CN102172806B - Image recognition technology based full-automatic welding system and operation method thereof - Google Patents
Image recognition technology based full-automatic welding system and operation method thereof Download PDFInfo
- Publication number
- CN102172806B CN102172806B CN 201010566521 CN201010566521A CN102172806B CN 102172806 B CN102172806 B CN 102172806B CN 201010566521 CN201010566521 CN 201010566521 CN 201010566521 A CN201010566521 A CN 201010566521A CN 102172806 B CN102172806 B CN 102172806B
- Authority
- CN
- China
- Prior art keywords
- image
- solder joint
- welded piece
- welding
- microcomputer
- 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
- 238000003466 welding Methods 0.000 title claims abstract description 146
- 238000000034 method Methods 0.000 title claims abstract description 33
- 238000005516 engineering process Methods 0.000 title claims abstract description 21
- 238000012545 processing Methods 0.000 claims abstract description 62
- 230000003287 optical effect Effects 0.000 claims abstract description 30
- 238000004364 calculation method Methods 0.000 claims abstract description 19
- 229910000679 solder Inorganic materials 0.000 claims description 188
- 230000007246 mechanism Effects 0.000 claims description 48
- 239000007787 solid Substances 0.000 claims description 26
- 230000008439 repair process Effects 0.000 claims description 23
- 230000008859 change Effects 0.000 claims description 6
- 238000010205 computational analysis Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 230000002452 interceptive effect Effects 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 3
- 230000005055 memory storage Effects 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 6
- 238000004519 manufacturing process Methods 0.000 abstract 1
- 230000026676 system process Effects 0.000 abstract 1
- 239000012467 final product Substances 0.000 description 6
- 230000006870 function Effects 0.000 description 4
- 239000003292 glue Substances 0.000 description 3
- 238000012546 transfer Methods 0.000 description 3
- 238000005219 brazing Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 230000008901 benefit Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009394 selective breeding Methods 0.000 description 1
Images
Landscapes
- Length Measuring Devices By Optical Means (AREA)
- Image Processing (AREA)
Abstract
The invention discloses an image recognition technology based full-automatic welding system and an operation method thereof. In the operation method, a microcomputer of the system processes acquisitions acquired by a vertical optical device in two processing modes (namely, a black-and-white processing mode and a color processing mode); in the black-and-white processing mode, pixelated image processing is carried out on cached images by using a binaryzation method; and the color processing mode refers to calculation on each pixel RGB (red, green, blue) chrominance component in the cached images; and finally, through comparing the calculation results obtained by using the black-and-white processing mode and the color processing mode, a welding spot on a welded workpiece is captured automatically, therefore, the full-automatic welding process is really realized, and further, the production efficiency and the welding quality are improved, and the labor intensity is reduced.
Description
Technical field
The present invention relates to a kind of robot welding system, be specifically related to a kind of full automatic welding welding system and operation method thereof based on image recognition technology.
Background technology
Granted publication has been number for the Chinese invention patent of CN100351721C discloses a kind of " welding equipment and operation method thereof with the automatic compute location of micro-image system ", the microoptic device that comprises the microcomputer that has mouse and display screen and be attached thereto, is arranged on three-dimensional servo control mechanism that the CCD camera on the microoptic device eyepiece, three-dimensional servomotor and workbench form, and ultrasonic brazing unit.The CCD camera is sent the welded piece image that the microoptic device amplifies into microcomputer.The automatic governor motion of micro-multiplication factor that microcomputer is directly controlled the microoptic device is determined the multiplication factor of microoptic device.Three-dimensional servomotor in the three-dimensional servo control mechanism is according to the moving working table movement of microcomputer instruction tape.Microcomputer control ultrasonic brazing unit is finished backguy and ultrasonic bonding.When first welded piece welding, the employing mouse is clicked the solder joint position of determining the welded piece image one by one and is deposited microcomputer on screen, the position that microcomputer is clicked according to mouse, calculate each solder joint and capillary and align when reaching the welding position the required curve movement of three-dimensional servo control mechanism, control three-dimensional servo control mechanism and drive the welded piece motion in proper order, make each solder joint reach welding position, welded piece orientation of living in, control welder successively to finish each required job step of welding; When welding the welded piece of same size again, click the current welded piece image-based of screen with mouse earlier and determine welded piece to be welded orientation on schedule, microcomputer calls the related data of first welded piece of storage automatically, calculates each solder joint data of welded piece to be welded automatically and finishes the location welding.Determine the solder joint position of welded piece image because the said equipment when changing different welded pieces, all needs the workman to click one by one again with mouse at every turn on screen, and can not realize the automatic seizure of solder joint; Thereby in specific operation process, following problem will occur: number of welds on the welded piece solid space is generally all many, area is less relatively owing to be distributed in, and this just requires the workman to click energy high concentration when determining bond pad locations carrying out mouse.If careless slightly, just the situation of leak source or one of wrong point or even a plurality of solder joints appears easily.This has not only increased working strength of workers, has reduced operating efficiency; And in case when leak source or wrong spot welding point occurring, each solder joint that microcomputer calculated and capillary align when reaching the welding position the required curve movement of three-dimensional servo control mechanism just deviation can occur, and then influence whole welding process, cause its welding result appearance mistake, welding quality to be difficult to guarantee.
