN106153641A (en) | 2016-06-20 | 2016-06-20 | A kind of object defect check system based on machine vision and method |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106153641A (en) |
Cited By (17)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN106871801A (en) * | 2017-02-24 | 2017-06-20 | 哈工大机器人集团(哈尔滨)华粹智能装备有限公司 | A kind of Machine Vision Inspecting System of the lithium battery parameter of automatic detection on a moving belt |
CN107014822A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of non-woven fabrics gauze mask defective vision detecting system and method |
CN107991316A (en) * | 2017-12-27 | 2018-05-04 | 深圳市山达士电子有限公司 | A kind of two-channel intelligent high-speed detecting machine and its automatic charging package detection production line |
CN108037126A (en) * | 2018-01-19 | 2018-05-15 | 苏州杰锐思自动化设备有限公司 | Circumference appearance detects module |
CN108037130A (en) * | 2017-12-01 | 2018-05-15 | 深圳市科陆电子科技股份有限公司 | The lug defect automatic testing method and automatic detection device of battery core |
CN108072668A (en) * | 2016-11-18 | 2018-05-25 | 中国科学院沈阳自动化研究所 | Bullet open defect automatic recognition system based on Photoelectric Detection |
CN108362705A (en) * | 2018-02-05 | 2018-08-03 | 中达电通股份有限公司 | A kind of toothbrush bristle quantity detection means and its detection method |
CN108393933A (en) * | 2018-05-25 | 2018-08-14 | 饶先梅 | A kind of robot production device for fast detecting |
CN108709888A (en) * | 2018-03-19 | 2018-10-26 | 哈尔滨理工大学 | A kind of cylindrical roller surface defect detection apparatus based on machine vision |
CN110220921A (en) * | 2019-05-22 | 2019-09-10 | 武汉科技大学 | A kind of steel tube surface image collecting device and acquisition method based on machine vision |
CN110530891A (en) * | 2019-08-29 | 2019-12-03 | 苏州市兆众自动化有限公司 | A kind of paper bottle production defect intelligent checking system and its application apparatus |
WO2020103324A1 (en) * | 2018-11-20 | 2020-05-28 | 深圳市维图视技术有限公司 | On-line mask detection system and method |
CN111693533A (en) * | 2020-06-11 | 2020-09-22 | 南通通富微电子有限公司 | Workpiece surface quality detection method and device and appearance machine |
CN111812110A (en) * | 2020-08-06 | 2020-10-23 | 东方蓝天钛金科技有限公司 | Fluorescent penetrant inspection visual identification system |
CN113109339A (en) * | 2021-04-08 | 2021-07-13 | 仲恺农业工程学院 | Control method of defect visual inspection equipment |
CN116689311A (en) * | 2023-08-08 | 2023-09-05 | 慧三维智能科技(苏州)有限公司 | Shaft side wall appearance detection device and shaft appearance detection device thereof |
CN110220921B (en) * | 2019-05-22 | 2024-04-19 | 武汉科技大学 | Steel pipe surface image acquisition device and acquisition method based on machine vision |
Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009115169A1 (en) * | 2008-03-20 | 2009-09-24 | Khs Ag | Transport system for bottles and device for the treatment of bottles |
CN201485022U (en) * | 2009-08-20 | 2010-05-26 | 潘新江 | Conveying device for fully-computerized capsule detector |
CN201689057U (en) * | 2010-05-13 | 2010-12-29 | 山东大学 | Intelligent empty bottle detecting machine based on client-server |
CN201703598U (en) * | 2010-05-19 | 2011-01-12 | 山东明佳包装检测科技有限公司 | Novel comprehensive intelligent bottle inspector detecting device |
CN102053091A (en) * | 2010-11-10 | 2011-05-11 | 王喆 | Vision detection device used for detecting defects on surface of plastic medicine bottle |
CN102167211A (en) * | 2011-04-08 | 2011-08-31 | 扬州福尔喜果蔬汁机械有限公司 | Rolling rod fruit charging method |
CN202256180U (en) * | 2011-09-29 | 2012-05-30 | 佛山华兴玻璃有限公司 | Glass bottle auxiliary detecting device |
CN203699337U (en) * | 2014-01-15 | 2014-07-09 | 薛伊婷 | Steering conveying mechanism |
CN104648983A (en) * | 2015-02-15 | 2015-05-27 | 中国大冢制药有限公司 | Automatic confluence device for ampoule transfusion product advancing bottle body |
CN104964980A (en) * | 2015-06-05 | 2015-10-07 | 电子科技大学 | Machine vision-based detection method for defect on end face of spark plug |
