CN202710486U - Automatic detection device of metal self-adhesive label - Google Patents
Automatic detection device of metal self-adhesive label Download PDFInfo
- Publication number
- CN202710486U CN202710486U CN 201220354805 CN201220354805U CN202710486U CN 202710486 U CN202710486 U CN 202710486U CN 201220354805 CN201220354805 CN 201220354805 CN 201220354805 U CN201220354805 U CN 201220354805U CN 202710486 U CN202710486 U CN 202710486U
- Authority
- CN
- China
- Prior art keywords
- industrial camera
- automatic detection
- detection device
- industrial
- slide rail
- 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.)
- Expired - Fee Related
Links
Images
Landscapes
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The utility model relates to an automatic detection device of a metal self-adhesive label. The automatic detection device of the metal self-adhesive label comprises a lighting source, an industrial camera, an assembly line belt, a laser sensor, an industrial personal computer, a stepper motor, a removing cylinder, a ball screw and a linear slide rail; the lighting source is arranged between the industrial camera and the assembly line belt; the industrial camera is connected with an output end of the laser sensor by virtue of an arranged hardware trigger port; the laser sensor and the removing cylinder are fixed on the assembly line belt; the industrial personal computer is respectively connected with the industrial camera, the stepper motor and the removing cylinder; the stepper motor drives the ball screw and the linear slide rail by virtue of a coupling; four slide blocks are arranged on the linear slide rail; and a nut on the ball screw is combined with the linear slide rail and the slide blocks by virtue of a connection block, and then drives the industrial camera to move up and down by virtue of a connection rod. Therefore, the automatic detection device based on machine vision provided by the utility model can achieve rapid and stable online detection and can automatically reject defective products.
Description
Technical field
The utility model relates to machine vision and automatic control technology field, particularly a kind of based on machine vision technique, can be automatically the defectives such as the disappearance of metal self-adhesion label on the metal self-adhesion label surface, crooked, skew, warpage be detected automatically, and the autonomous detection device that defective products is got rid of automatically.
Background technology
Metal self-adhesion label is a kind of novel package decoration technique, this technology mainly be on the noble metal sheet materials such as gold, silver, copper by modes such as etchings to obtain pattern or character, its back side is the super glue sticky paper, it is positive to be the transfer film.Tear the back side sticky paper during use off label is bonded in decorated metal self-adhesion label surface, removing surface transfer film gets final product again, it is topmost label kind on present high-end metal self-adhesion label such as the packing boxes such as famous and precious medicine, cosmetics, commemorative coin, jewelry and the artwork, and because its technique is simple and convenient, be the best alternative metals self-adhesion label of various adhesive labels, aluminium board, bronze medal, iron board etc., market has a wide range of applications.
Mainly there is following defective in metal self-adhesion label technique of the prior art:
The one, lighting source.Because the metal self-adhesive label is precious metal material, its gluing of surfaces generally also adopts paint baking, and approximate mirror effect is reflective serious, adopts conventional lighting system easily to form inverted image on the surface.Its material is different simultaneously, and the color of metal label and background is also different, and monochromatic illumination is all adopted in now general illumination, and its applicability is very narrow, is difficult to obtain preferably image of contrast; Three look lighting systems also occurred abroad adopting, but its control is manual control, namely by manually regulating three kinds of LED brightness of red, green, blue, does not have concrete adjustment criteria, can only be controlled by artificial feel, namely science not is also inconvenient.
The 2nd, sensor.Because detected metal self-adhesion label difference in thickness is very large, the thickest 230mm, the thinnest only 2mm of surpassing.And metal self-adhesion label is placed on the streamline belt and thereupon motion, and belt edge is shaken up and down, as adopts common correlation photoelectric sensor, is easily shaken by belt edge and the false triggering detection that makes the mistake.And adopt common reflective photoelectric sensor, can't adapt to again the metal self-adhesion label that thickness differs greatly.
The 3rd, the position adjustment such as camera, light source.Metal self-adhesion label difference in thickness is very large, is not on the focal plane such as metal self-adhesion label surface, and image is then smudgy and can't detect.General settling mode is manually to adjust camera and light source position, but in the actual production of the many kinds of small batch, causes very large workload, has also delayed many production times.Another kind of settling mode is the electronic automatic focus-adjusting lens that adopts external import, but this type of camera lens is expensive especially, brings very large equipment cost, the simultaneously also motion of uncontrollable light source.