Summary of the invention
Technical problem to be solved by this invention is the deficiency that existing robot welding system inefficiency, welding quality are difficult to guarantee, a kind of full automatic welding welding system and operation method thereof based on image recognition technology is provided.
For addressing the above problem, a kind of full automatic welding welding system that the present invention is designed based on image recognition technology, mainly three-dimensional servo control mechanism, vertical optical device and the welder that links to each other by microcomputer and with microcomputer constituted; Wherein have keyboard, mouse and display screen on the microcomputer; Three-dimensional servo control mechanism comprises the fixedly workbench and the three-dimensional servomotor of controlling working table movement of welded piece; The operating surface of fixing welded piece is relative on vertical optical device and the three-dimensional servo control mechanism, and it mainly is made up of with the vertical CCD camera on being arranged on microoptic device eyepiece the microoptic device; The vertical CCD camera of vertical optical device is sent the welded piece image that the microoptic device amplifies into microcomputer;
Select system's relevant parameter is preset by keyboard input or mouse, and click a solder joint on the welded piece solid space as the original bench mark point with mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default;
Microcomputer is according to selected path of welding and above-mentioned solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position; Controlling three-dimensional servo control mechanism drives welded piece motion in proper order, makes each solder joint arrive successively that welding position, welded piece are in required orientation, the control welder is finished each required job step of welding;
In addition, native system also further includes Z axle horizontal optical system, and this Z axle horizontal optical system comprises calibration light source and horizontal CCD camera, and is located at the Z axle horizontal correcting unit in microcomputer; The horizontal CCD camera links to each other with Z axle horizontal correcting unit in the microcomputer; All the rectifying plane with welded piece is relative with the horizontal CCD camera for calibration light source, its Z axle horizontal optical system to the rectifying plane of welded piece and vertical optical device perpendicular to the operating surface of welded piece;
When fixing first welded piece, guarantee that the operating surface of welded piece is in level; The horizontal CCD camera is sent the image of the rectifying plane that obtained into microcomputer, and shows via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane; Again fixedly during the welded piece of same size, microcomputer is caught image residing Y-axis bound scope on display screen of the rectifying plane of welded piece to be corrected automatically, and itself and N Reference Alignment scope compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.
In the such scheme, by a keyboard entry or the parameter preset selected of mouse include the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, magnitude range, solder joint recognition mode and/or the path of welding of each grabgraf.
Be provided with central processing unit, welding control unit, three-dimensional motion control module, micro-multiplication factor record cell, solder joint picture storing unit, solder joint image identification unit, solder joint image comparison analytic unit, workpiece solder joint positioning unit, bond pad locations memory cell and computational analysis unit in the described microcomputer of such scheme;
Central processing unit is finished the core control of whole microcomputer;
Welding control unit control welder is finished each required job step of welding;
The three-dimensional motion control module is controlled the operation of three-dimensional servomotor;
The multiplication factor of micro-multiplication factor recording unit records storage microoptic device;
Solder joint picture storing unit stores the image information of the welded piece that vertical CCD camera obtains;
The solder joint image identification unit is used to automatically identify the solder joint on the image of welded piece of solder joint picture storing element memory storage;
Solder joint image comparison analytic unit is compared the solder joint that the solder joint image identification unit identifies with solder joint parameter information by a keyboard entry, filtering interfering information keeps effective solder joint information;
Workpiece solder joint positioning unit positions storage with effective solder joint information that solder joint image comparison analytic unit retains;
Position data after locate according to each solder joint of bond pad locations memory cell stores the computational analysis unit is calculated the welded piece curve movement, sends into the three-dimensional motion control module, sends instruction to three-dimensional servo control mechanism.
As the further improvement of such scheme, also be provided with the repair welding unit in the described microcomputer, this repair welding unit links to each other with welder and three-dimensional servo control mechanism.