CN205192965U (en) * | 2015-11-20 | 2016-04-27 | 山东科技大学 | Bottle embryo detecting system based on it is embedded |
-
2016
- 2016-06-20 CN CN201610447672.6A patent/CN106153641A/en active Pending
Patent Citations (11)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2009115169A1 (en) * | 2008-03-20 | 2009-09-24 | Khs Ag | Transport system for bottles and device for the treatment of bottles |
CN201485022U (en) * | 2009-08-20 | 2010-05-26 | 潘新江 | Conveying device for fully-computerized capsule detector |
CN201689057U (en) * | 2010-05-13 | 2010-12-29 | 山东大学 | Intelligent empty bottle detecting machine based on client-server |
CN201703598U (en) * | 2010-05-19 | 2011-01-12 | 山东明佳包装检测科技有限公司 | Novel comprehensive intelligent bottle inspector detecting device |
CN102053091A (en) * | 2010-11-10 | 2011-05-11 | 王喆 | Vision detection device used for detecting defects on surface of plastic medicine bottle |
CN102167211A (en) * | 2011-04-08 | 2011-08-31 | 扬州福尔喜果蔬汁机械有限公司 | Rolling rod fruit charging method |
CN202256180U (en) * | 2011-09-29 | 2012-05-30 | 佛山华兴玻璃有限公司 | Glass bottle auxiliary detecting device |
CN203699337U (en) * | 2014-01-15 | 2014-07-09 | 薛伊婷 | Steering conveying mechanism |
CN104648983A (en) * | 2015-02-15 | 2015-05-27 | 中国大冢制药有限公司 | Automatic confluence device for ampoule transfusion product advancing bottle body |
CN104964980A (en) * | 2015-06-05 | 2015-10-07 | 电子科技大学 | Machine vision-based detection method for defect on end face of spark plug |
CN205192965U (en) * | 2015-11-20 | 2016-04-27 | 山东科技大学 | Bottle embryo detecting system based on it is embedded |
Non-Patent Citations (4)
Title |
---|
封帆等: "基于智能机器视觉的针剂生产线安瓶检测识别系统", 《自动化博览》 * |
康晶: "图像分析在工业生产中的应用-啤酒瓶缺陷检测及关键技术研究", 《中国优秀硕士学位论文全文数据库》 * |
张亚楠: ""基于图像的线路板参数测量方法研究"", 《中国优秀硕士学位论文全文数据库》 * |
王为农等: "《影像测量仪技术基础》", 31 March 2010, 中国商业出版社 * |
Cited By (23)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108072668A (en) * | 2016-11-18 | 2018-05-25 | 中国科学院沈阳自动化研究所 | Bullet open defect automatic recognition system based on Photoelectric Detection |
CN106871801B (en) * | 2017-02-24 | 2018-03-13 | 中山慧能科技有限公司 | A kind of Machine Vision Inspecting System of the lithium battery parameter of automatic detection on a moving belt |
CN106871801A (en) * | 2017-02-24 | 2017-06-20 | 哈工大机器人集团(哈尔滨)华粹智能装备有限公司 | A kind of Machine Vision Inspecting System of the lithium battery parameter of automatic detection on a moving belt |
CN107014822A (en) * | 2017-02-28 | 2017-08-04 | 深圳市维图视技术有限公司 | A kind of non-woven fabrics gauze mask defective vision detecting system and method |
CN108037130A (en) * | 2017-12-01 | 2018-05-15 | 深圳市科陆电子科技股份有限公司 | The lug defect automatic testing method and automatic detection device of battery core |
CN107991316A (en) * | 2017-12-27 | 2018-05-04 | 深圳市山达士电子有限公司 | A kind of two-channel intelligent high-speed detecting machine and its automatic charging package detection production line |
CN108037126A (en) * | 2018-01-19 | 2018-05-15 | 苏州杰锐思自动化设备有限公司 | Circumference appearance detects module |
CN108362705A (en) * | 2018-02-05 | 2018-08-03 | 中达电通股份有限公司 | A kind of toothbrush bristle quantity detection means and its detection method |
CN108709888A (en) * | 2018-03-19 | 2018-10-26 | 哈尔滨理工大学 | A kind of cylindrical roller surface defect detection apparatus based on machine vision |
CN108393933B (en) * | 2018-05-25 | 2021-03-12 | 中重(泸州)特种机器人有限公司 | Quick detection device is used in robot production |
CN108393933A (en) * | 2018-05-25 | 2018-08-14 | 饶先梅 | A kind of robot production device for fast detecting |
WO2020103324A1 (en) * | 2018-11-20 | 2020-05-28 | 深圳市维图视技术有限公司 | On-line mask detection system and method |
CN110220921B (en) * | 2019-05-22 | 2024-04-19 | 武汉科技大学 | Steel pipe surface image acquisition device and acquisition method based on machine vision |
CN110220921A (en) * | 2019-05-22 | 2019-09-10 | 武汉科技大学 | A kind of steel tube surface image collecting device and acquisition method based on machine vision |