The 4th, manual work.When the bonding process of metal self-adhesion label, owing to connecting less between the ultra-thin and pattern character of label or without connection, be easy to cause the defectives such as character or pattern disappearance, crooked, skew, warpage, have a strong impact on metal self-adhesion label outward appearance, this is unallowed in the production of metal self-adhesion label.Be aborning at present the artificial visual inspection, but since pattern be metal material and reflective serious, character pattern is more, metal self-adhesion label kind is a lot, production quantity is larger, easily cause visual fatigue so that the phenomenon such as undetected, false retrieval occurs, and detection efficiency is low.At home in the package manufacturing enterprise, there is no and to carry out automatic detection device to this at present.
Therefore, how realizing automatically detecting this type of defective in technique, improve metal self-adhesion label personal style and attitude matter, reduce production costs, is the main bugbear that wrapping enterprise adopts metal self-adhesion label.
The utility model content
Technical problem to be solved in the utility model provides a kind of automatic detection device of the metal self-adhesion label based on machine vision technique, is used for solving artificial visual and checks that metal self-adhesion label is prone to the problems such as undetected, false retrieval.
The technical scheme that the utility model solves the problems of the technologies described above is as follows: a kind of automatic detection device of metal self-adhesion label comprises lighting source, industrial camera, streamline belt, laser sensor, industrial computer, step motor, eliminating cylinder, ball screw and line slide rail;
Described streamline belt is used for transmitting metal self-adhesion label in the described automatic detection device course of work.
Described lighting source is placed between industrial camera and the streamline belt, and described industrial camera and described lighting source coaxial cable be perpendicular to the streamline belt surface, the illumination when described lighting source is used for imaging;
Described industrial camera is connected with the output terminal of described laser sensor by the hardware trigger port that arranges, and is used for obtaining the triggering sensing by described laser sensor, realizes optical imagery; Described laser sensor is fixed on the described streamline belt, and the photoelectricity that laser is dropped on the streamline belt is positioned at the partially front place of industrial camera vertical central, for the consistance of the image that guarantees to collect at every turn;
Described industrial computer is connected with industrial camera, step motor and eliminating cylinder respectively;
Described step motor drives ball screw by shaft coupling and is rotated, and driving line slide rail, described line slide rail is provided with four slide blocks, nut on the ball screw makes to rotatablely move by contiguous block and line slide rail and slide block combination and changes rectilinear motion into, moves up and down thereby drive industrial camera by connecting link; Here, the composite set of nut, line slide rail and the slide block of ball screw can be installed on the web joint, drive connecting link by web joint again.
Described eliminating cylinder is installed on the streamline belt, and is connected with described industrial computer by solenoid valve, is used for defective products is got rid of.Except getting rid of cylinder, can also adopt other can realize getting rid of the device of defective products function.
In the utility model, lighting source, industrial camera and laser sensor have consisted of image collecting device, and it be used for to realize optical imagery and Collect conversion, and with the image signal transmission that collects to described industrial computer.
Described industrial computer namely is the industry control PC, is used for the control of realization system and image processing function.
In addition, step motor, ball screw, shaft coupling, connecting link, line slide rail and slide block etc. have consisted of a servo drive, are used for adjusting and driving above-mentioned image collecting device moving up and down.Servo drive drives image collecting device according to metal self-adhesion label thickness difference and moves up and down, to guarantee to obtain best image.
On the basis of technique scheme, the utility model can also be done following improvement.
Further, on the described industrial camera industrial lens is installed, has consisted of an optical imaging device.