The operation method of a kind of full automatic welding welding system based on image recognition technology of the present invention comprises the steps:
The a welded piece is fixed on the workbench of three-dimensional servo control mechanism, and vertical CCD camera is sent welded piece enlarged image data into the microcomputer image-display units and shown on the screen of microcomputer;
B selects by keyboard input or mouse system's relevant parameter is preset, and selects promptly to lock after the micro-multiplication factor according to welded piece size adjustment to be welded simultaneously, and this multiplication factor is stored at the multiplication factor record cell simultaneously;
C catches all solder joints on the welded piece solid space; Promptly the process that all solder joints on the welded piece solid space are caught is specially: select system's relevant parameter is preset by keyboard input or mouse, and click a solder joint on the welded piece solid space as the original bench mark point with mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default;
D deposits all bond pad locations on the welded piece solid space in the bond pad locations memory cell;
E Microcomputer Calculation analytic unit is according to the selected welding manner and the solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position;
F central processing unit notice motion control unit sends instruction to three-dimensional servo control mechanism, and three-dimensional servo control mechanism is delivered to welding position under the capillary according to instruction with the solder joint order;
After the three-dimensional servo control mechanism of g is whenever finished moving into place of a solder joint, beam back signal to central processing unit, central processing unit notice welding control unit sends instruction to welder, carry out welding procedure and finish each required job step of welding, to weld end signal after welding is finished and send central processing unit back to, continue data call, location and the welding of next solder joint; All solder joint welding to this welded piece solid space finish;
In addition, this method also further includes following welded piece level correction step, when promptly fixing first welded piece, guarantees that the operating surface of welded piece is in level; The horizontal CCD camera is sent the image of the rectifying plane that obtained into microcomputer, and shows via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane; Again fixedly during the welded piece of same size, microcomputer is caught image residing Y-axis bound scope on display screen of the rectifying plane of welded piece to be corrected automatically, and itself and N Reference Alignment scope compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.
Among the above-mentioned steps b by a keyboard entry or the parameter preset selected of mouse include the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, magnitude range, solder joint recognition mode and/or the path of welding of each grabgraf.
As the further improvement of such scheme, the operation method of described full automatic welding welding system based on image recognition technology comprises that also promising selected solder joint or welding carry out the repair welding step.
Compared with prior art, the present invention has following characteristics:
1, based on the image automatic identification technology, by each solder joint on automatic seizure of microcomputer and the location welded piece solid space, having automated graphics simultaneously calculates, analyzes and the control positioning function, control system is directly obtained data from micro-image by optical system, chip and the support solder joint coordinate to be welded to each part calculate immediately, control three-dimensional servo control mechanism again and solder joint is delivered to are carried out automation welding under the capillary; Need not the workman and manually catch and tack weld, thus truly the full automatic welding termination process that has realized;
2, Z axle horizontal optical system can be preserved calibration parameter for the first time automatically, so that can carrying out automatic horizontal to the welded piece that is fixed on the workbench once more and have a same size, system proofreaies and correct, so both reduce working strength of workers, improved operating efficiency, improved the precision of the correction of welded piece again;
3, the automatic repair welding function that possesses of system can reduce because of solder joint has the glue welding and do not reach the standard grade or the occurrence probability of problem such as bonding wire fracture.
Description of drawings
Fig. 1 is the structured flowchart of a kind of full automatic welding welding system based on image recognition technology of the present invention;
Fig. 2 is the inner theory diagram of microcomputer;
Fig. 3 is the flow chart of the operation method of a kind of full automatic welding welding system based on image recognition technology of the present invention.
The specific embodiment
The structured flowchart of a kind of full automatic welding welding system based on image recognition technology of the present invention as shown in Figure 1, its three-dimensional servo control mechanism that mainly links to each other by microcomputer and with microcomputer, vertical optical device and welder are constituted; Wherein have keyboard, mouse and display screen on the microcomputer.Three-dimensional servo control mechanism comprises the fixedly workbench and the three-dimensional servomotor of controlling working table movement of welded piece.The operating surface of fixing welded piece is relative on vertical optical device and the three-dimensional servo control mechanism, and it mainly is made up of with the vertical CCD camera on being arranged on microoptic device eyepiece the microoptic device.The vertical CCD camera of vertical optical device is sent the welded piece image that the microoptic device amplifies into microcomputer.Keyboard and mouse are used for the predetermined system relevant parameter, and click a solder joint on the welded piece solid space as the original bench mark point with mouse.