CN110530891B (en) * | 2019-08-29 | 2022-03-22 | 苏州市兆众自动化有限公司 | Defect intelligent detection system for paper bottle production and application equipment thereof |
CN110530891A (en) * | 2019-08-29 | 2019-12-03 | 苏州市兆众自动化有限公司 | A kind of paper bottle production defect intelligent checking system and its application apparatus |
CN111693533B (en) * | 2020-06-11 | 2023-01-20 | 南通通富微电子有限公司 | Workpiece surface quality detection method and device and appearance machine |
CN111693533A (en) * | 2020-06-11 | 2020-09-22 | 南通通富微电子有限公司 | Workpiece surface quality detection method and device and appearance machine |
CN111812110A (en) * | 2020-08-06 | 2020-10-23 | 东方蓝天钛金科技有限公司 | Fluorescent penetrant inspection visual identification system |
CN113109339A (en) * | 2021-04-08 | 2021-07-13 | 仲恺农业工程学院 | Control method of defect visual inspection equipment |
CN113109339B (en) * | 2021-04-08 | 2022-05-27 | 仲恺农业工程学院 | Control method of defect visual inspection equipment |
CN116689311A (en) * | 2023-08-08 | 2023-09-05 | 慧三维智能科技(苏州)有限公司 | Shaft side wall appearance detection device and shaft appearance detection device thereof |
CN116689311B (en) * | 2023-08-08 | 2023-10-20 | 慧三维智能科技(苏州)有限公司 | Shaft side wall appearance detection device and shaft appearance detection device thereof |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN106153641A (en) | A kind of object defect check system based on machine vision and method | |
CN104359916B (en) | Detection method and detection device for detecting foreign matters in bottled liquid based on reflected light path design | |
CN108405368B (en) | A kind of comprehensive automatic detecting machine of intelligent control | |
CN105523369A (en) | Casing pipe detection equipment and detection method thereof | |
CN106645177B (en) | Battery-shell surface vision-based detection assembly line and its inner subface inspection device | |
CN102941189B (en) | Controllable rotary angle insection tray device of fruit sorter | |
CN101692056A (en) | Method and device for automatically detecting impurities in health wine | |
CN207036716U (en) | A kind of packaging bag defect detecting device | |
CN106622991A (en) | Cylindrical work piece surface detecting assembly line, detection method and gesture adjusting device | |
CN102735694A (en) | Automatic capsule detection machine | |
CN106093060A (en) | A kind of cylinder products appearance detecting system | |
CN111282837A (en) | Online visual detection device for medicinal glass bottles | |
CN205861570U (en) | A kind of full-automatic horizontal candling apparatus | |
CN210665538U (en) | Machine is examined to full-automatic lamp of oral liquid | |
CN208888144U (en) | A kind of translucent or opaque empty bottle outer wall testing agency | |
CN106153640A (en) | The omnidirectional images of a kind of transparent object obtains system and method | |
CN201689061U (en) | Detecting device for transparent bottle wall of cylinder | |
CN106044060A (en) | Conveying and self-rotating device and method for cylindrical or columnar object | |
CN106871801B (en) | A kind of Machine Vision Inspecting System of the lithium battery parameter of automatic detection on a moving belt | |
CN205015291U (en) | Medicine bottle detects machine with particle impurity detects function | |
CN205010982U (en) | It detects machine to revolve bottle formula medicine bottle | |
CN202502057U (en) | Transparent capsule quality inspection machine | |
CN106044119B (en) | A kind of industrial production line automatic charging device and control method | |
CN219689303U (en) | Food bottle lid plastic envelope membrane detects machine | |
CN209182252U (en) | Roller surface defect detection equipment |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
C06 | Publication | ||
PB01 | Publication | ||
C10 | Entry into substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20161123 |
|
WD01 | Invention patent application deemed withdrawn after publication |