Further, described lighting source adopts dome-type dome diffuse reflective structure, two circle red, green, blues, three paths of LEDs lamps are installed on it, and be provided with light source controller on the described lighting source, be used for independent output triple channel electric current with control red, green, blue three paths of LEDs lamps, described light source controller also is connected with described industrial computer, is used for to the colouring information of described industrial computer transmission metal self-adhesion label.For dome-type dome diffuse reflective structure, namely be to form a dome-type outer cover with the stainless steel punching press, inner spray is with the graininess white paint, in ball-type outer cover bottom one annular LED light circle is installed, the annular external diameter is identical with ball-type outer cover base diameter, internal diameter is less than external diameter 40mm, upper installation two circle LED, LED adopts three kinds of colors of red, green, blue to mix successively arrangement, every kind of separately constant current control of LED, form the light of different colours with control LED luminosity by the control electric current, with the detected metal self-adhesion label of applicable different colours.Particularly, LED is luminous during energising, light upwards and around shine on the graininess white paint of ball-type outer cover again that secondary reflection gets off, form the diffuse illumination object to be detected, avoided the reflective shortcoming of common annular, bar shaped or flat illumination fully.Light source controller can independently be exported the triple channel electric current, with control red, green, blue three paths of LEDs lamps.In addition, light source controller links to each other with industrial computer by the RS232 communication interface, automatically generates the optimal illumination parameter by industrial computer according to metal self-adhesion label color.
Further, described laser sensor is fixed on the described streamline belt by universal installation frame.
Further, described industrial computer is connected with industrial camera, step motor and eliminating cylinder respectively by three PCI slots are set at described industrial computer to be realized, described three PCI slots are respectively applied to in-line image capture card, motion control card and digital I/O card, described image pick-up card links to each other with industrial camera by data line, described motion control card is connected with described step motor, and described digital I/O cartoon is crossed solenoid valve and is connected with the eliminating cylinder;
Here, described image pick-up card provides the IEEE1394 interface, link to each other with industrial camera by data line, the parameters such as control every time shutter of camera, gain when system initialization, when starting working, camera obtains the image information that collects, and it is regional that it is deposited specified memory, and the mode with message sends described industrial computer processing to again;
Described digital I/O card also is used for accepting to start stop signal and driving acoustic-optic alarm except being used for driving described eliminating cylinder action simultaneously.
Described motion control card is used for controlling described step motor by the output pulse, thereby controls further the congregation of camera etc.
Further, described industry control function realizes automatically generating optimal illumination parameter function, Template Learning function or automatic detection function by software form.
Further, the mode of calculation of characteristic parameters and template matches is adopted in the realization of described automatic detection function: described calculation of characteristic parameters comprises the calculating of area, girth, circularity, center of gravity, length and width and the anglec of rotation to the target area; Described template matches refers to seek the target area in image, calculate similarity, differentiates current goal whether identical with target setting.Here, described area is exactly the quantity of pixel in the target area; Described girth refers on the outline line of target area spacing sum between the pixel; Described circularity refers on the basis of area and perimeter, calculates a characteristic quantity of target shape complexity, specifically is calculated as
Be the circle of r for radius, area is π r
2, girth is 2 π r, if shape near circle, then this value is more near 1, shape is more complicated, then should the value less; The mean value of pixel coordinate in the described center of gravity feeling the pulse with the finger-tip mark; Described length and width comprise the length and width of target minimum parallel axle boundary rectangle and the length and width of the minimum boundary rectangle of any direction; The angle of major axis and transverse axis in the oval feature in described anglec of rotation feeling the pulse with the finger-tip mark zone.
The beneficial effects of the utility model are:
The one, camera, light source etc. can be by moving up and down by servo driving, to obtain the image of best sharpness;
The 2nd, adopt dome-type dome diffuse reflective structure light source, avoided the reflective shortcoming of ordinary light source.Colour brightness is controlled, cooperates to be fit to different metal self-adhesion label by software, can automatically generate the optimal illumination parameter, obtains the optimum contrast image;
The 3rd, adopt laser sensor as trigger pip, even the metal self-adhesion label that guarantees thickness 2mm also can stable detection;
The 4th, adopt that template matches and calculation of characteristic parameters are unified to be differentiated, that effectively detects metal self-adhesion label falls word, crooked, skew, warpage defective, replaces manual work fully, and stable performance is respond well.
Therefore, a kind of automatic detection device based on machine vision that the utility model provides can be realized quick, stable online detection, and defective products is got rid of automatically.
Description of drawings
Fig. 1 is the one-piece construction figure of automatic detection device described in the utility model;
Fig. 2 is the process flow diagram of the generation optimal illumination parameter of the utility model embodiment;
Fig. 3 is the Template Learning functional flow diagram of the utility model embodiment;
Fig. 4 is the automatic detection function process flow diagram of the utility model embodiment.