Be provided with central processing unit, welding control unit, three-dimensional motion control module, micro-multiplication factor record cell, solder joint picture storing unit, solder joint image identification unit, solder joint image comparison analytic unit, workpiece solder joint positioning unit, bond pad locations memory cell and computational analysis unit in the above-mentioned microcomputer.Central processing unit is finished the core control of whole microcomputer.Welding control unit control welder is finished each required job step of welding.The three-dimensional motion control module is controlled the operation of three-dimensional servomotor.The multiplication factor of micro-multiplication factor recording unit records storage microoptic device.Solder joint picture storing unit stores the image information of the welded piece that vertical CCD camera obtains.The solder joint image identification unit is used to automatically identify the solder joint on the image of welded piece of solder joint picture storing element memory storage.Solder joint image comparison analytic unit is compared the solder joint that the solder joint image identification unit identifies with solder joint parameter information by a keyboard entry, filtering interfering information keeps effective solder joint information.Workpiece solder joint positioning unit positions storage with effective solder joint information that solder joint image comparison analytic unit retains.Position data after locate according to each solder joint of bond pad locations memory cell stores the computational analysis unit is calculated the welded piece curve movement, sends into the three-dimensional motion control module, sends instruction to three-dimensional servo control mechanism.Referring to Fig. 2.
Select system's relevant parameter is preset by keyboard input or mouse, and click a solder joint on the welded piece solid space as the original bench mark point with mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default.Microcomputer is according to the selected welding manner and the solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position.Controlling three-dimensional servo control mechanism drives welded piece motion in proper order, makes each solder joint arrive successively that welding position, welded piece are in required orientation, the control welder is finished each required job step of welding.
The parameter preset that by a keyboard entry above-mentioned or mouse is selected includes the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, magnitude range, solder joint recognition mode and/or the path of welding of each grabgraf.Wherein the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, the solder joint parameter informations such as magnitude range of each grabgraf are used for filtering interfering, improve the automatic seizure precision of solder joint; These parameter informations not only can be keyed in before system's operation, and in system's running, also can finely tune according to the solder joint parameter information of having caught that shows on the screen, improve precision and efficient that solder joint is caught thus.Carrying out concrete solder joint when catching, we both can progressively catch automatically to all solder joints by selecting the solder joint recognition mode; Also can be according to the Given information of device to be welded on the welded piece solid space, extrapolate other solder joints on this device by a solder joint in the capture device.As the device on the known welded piece solid space is diode chip for backlight unit, and the distance between two utmost points of known diode chip for backlight unit, when microcomputer captures bond pad locations information of this diode chip for backlight unit automatically like this, just can extrapolate the positional information of another solder joint by the distance between 2 solder joints on the diode chip for backlight unit.Path of welding then is the connected mode that is used to limit between the solder joint, with different electric connecting modes such as the parallel connection that obtains chip and/or series connection.In the preferred embodiment of the present invention, path of welding comprises measures enlargement ratio and compiling data file.Measuring enlargement ratio is to provide a fixed value at present procedure present image is measured, as the reference data of the automatic location Calculation in back.Native system X, Y coordinate have independently enlargement ratio measurement function, make it accurately higher.The compiling data file is preset the soldered wires mode of chip on the welded piece solid space, preserves so that directly transfer when using same model in the future and gets final product (measuring under the constant situation of enlargement ratio).
In order to guarantee welding effect, the solder side that is fixed on the welded piece on the workbench must be in level.Therefore the present invention also further includes Z axle horizontal optical system, and this Z axle horizontal optical system comprises calibration light source and horizontal CCD camera, and is located at the Z axle horizontal correcting unit in microcomputer; The horizontal CCD camera links to each other with Z axle horizontal correcting unit in the microcomputer; All the rectifying plane with welded piece is relative with the horizontal CCD camera for calibration light source, its Z axle horizontal optical system to the rectifying plane of welded piece and vertical optical device perpendicular to the operating surface of welded piece.When fixing first welded piece, guarantee that the operating surface of welded piece is in level; The horizontal CCD camera is sent the image of the rectifying plane that obtained into microcomputer, and shows via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane; Again fixedly during the welded piece of same size, microcomputer is caught image residing Y-axis bound scope on display screen of the rectifying plane of welded piece to be corrected automatically, and itself and N Reference Alignment scope compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.The effect of Z axle automatic horizontal calibration is: for the first time fixedly during welded piece, the operating surface of guaranteeing welded piece by naked eyes is in level, and the related levels parameter information preserved, can directly transfer the automatic calibration that correspondingly horizontal parameter information can be realized welded piece when re-using the welded piece of same specification after like this.
For avoiding not reaching the standard grade or the generation of problem such as bonding wire broken string because of solder joint has glue welding, the present invention also disposes the repair welding unit that also is provided with in the microcomputer to selected solder joint or welding repair welding, and this repair welding unit links to each other with welder and three-dimensional servo control mechanism.According to program setting, the repair welding unit can be selected the tip-in pattern or mend ray mode.When carrying out welding, can select after the first tip-in tip-in pattern behind bonding wire pattern or the first bonding wire.Mend ray mode following time when the repair welding unit is in, the workman moves to a solder joint with cursor, pins left mouse button and does not put to be dragged to and unclamp mouse behind two solder joints and get final product.When the repair welding unit was in repair welding ball pattern, the solder joint that mouse the is moved to required repair welding left button of clicking the mouse got final product.
The flow chart of the operation method of a kind of full automatic welding welding system based on image recognition technology of the present invention as shown in Figure 3, specifically includes following steps:
The a welded piece is fixed on the workbench of three-dimensional servo control mechanism, and vertical CCD camera is sent welded piece enlarged image data into the microcomputer image-display units and shown on the screen of microcomputer;
In order to satisfy welding requirements, to guarantee that the solder side that is fixed on the welded piece on the workbench must be in level.Welded piece is fixed on and fixedly is divided into two kinds of situations: when the welded piece on being fixed on workbench is new spec, needs artificial naked eyes to proofread and correct the operating surface of guaranteeing welded piece and be in level; This moment, the horizontal CCD camera was sent the image of the rectifying plane that obtained into microcomputer, and showed via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane and preserves.Welded piece on being fixed on workbench with before the welded piece fixed when consistent, need microcomputer guarantee that automatically the operating surface of welded piece is in level; This moment is by artificial selection welded piece type, microcomputer is transferred the N Reference Alignment scope, and obtain the image of the rectifying plane of this welded piece by the horizontal CCD camera, the image of this rectifying plane residing Y-axis bound scope and N Reference Alignment scope on display screen are compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.
Before b welds the welded piece of a certain new spec, at first to come the relevant parameter of system is preset, as the diameter of the shared territory value range of the bound of solder joint area, the solder joint that is captured, solder joint, magnitude range, solder joint recognition mode and/or the path of welding etc. of each grabgraf by mouse or keyboard; Simultaneously, the workman selects promptly to lock after the micro-multiplication factor according to welded piece size adjustment to be welded, and this multiplication factor is stored at the multiplication factor record cell simultaneously.
Wherein the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, the solder joint parameter informations such as magnitude range of each grabgraf are used for filtering interfering, improve the automatic seizure precision of solder joint; These parameter informations not only can be keyed in before system's operation, and in system's running, also can finely tune according to the solder joint parameter information of having caught that shows on the screen, improve precision and efficient that solder joint is caught thus.Carrying out concrete solder joint when catching, we both can progressively catch automatically to all solder joints by selecting the solder joint recognition mode; Also can be according to the Given information of device to be welded on the welded piece solid space, extrapolate other solder joints on this device by a solder joint in the capture device.As the device on the known welded piece solid space is diode chip for backlight unit, and the distance between two utmost points of known diode chip for backlight unit, when microcomputer captures bond pad locations information of this diode chip for backlight unit automatically like this, just can extrapolate the positional information of another solder joint by the distance between 2 solder joints on the diode chip for backlight unit.Path of welding then is the connected mode that is used to limit between the solder joint, with different electric connecting modes such as the parallel connection that obtains chip and/or series connection.In the preferred embodiment of the present invention, path of welding comprises measures enlargement ratio and compiling data file.Measuring enlargement ratio is to provide a fixed value at present procedure present image is measured, as the reference data of the automatic location Calculation in back.Native system X, Y coordinate have independently enlargement ratio measurement function, make it accurately higher.The compiling data file is preset the soldered wires mode of chip on the welded piece solid space, preserves so that directly transfer when using same model in the future and gets final product (measuring under the constant situation of enlargement ratio).
C catches all solder joints on the welded piece solid space; Promptly the process that all solder joints on the welded piece solid space are caught is specially: select system's relevant parameter is preset by keyboard input or mouse, and click a solder joint on the welded piece solid space as the original bench mark point with mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default;
D deposits all bond pad locations on the welded piece solid space in the bond pad locations memory cell;
E Microcomputer Calculation analytic unit is according to the selected welding manner and the solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position;
F central processing unit notice motion control unit sends instruction to three-dimensional servo control mechanism, and three-dimensional servo control mechanism is delivered to welding position under the capillary according to instruction with the solder joint order;
After the three-dimensional servo control mechanism of g is whenever finished moving into place of a solder joint, beam back signal to central processing unit, central processing unit notice welding control unit sends instruction to welder, carry out welding procedure and finish each required job step of welding, to weld end signal after welding is finished and send central processing unit back to, continue data call, location and the welding of next solder joint; All solder joint welding to this welded piece solid space finish.