In the accompanying drawing, the list of parts of each label representative is as follows:
1, metal self-adhesion label, 2, industrial computer, 3, lighting source, 4, industrial lens, 5, industrial camera, 6, connecting link, 7, ball screw, 8, the streamline belt, 9, line slide rail, 10, slide block, 11, the ball screw contiguous block, 12, ball screw nut, 13, shaft coupling, 14, step motor, 15, laser sensor, 16, universal installation frame, 17, get rid of cylinder.
Embodiment
Below in conjunction with accompanying drawing principle of the present utility model and feature are described, institute gives an actual example and only is used for explaining the utility model, is not be used to limiting scope of the present utility model.
As shown in Figure 1, the automatic detection device of the described a kind of metal self-adhesion label 1 of the present embodiment comprises lighting source 3, industrial camera 5, streamline belt 8, laser sensor 15, industrial computer 2, step motor 14, gets rid of cylinder 17, ball screw 7 and line slide rail 9;
Described streamline belt 8 is used for transmitting metal self-adhesion label in the described automatic detection device course of work.
Described lighting source 3 is placed between industrial camera 5 and the streamline belt 8, and described industrial camera 5 and described lighting source 3 coaxial cables are perpendicular to streamline belt 8 surfaces, the illumination when described lighting source 3 is used for imaging;
Described industrial camera 5 is connected with the output terminal of described laser sensor 15 by the hardware trigger port that arranges, and is used for obtaining the triggering sensing by described laser sensor 15, realizes optical imagery; Described laser sensor 15 is fixed on the described streamline belt 8 by universal installation frame 16, and the photoelectricity that laser is dropped on the streamline belt is positioned at the partially front place of industrial camera vertical central, for the consistance of the image that guarantees to collect at every turn;
Described industrial computer 2 is respectively with industrial camera 5, step motor 14 be connected cylinder 17 and be connected, and described industrial computer 2 can also comprise the part accessory circuit;
Described step motor 14 drives ball screw 7 by shaft coupling 13 and is rotated, and driving line slide rail 9, described line slide rail 9 is provided with four slide blocks 10, nut 12 on the ball screw makes to rotatablely move with line slide rail 9 and slide block 10 combinations by contiguous block 11 and changes rectilinear motion into, moves up and down thereby drive industrial camera 5 by connecting link 6; Here, the composite set of ball screw nut 12, line slide rail 9 and slide block 10 can be installed on the web joint, drive connecting link 6 by web joint again.
Described eliminating cylinder 17 is installed on the streamline belt, and is connected with described industrial computer 2 by solenoid valve, is used for defective products is got rid of.Except getting rid of cylinder 17, can also adopt other can realize getting rid of the device of the function of defective products.
In the present embodiment, lighting source 3, industrial camera 5 and laser sensor 15 have consisted of image collecting device, and it be used for to realize optical imagery and Collect conversion, and with the image signal transmission that collects to described industrial computer 2.
Described industrial computer 2 namely is the industry control PC, is used for the control of realization system and image processing function.
In addition, step motor 14, ball screw 7, shaft coupling 13, connecting link 6, line slide rail 9 and slide block 10 etc. have consisted of a servo drive, are used for adjusting and driving above-mentioned image collecting device moving up and down.Servo drive drives image collecting device according to metal self-adhesion label thickness difference and moves up and down, to guarantee to obtain best image.
In addition, on the described industrial camera 5 industrial lens 4 is installed, has consisted of an optical imaging device.The present embodiment adopts face battle array 5,000,000 pixel CCD industrial cameras, the IEEE1394 interface, the shortest electronic shutter speed is 30us, frame per second is 6fps, the streamline linear speed is generally per minute 5-10 rice, can satisfy the demand of streamline metal self-adhesion label detection of dynamic, and industrial lens adopts the tight shot of five mega pixel resolution.