For avoiding not reaching the standard grade or the generation of problem such as bonding wire broken string because of solder joint has glue welding, the repair welding step that the present invention also carries out repair welding for selected solder joint or bonding wire.Described repair welding step is according to program setting, and the repair welding unit can select single tip-in pattern, the single benefit to mend ray mode after ray mode, the first tip-in, mend tip-in pattern behind the line earlier ...Mend ray mode following time when the repair welding unit is in, the workman moves to a solder joint with cursor, pins left mouse button and does not put to be dragged to and unclamp mouse behind two solder joints and get final product.When the repair welding unit was in repair welding ball pattern, the solder joint that mouse the is moved to required repair welding left button of clicking the mouse got final product.
Claims (7)
1. based on the full automatic welding welding system of image recognition technology, mainly three-dimensional servo control mechanism, vertical optical device and the welder that links to each other by microcomputer and with microcomputer constituted; Wherein have keyboard, mouse and display screen on the microcomputer; Three-dimensional servo control mechanism comprises the fixedly workbench and the three-dimensional servomotor of controlling working table movement of welded piece; The operating surface of fixing welded piece is relative on vertical optical device and the three-dimensional servo control mechanism, and it mainly is made up of with the vertical CCD camera on being arranged on microoptic device eyepiece the microoptic device; The vertical CCD camera of vertical optical device is sent the welded piece image that the microoptic device amplifies into microcomputer;
Select system's relevant parameter is preset by keyboard input or mouse, and click a solder joint on the welded piece solid space as the original bench mark point with mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default;
Microcomputer is according to selected path of welding and above-mentioned solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position; Controlling three-dimensional servo control mechanism drives welded piece motion in proper order, makes each solder joint arrive successively that welding position, welded piece are in required orientation, the control welder is finished each required job step of welding;
It is characterized in that:
Also further include Z axle horizontal optical system, this Z axle horizontal optical system comprises calibration light source and horizontal CCD camera, and is located at the Z axle horizontal correcting unit in microcomputer; The horizontal CCD camera links to each other with Z axle horizontal correcting unit in the microcomputer; All the rectifying plane with welded piece is relative with the horizontal CCD camera for calibration light source, its Z axle horizontal optical system to the rectifying plane of welded piece and vertical optical device perpendicular to the operating surface of welded piece;
When fixing first welded piece, guarantee that the operating surface of welded piece is in level; The horizontal CCD camera is sent the image of the rectifying plane that obtained into microcomputer, and shows via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane; Again fixedly during the welded piece of same size, microcomputer is caught image residing Y-axis bound scope on display screen of the rectifying plane of welded piece to be corrected automatically, and itself and N Reference Alignment scope compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.
2. according to the described full automatic welding welding system of claim 1, it is characterized in that based on image recognition technology: by a keyboard entry or the parameter preset selected of mouse include the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, magnitude range, solder joint recognition mode and/or the path of welding of each grabgraf.
3. according to the described full automatic welding welding system of claim 1, it is characterized in that: be provided with central processing unit, welding control unit, three-dimensional motion control module, micro-multiplication factor record cell, solder joint picture storing unit, solder joint image identification unit, solder joint image comparison analytic unit, workpiece solder joint positioning unit, bond pad locations memory cell and computational analysis unit in the microcomputer based on image recognition technology;
Central processing unit is finished the core control of whole microcomputer;
Welding control unit control welder is finished each required job step of welding;
The three-dimensional motion control module is controlled the operation of three-dimensional servomotor;
The multiplication factor of micro-multiplication factor recording unit records storage microoptic device;
Solder joint picture storing unit stores the image information of the welded piece that vertical CCD camera obtains;
The solder joint image identification unit is used to automatically identify the solder joint on the image of welded piece of solder joint picture storing element memory storage;
Solder joint image comparison analytic unit is compared the solder joint that the solder joint image identification unit identifies with solder joint parameter information by a keyboard entry, filtering interfering information keeps effective solder joint information;
Workpiece solder joint positioning unit positions storage with effective solder joint information that solder joint image comparison analytic unit retains;
Position data after locate according to each solder joint of bond pad locations memory cell stores the computational analysis unit is calculated the welded piece curve movement, sends into the three-dimensional motion control module, sends instruction to three-dimensional servo control mechanism.
4. according to any described full automatic welding welding system based on image recognition technology in the claim 1～3, it is characterized in that: also be provided with the repair welding unit in the described microcomputer, this repair welding unit links to each other with welder and three-dimensional servo control mechanism.
5. based on the operation method of the full automatic welding welding system of image recognition technology, comprise the steps:
The a welded piece is fixed on the workbench of three-dimensional servo control mechanism, and vertical CCD camera is sent welded piece enlarged image data into the microcomputer image-display units and shown on the screen of microcomputer;
B selects by keyboard input or mouse system's relevant parameter is preset, and selects promptly to lock after the micro-multiplication factor according to welded piece size adjustment to be welded simultaneously, and this multiplication factor is stored at the multiplication factor record cell simultaneously;
C catches all solder joints on the welded piece solid space; Promptly select system's relevant parameter is preset, and click a solder joint on the welded piece solid space as the original bench mark point with mouse by keyboard input or mouse; The vertical optical device carries out IMAQ to welded piece, gather image at every turn after, microcomputer is this image application buffer memory; Image to above-mentioned buffer memory adopts black and white and colored two kinds of processing modes, and wherein the black and white processing mode is with binarization method cache image to be carried out pixel image to handle, and color processing mode then is that each the pixel RGB chrominance component in the cache image is calculated; Result of calculation with black and white processing mode and color processing mode gained compares at last, if both result of calculation unanimities are caught image and finished, obtains the solder joint coordinate on this cache image; Otherwise will gather image once more, and repeat the black and white of cache image and the comparison of color processing and image; Do not capture the solder joint image yet as catching in the number of times, will change the artificial solder joint coordinate of determining automatically in the circulation of default;
D deposits all bond pad locations on the welded piece solid space in the bond pad locations memory cell;
E Microcomputer Calculation analytic unit is according to the selected welding manner and the solder joint coordinate information of catching, and calculates the required curve movement of three-dimensional servo control mechanism when the position that the capillary of each solder joint and welder aligns arrives the welding position;
F central processing unit notice motion control unit sends instruction to three-dimensional servo control mechanism, and three-dimensional servo control mechanism is delivered to welding position under the capillary according to instruction with the solder joint order;
After the three-dimensional servo control mechanism of g is whenever finished moving into place of a solder joint, beam back signal for the central processing unit of microcomputer, central processing unit notice welding control unit sends instruction to welder, carry out welding procedure and finish each required job step of welding, to weld end signal after welding is finished and send central processing unit back to, continue data call, location and the welding of next solder joint; All solder joint welding to this welded piece solid space finish;
It is characterized in that:
Also further include welded piece level correction step, when promptly fixing first welded piece, guarantee that the operating surface of welded piece is in level; The horizontal CCD camera is sent the image of the rectifying plane that obtained into microcomputer, and shows via display screen; The mouse frame is chosen the highlighted part of being shone by light source in the image of stating rectifying plane, is used as the N Reference Alignment scope with highlighted part residing Y-axis bound scope on display screen in the image of determining this rectifying plane; Again fixedly during the welded piece of same size, microcomputer is caught image residing Y-axis bound scope on display screen of the rectifying plane of welded piece to be corrected automatically, and itself and N Reference Alignment scope compared, when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops in the N Reference Alignment scope on display screen, then microcomputer is judged welded piece level to be corrected; And when the image of the rectifying plane of welded piece to be corrected when residing Y-axis bound scope drops on outside the N Reference Alignment scope on display screen, then microcomputer judges that welded piece to be corrected is out-of-level, and microcomputer is controlled three-dimensional servo control mechanism and driven the operating surface that welded piece rotates to welded piece to be corrected and be horizontal.
6. according to the operation method of the described full automatic welding welding system based on image recognition technology of claim 5, it is characterized in that: among the described step b by a keyboard entry or the parameter preset selected of mouse include the shared territory value range of the bound of solder joint area, the solder joint that is captured, the diameter of solder joint, magnitude range, solder joint recognition mode and/or the path of welding of each grabgraf.