Described lighting source 3 adopts dome-type dome diffuse reflective structure, two circle red, green, blues, three paths of LEDs lamps are installed on it, and be provided with light source controller on the described lighting source, be used for independent output triple channel electric current with control red, green, blue three paths of LEDs lamps, described light source controller also is connected with described industrial computer 2, is used for to the colouring information of described industrial computer 2 transmission metal self-adhesion labels.For dome-type dome diffuse reflective structure, namely be to form a dome-type outer cover with the stainless steel punching press, inner spray is with the graininess white paint, in ball-type outer cover bottom one annular LED light circle is installed, the annular external diameter is identical with ball-type outer cover base diameter, internal diameter is less than external diameter 40mm, upper installation two circle LED, LED adopts three kinds of colors of red, green, blue to mix successively arrangement, every kind of separately constant current control of LED, form the light of different colours with control LED luminosity by the control electric current, with the detected metal self-adhesion label of applicable different colours.Particularly, LED is luminous during energising, light upwards and around shine on the graininess white paint of ball-type outer cover again that secondary reflection gets off, form the diffuse illumination object to be detected, avoided the reflective shortcoming of common annular, bar shaped or flat illumination fully.Light source controller can independently be exported the triple channel electric current, with control red, green, blue three paths of LEDs lamps.In addition, light source controller links to each other with industrial computer by the RS232 communication interface, automatically generates the optimal illumination parameter by industrial computer 2 according to metal self-adhesion label color.
Described industrial computer 2 is respectively with industrial camera 5, step motor 14 be connected cylinder 17 and be connected by three PCI slots are set at described industrial computer 2 and realize, described three PCI slots are respectively applied to in-line image capture card, motion control card and digital I/O card, described image pick-up card links to each other with industrial camera 5 by data line, described motion control card is connected with described step motor 14, and described digital I/O cartoon is crossed solenoid valve and is connected with eliminating cylinder 17;
Here, described image pick-up card provides the IEEE1394 interface, link to each other with industrial camera 5 by data line, the parameters such as control every time shutter of camera, gain when system initialization, when starting working, camera obtains the image information that collects, and it is regional that it is deposited specified memory, and the mode with message sends described industrial computer 2 processing to again;
Described digital I/O card drives described eliminating cylinder 17 actions except being used for, and also is used for simultaneously accepting to start stop signal and drives acoustic-optic alarm (described automatic detection device can be established a cover sound and light detecting device certainly).
Described motion control card is used for controlling described step motor 14 by the output pulse, thereby controls further the congregation of camera etc.
Industrial computer described in the present embodiment 2 can realize automatically generating optimal illumination parameter function, Template Learning function or automatic detection function by software form.And the mode of calculation of characteristic parameters and template matches is adopted in the realization of described automatic detection function: described calculation of characteristic parameters comprises the calculating of area, girth, circularity, center of gravity, length and width and the anglec of rotation to the target area; Described template matches refers to seek the target area in image, calculate similarity, differentiates current goal whether identical with target setting.Here, described area is exactly the quantity of pixel in the target area; Described girth refers on the outline line of target area spacing sum between the pixel; Described circularity refers on the basis of area and perimeter, calculates a characteristic quantity of target shape complexity, specifically is calculated as
Be the circle of r for radius, area is π r
2, girth is 2 π r, if shape near circle, then this value is more near 1, shape is more complicated, then should the value less; The mean value of pixel coordinate in the described center of gravity feeling the pulse with the finger-tip mark; Described length and width comprise the length and width of target minimum parallel axle boundary rectangle and the length and width of the minimum boundary rectangle of any direction; The angle of major axis and transverse axis in the oval feature in described anglec of rotation feeling the pulse with the finger-tip mark zone.
The principle of work of the described automatic detection device of the present embodiment is: at the beginning of System Operation, first with the characteristics of image of the metal self-adhesion label 1 that detects and highly wait to arrange and be stored as Parameter File.Access Parameter File before the detection, system carries out initialization, by servo-actuating device camera light source etc. is urged to desired location simultaneously.Metal self-adhesion label 1 flows through from streamline belt 8 during work, responded to and 5 captures of output signal triggering industrial camera by laser sensor 15, and image is transmitted industrial computer 2 process, after finishing dealing with, as then giving a green light by digital I/O card control for non-defective unit, such as the warning that then sends out a warning for defective products, solenoid valve gets electric simultaneously, get rid of cylinder 17 actions, this defective products is released.
Software agent function for the present embodiment mentioned above: automatically generate optimal illumination parameter, Template Learning function and automatic detection function.
Generate optimal illumination parameter function process flow diagram as shown in Figure 2, at first input metal tag prospect and background color.System generates the initialization lighting parameter according to known cold and warm tone principle, be red, green, blue three paths of LEDs drive currents, begin to gather image, image is carried out histogram calculation, calculate its prospect and background central value and deviation separately, then pass through RS232 serial ports output command to light source controller, adjust successively three tunnel driving current value, continue to gather image and calculating prospect and background central value and deviation separately.After No. three driving circuits are all adjusted end in setting range, by more separately central value and deviation numerical value, the different maximum of its center value difference and deviation minimum be that three tunnel driving current value are the optimal illumination parameter.Be in course of adjustment, system selects the specific region in the image to process, and controls simultaneously adjusting range, has greatly shortened the adjustment time.Generally at 2-3 within second, namely capable of regulating obtains the optimal illumination parameter, and then this metalloid self-adhesion label board all throws light on according to this parameter aborning.
The Template Learning functional flow diagram as shown in Figure 3.At first read in image, and image is shown.Then image is carried out medium filtering, utilize the interior gray scale Mesophyticum of neighborhood of pixels for the gray-scale value of this pixel, the salt-pepper noise in the filtering image.Again image is carried out Threshold segmentation, the method of Threshold segmentation is a lot, in the utility model, adopt large Tianjin method to cut apart, specific implementation process is: establishing t is image segmentation threshold, thereby differentiate prospect and the background of image, the total pixel ratio of foreground pixel and image is w0 in the image, and the prospect average gray is u0; The total pixel ratio of background and image is w1 in the image, the background average gray is u1, thereby calculate image averaging gray scale u=w0*u0+w1*u1, inter-class variance g=w0*(u0-u) * (u0-u)+w1*(u1-u) * (u1-u), if t is from 0 to 255, calculate the value of inter-class variance g, the t value that take this value maximum is is as segmentation threshold.With this threshold value image is carried out binary conversion treatment, obtain the new image of a width of cloth.
Secondly, extract exactly character pattern, wherein comprise Mathematical Morphology and process and zone marker.Except character pattern, generally also have a lot of noise spots in the image after the binary conversion treatment.The dilation and erosion mode of processing by Mathematical Morphology in the utility model is removed noise spot.Carry out again zone marker and process, namely by scan image, the pixel that links together is enclosed identical mark.What character patterns are quantity by the statistics mark namely had in the image as can be known.
Again namely carry out the calculating of characteristic parameter.Area, girth, circularity, center of gravity, length and width and the anglec of rotation to the character pattern marked region in the utility model are calculated.The calculating of area namely is the pixel quantity of each marked region of statistics; Circumference calculating then is the outline line that extracts the zone, calculates the length of its outline line; The circularity calculating formula is calculated the complex-shaped degree of marked region on the basis of area and perimeter, in the utility model with following formula
Center of gravity is exactly to ask the mean value of marked region pixel coordinate, utilizes formula in the utility model
The zoning center of gravity; Obtain on the basis according to outer contour in center of gravity, obtain minimum boundary rectangle, thus length and width and the anglec of rotation to zone marker.After all the marked region statistics is finished, all deposit parameters and setting tolerance in Parameter File.
Recovery image then utilizes the marked region that counts on previously to enlarge certain pixel and obtain each corresponding subimage in Recovery image again, and each subimage is carried out edge extracting.Adopt the cany edge detection operator to extract the edge in the utility model, the gradient that it at first calculates G (x, y) obtains the partial differential value f for x and y
xAnd f
y, formula is:
Its gradient amplitude and gradient direction are:
The dual threshold operation is adopted on all limits among the last again refinement M again, obtains the edge of character pattern.Process successively the edge image that obtains each marked region and be stored as template image, at last it is deposited in the Parameter File.Obtain simultaneously in the image a certain zone as the positioning datum image of whole image, deposit in the Parameter File in the lump as template going.
The automatic detection function process flow diagram as shown in Figure 4.In automatic detection function, system is in waiting status, waits for the message that camera hardware triggers.Carry out capture when laser sensor induction metal self-adhesion label triggers camera, and when delivering to internal memory by the IEEE1394 interface, system receives message, namely begins automatic detection function.In the utility model, at first image is carried out medium filtering, remove noise, then adopt the cany edge detection operator to carry out edge extracting.Benchmark image progressive die plate coupling in image, calculate its related function:
To judge in this inside, zone whether be present in the topography that template is consistent, obtain again position and the anglec of rotation of positioning datum image, carry out picture position and angle compensation according to this value.Then according to the quantity of character pattern, call successively the template of each character pattern, in the scope of the regional expansion some pixels of its theory, carry out template matches, obtain again match map the position of image and the anglec of rotation.If its correlativity, position and the anglec of rotation in setting specification, are then carried out following operation, are bad otherwise directly declare this character pattern.Recovery image, passing threshold are cut apart this character of extraction, calculate its area, girth, circularity, center of gravity, length and width and the anglec of rotation, judge mutually with its setting specification again, judge then that as surpassing specification this character pattern is bad.