7. according to the operation method of claim 5 or 6 described full automatic welding welding systems based on image recognition technology, it is characterized in that also comprising that promising selected solder joint or welding carry out the repair welding step.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010566521 CN102172806B (en) | 2010-12-01 | 2010-12-01 | Image recognition technology based full-automatic welding system and operation method thereof |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201010566521 CN102172806B (en) | 2010-12-01 | 2010-12-01 | Image recognition technology based full-automatic welding system and operation method thereof |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102172806A CN102172806A (en) | 2011-09-07 |
| CN102172806B true CN102172806B (en) | 2013-07-31 |
Family
ID=44516151
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201010566521 Expired - Fee Related CN102172806B (en) | 2010-12-01 | 2010-12-01 | Image recognition technology based full-automatic welding system and operation method thereof |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102172806B (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021127762A1 (en) * | 2019-12-23 | 2021-07-01 | Petróleo Brasileiro S.A. - Petrobras | Computer method for automatic correction of welding |
Families Citing this family (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103418944B (en) * | 2012-05-21 | 2016-08-17 | 上海航天设备制造总厂 | A kind of image identification system for solar array automatic welding |
| CN111570973A (en) * | 2013-03-14 | 2020-08-25 | 林肯环球股份有限公司 | System and method for exporting or using welding sequencer data for external systems |
| CN105160645B (en) * | 2015-01-24 | 2017-09-19 | 六安志成智能科技有限公司 | Track the welding torch displacement control system of position while welding |
| CN104766333A (en) * | 2015-04-20 | 2015-07-08 | 吉林大学 | Vehicle door point welding robot path correction method based on stereoscopic vision |
| CN105500693A (en) * | 2016-01-29 | 2016-04-20 | 上海甬兴塑胶有限公司 | Ultrasonic wave welding device for injection molding workpiece |
| CN106780426B (en) * | 2016-09-21 | 2023-05-23 | 南京师范大学 | Integrated circuit board multi-form welding spot positioning method based on color feature model |
| CN107511576B (en) * | 2017-08-18 | 2019-12-10 | 湖南军成科技有限公司 | Method and system for positioning welding point |
| CN111014995B (en) * | 2018-10-09 | 2021-10-08 | 中冶赛迪工程技术股份有限公司 | Robot welding method and system for nonstandard unstructured operation environment |
| CN114697529B (en) * | 2020-12-30 | 2024-05-28 | 富联裕展科技(深圳)有限公司 | Visual information self-adaptive adjustment method and device, welding equipment and storage medium |
| CN114772592B (en) * | 2022-06-21 | 2022-09-16 | 成都沃特塞恩电子技术有限公司 | Diamond cultivation equipment adjusting method and device, electronic equipment and storage medium |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101050961A (en) * | 2007-04-30 | 2007-10-10 | 西南林学院 | Method for fine correcting satellite remote sensing image geometry based on topographic line |
| CN100351721C (en) * | 2005-07-05 | 2007-11-28 | 李光 | Welding apparatus with micropicture automatic-computing positioning system and operation mode |
| CN100394266C (en) * | 2005-03-24 | 2008-06-11 | 爱普生映像元器件有限公司 | Image display device and electronic apparatus |
-
2010
- 2010-12-01 CN CN 201010566521 patent/CN102172806B/en not_active Expired - Fee Related
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN100394266C (en) * | 2005-03-24 | 2008-06-11 | 爱普生映像元器件有限公司 | Image display device and electronic apparatus |
| CN100351721C (en) * | 2005-07-05 | 2007-11-28 | 李光 | Welding apparatus with micropicture automatic-computing positioning system and operation mode |
| CN101050961A (en) * | 2007-04-30 | 2007-10-10 | 西南林学院 | Method for fine correcting satellite remote sensing image geometry based on topographic line |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021127762A1 (en) * | 2019-12-23 | 2021-07-01 | Petróleo Brasileiro S.A. - Petrobras | Computer method for automatic correction of welding |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102172806A (en) | 2011-09-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102172806B (en) | Image recognition technology based full-automatic welding system and operation method thereof | |
| CN201906929U (en) | Full-automatic welding system based on picture identification technology | |
| CN107584263B (en) | A kind of vision positioning guide device and bootstrap technique | |
| US9959451B2 (en) | Image inspection device, image inspection method and image inspection program | |
| CN204944428U (en) | For detecting the equipment that there is at least one and connect artifact sample | |
| CN100447525C (en) | Measuring method of image measuring instrument | |
| CN105025290B (en) | A kind of inclined method between adjust automatically camera module sensor and camera lens | |
| CN104384762B (en) | A kind of welding machine mobile control system and control method | |
| CN110137098B (en) | Method and system for inspecting wafer defects | |
| KR20160107128A (en) | Vision system for training an assembly system through virtual assembly of objects | |
| CN103231162A (en) | Device and method for visual detection of welding quality of robot | |
| CN110064819A (en) | The extraction of cylinder longitudinal seam characteristic area, welding seam tracking method and system based on structure light | |
| KR100805486B1 (en) | A system and a method of measuring a display at multi-angles | |
| CN106468572B (en) | Articulated robot dynamic property evaluation system | |
| CN110695520B (en) | Vision-based full-automatic galvanometer field calibration system and calibration method thereof | |
| CN104708158A (en) | Automatic circuit board welding method | |
| CN111161208A (en) | Image detection method and device | |
| CN112964721A (en) | Multi-target random array flying shooting visual detection method and detection system | |
| CN109447941B (en) | Automatic registration and quality detection method in welding process of laser soldering system | |
| US20020040922A1 (en) | Multi-modal soldering inspection system | |
| CN109894776A (en) | The automatic compensating method of seam track | |
| CN112730442A (en) | Automatic online product surface defect detection device and system based on machine vision | |
| JP6140255B2 (en) | Image processing apparatus and image processing method | |
| CN110961778A (en) | Method for automatically identifying welding area of welding workpiece, computer device and computer-readable storage medium | |
| JP3680785B2 (en) | Electronic component bonding apparatus and bonding method |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20130731 |