The above only is preferred embodiment of the present utility model, and is in order to limit the utility model, not all within spirit of the present utility model and principle, any modification of doing, is equal to replacement, improvement etc., all should be included within the protection domain of the present utility model.
Claims (5)
1. the automatic detection device of a metal self-adhesion label is characterized in that, comprises lighting source, industrial camera, streamline belt, laser sensor, industrial computer, step motor, eliminating cylinder, ball screw and line slide rail;
Described lighting source is placed between industrial camera and the streamline belt, and described industrial camera and described lighting source coaxial cable are perpendicular to the streamline belt surface;
Described industrial camera is connected with the output terminal of described laser sensor by the hardware trigger port that arranges, and described laser sensor is fixed on the described streamline belt, and the photoelectricity that laser is dropped on the streamline belt is positioned at the partially front place of industrial camera vertical central;
Described industrial computer is connected with industrial camera, step motor and eliminating cylinder respectively;
Described step motor drives ball screw by shaft coupling and is rotated, and driving line slide rail, described line slide rail is provided with four slide blocks, nut on the ball screw makes to rotatablely move by contiguous block and line slide rail and slide block combination and changes rectilinear motion into, moves up and down thereby drive industrial camera by connecting link;
Described eliminating cylinder is installed on the streamline belt, and is connected with described industrial computer by solenoid valve.
2. automatic detection device according to claim 1 is characterized in that, on the described industrial camera industrial lens is installed.
3. automatic detection device according to claim 1, it is characterized in that, described lighting source adopts dome-type dome diffuse reflective structure, two circle red, green, blues, three paths of LEDs lamps are installed on it, and be provided with light source controller on the described lighting source, be used for independent output triple channel electric current with control red, green, blue three paths of LEDs lamps, described light source controller also is connected with described industrial computer, is used for to the colouring information of described industrial computer transmission metal self-adhesion label.
4. automatic detection device according to claim 1 is characterized in that, described laser sensor is fixed on the described streamline belt by universal installation frame.
5. automatic detection device according to claim 1, it is characterized in that, described industrial computer is connected with industrial camera, step motor and eliminating cylinder respectively by three PCI slots are set at described industrial computer to be realized, described three PCI slots are respectively applied to in-line image capture card, motion control card and digital I/O card, described image pick-up card links to each other with industrial camera by data line, described motion control card is connected with described step motor, and described digital I/O cartoon is crossed solenoid valve and is connected with the eliminating cylinder.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220354805 CN202710486U (en) | 2012-07-20 | 2012-07-20 | Automatic detection device of metal self-adhesive label |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201220354805 CN202710486U (en) | 2012-07-20 | 2012-07-20 | Automatic detection device of metal self-adhesive label |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN202710486U true CN202710486U (en) | 2013-01-30 |
Family
ID=47590777
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201220354805 Expired - Fee Related CN202710486U (en) | 2012-07-20 | 2012-07-20 | Automatic detection device of metal self-adhesive label |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN202710486U (en) |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103134427A (en) * | 2013-03-07 | 2013-06-05 | 苏州吉视电子科技有限公司 | Recognizing device and method for ring parts |
| CN104198492A (en) * | 2014-09-12 | 2014-12-10 | 苏州石丸英合精密机械有限公司 | Belt surface visual contrast detector of belt test machine |
| CN105445286A (en) * | 2015-11-10 | 2016-03-30 | 沈阳建筑大学 | Method and device for detecting quality defects of mark on glass fuse tube products |
| CN106903075A (en) * | 2017-04-26 | 2017-06-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of vamp logo multi-directional visions detection method and system |
| CN107677686A (en) * | 2017-09-28 | 2018-02-09 | 京东方科技集团股份有限公司 | Equipment of the light through window integrating device and using the device |
| CN107806895A (en) * | 2017-12-13 | 2018-03-16 | 张家港市沙源检测技术有限公司 | A kind of and supporting slope mechanism of metal detection instrument |
| CN109030507A (en) * | 2018-08-22 | 2018-12-18 | 合肥通视智能技术有限公司 | A kind of coding detection verifying equipment |
| CN111504192A (en) * | 2020-05-07 | 2020-08-07 | 东华大学 | Compressor appearance detection method based on machine vision |
-
2012
- 2012-07-20 CN CN 201220354805 patent/CN202710486U/en not_active Expired - Fee Related
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103134427A (en) * | 2013-03-07 | 2013-06-05 | 苏州吉视电子科技有限公司 | Recognizing device and method for ring parts |
| CN104198492A (en) * | 2014-09-12 | 2014-12-10 | 苏州石丸英合精密机械有限公司 | Belt surface visual contrast detector of belt test machine |
| CN105445286A (en) * | 2015-11-10 | 2016-03-30 | 沈阳建筑大学 | Method and device for detecting quality defects of mark on glass fuse tube products |
| CN106903075A (en) * | 2017-04-26 | 2017-06-30 | 佛山市南海区广工大数控装备协同创新研究院 | A kind of vamp logo multi-directional visions detection method and system |
| CN107677686A (en) * | 2017-09-28 | 2018-02-09 | 京东方科技集团股份有限公司 | Equipment of the light through window integrating device and using the device |
| CN107677686B (en) * | 2017-09-28 | 2021-01-26 | 京东方科技集团股份有限公司 | Light transmission window integrated device and equipment adopting same |
| CN107806895A (en) * | 2017-12-13 | 2018-03-16 | 张家港市沙源检测技术有限公司 | A kind of and supporting slope mechanism of metal detection instrument |
| CN109030507A (en) * | 2018-08-22 | 2018-12-18 | 合肥通视智能技术有限公司 | A kind of coding detection verifying equipment |
| CN111504192A (en) * | 2020-05-07 | 2020-08-07 | 东华大学 | Compressor appearance detection method based on machine vision |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN202710486U (en) | Automatic detection device of metal self-adhesive label | |
| CN108765416A (en) | PCB surface defect inspection method and device based on fast geometric alignment | |
| US7773214B2 (en) | Apparatus and methods for container inspection | |
| CN100376889C (en) | Intelligent digital graphics context detection system and detection method | |
| CN109001212A (en) | A kind of stainless steel soup ladle defect inspection method based on machine vision | |
| CN208207914U (en) | PCB surface defect detecting device based on fast geometric alignment | |
| CN201434842Y (en) | Automatic chip appearance defect detection system | |
| CN102601063B (en) | Automatic identifying and grading method for bamboo chips | |
| CN106793560A (en) | Circuit board hot-press arrangement and its method with target location offset adjustment function | |
| CN108520260A (en) | The recognition methods of visible foreign matters in bottled oral solution | |
| CN1987435A (en) | Light source configuration method for online detecting system of stip cigarette packing quality | |
| CN114113129B (en) | Lens micro defect recognition and grabbing system and method | |
| CN109410594A (en) | A kind of Vehicle License Plate Recognition System and method | |
| JPH0624424A (en) | Facing/p0sitioning method for container | |
| CN109596058A (en) | The size detection recognition methods of plastic workpiece | |
| CN107246841A (en) | A kind of multiple views grouping system device of auto parts and components | |
| CN209387542U (en) | A kind of detection device for eliminating material shade | |
| CN208752021U (en) | A kind of machine vision device detecting household plate surface defect | |
| CN103091332A (en) | Detection method and detection system of U-shaped powder pipe based on machine vision | |
| CN105312251A (en) | Rhinestone inspection equipment | |
| CN103134427B (en) | Annulus Parts Recognition method | |
| CN113237895B (en) | Metal surface defect detection system based on machine vision | |
| CN205128441U (en) | Removing devices and water brill check -out set | |
| CN113624771A (en) | PCBA rubber coating intelligence visual detection equipment | |
| CN203083937U (en) | Machine vision-based detection system of U-shaped powder tube |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130130 Termination date: 20140720 |
|
| EXPY | Termination of patent right or utility model |