CN105149794A - Intelligent laser trimming system and method based on binocular vision - Google Patents
Intelligent laser trimming system and method based on binocular vision Download PDFInfo
- Publication number
- CN105149794A CN105149794A CN201510509465.4A CN201510509465A CN105149794A CN 105149794 A CN105149794 A CN 105149794A CN 201510509465 A CN201510509465 A CN 201510509465A CN 105149794 A CN105149794 A CN 105149794A
- Authority
- CN
- China
- Prior art keywords
- image
- binocular vision
- cutting
- video camera
- vision
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/36—Removing material
- B23K26/38—Removing material by boring or cutting
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
- B23K—SOLDERING OR UNSOLDERING; WELDING; CLADDING OR PLATING BY SOLDERING OR WELDING; CUTTING BY APPLYING HEAT LOCALLY, e.g. FLAME CUTTING; WORKING BY LASER BEAM
- B23K26/00—Working by laser beam, e.g. welding, cutting or boring
- B23K26/02—Positioning or observing the workpiece, e.g. with respect to the point of impact; Aligning, aiming or focusing the laser beam
- B23K26/03—Observing, e.g. monitoring, the workpiece
- B23K26/032—Observing, e.g. monitoring, the workpiece using optical means
Abstract
The invention discloses an intelligent laser trimming system and method based on binocular vision. The system comprises an industrial personal computer, a scanner and a box-type cutting platform. A mechanical hand, servo motors, a vision moving platform, a lighting system and a CCD binocular vision camera are installed in the box-type cutting platform. The industrial personal computer controls the CCD binocular vision camera installed on the vision moving platform to move through the servo motors. The mechanical hand is connected with the industrial personal computer through an arranged mechanical hand control cabinet. The vision moving platform is connected with the industrial personal computer through an arranged vision moving platform control cabinet. The system is high in intelligence degree, the automatic calibration and automatic scanned drawing programming functions are achieved, and production efficiency is greatly improved.
Description
Technical field
The present invention relates to a kind of Intelligent Laser trimming system based on binocular vision and method thereof, belong to image procossing, visual pattern process and manipulator laser cutting technique field.
Background technology
The cast member thicker for burr or material is harder, burr removing method conventional is at present manipulator laser cutting.Conventional laser cutting method has teaching manipulator to cut and laser-vision sensing.
Teaching manipulator cutter utilizes that the teaching machine of manipulator carries out the coordinate of cut point, attitude is searched one by one, recorded to realize cutting operation.Not only efficiency is low but also dependence human eye judges that cutting accuracy is difficult to be guaranteed for this method.
Laser-vision sensing has initiative, noncontact, the three-dimensional information that can obtain object, precision advantages of higher.But laser vision system cost is higher, is used in cutting surplus and only need ensures, in 3 ~ 8mm spray trimming operation, to there is no need completely.In addition laser sensor higher to environmental requirement, can not long-time continuous operation, inefficiency.
Summary of the invention
The technical problem to be solved in the present invention is: propose a kind of Intelligent Laser trimming system based on binocular vision and method thereof, binocular vision is utilized to obtain the image information of workpiece blank part and mate, from realizing manipulator cutting operation with the cutting profile curve that scanning obtains.
Technical scheme of the present invention is as follows:
Based on an Intelligent Laser trimming system for binocular vision, comprise industrial computer, scanner, box Cutting platform, manipulator, servomotor, vision mobile platform, illuminator and CCD binocular vision video camera are installed in described box Cutting platform; Described industrial computer is arranged on the CCD binocular vision camera motion on vision mobile platform by Serve Motor Control, described manipulator is connected with industrial computer by arranging a manipulator control cabinet, and described vision mobile platform is connected with industrial computer by arranging a vision mobile platform switch board;
Described illuminator adopts machine vision light source, and brightness-adjustable and crevice projection angle, be contained in box Cutting platform top, ensures that CCD binocular vision video camera photographs the processing site figure of clear no-reflection;
Described industrial computer is connected with scanner, and industrial computer obtains operation drawing information by scanner, profile information and dimension information in abstraction process drawing;
The video that described industrial computer obtains and display CCD binocular vision video camera is taken, in order to adjust machine vision light source to obtain processing site image clearly; The profile of image scene and operation drawing matches acquisition cutting path, automatically controls driving device hand and completes cutting operation.
Above-mentioned box Cutting platform three side sealing closes, and disturbs for preventing external environment condition light.
Based on an Intelligent Laser method for cutting edge for binocular vision, utilize above-mentioned system, its step is as follows:
(1), start industrial computer, start two threads simultaneously;
(2), thread 1: the cutting action drawing put in advance is on the scanner scanned, after the operation picture that scanning obtains is carried out Image semantic classification, extract cutting profile curve and dimension information again, contour curve is discrete and arrange, obtain one group of orderly machining feature point coordinates;
(3), thread 2: the automatic acquisition task performing processing site optimized image;
(3a), first adjust illuminator, guarantee that CCD binocular vision video camera obtains image scene clearly;
(3b), box Cutting platform surface be provided with the special identifier calibration point being supplied to image automatic Calibration in advance, each calibration point represents a coordinate position under robot coordinate system;
(3c), processing site single-frame images is obtained by CCD binocular vision video camera, comprise left order, right order two pictures, Image semantic classification is carried out to left order image, calibration point in the left order image of automatic identification, use Tsai two-step method to carry out automatic Calibration and distortion process to left order image, obtain the mapping relations of left order image and robot coordinate system;
(3d), identify and the geometric center of workpiece blank in left order image solve the relative position relation of the geometric center of CCD binocular vision video camera center and workpiece blank according to the calibration result of step (3c);
(3e), according to the relative position relation of step (3d), servomotor drive is driven to be mounted with the vision mobile platform motion of CCD binocular vision video camera, make directly over CCD binocular vision video camera central motion to workpiece blank center, obtain processing site optimized image;
(3f), CCD binocular vision video camera mechanism system is semi-closed loop system, need to carry out repetition step (3c), judge whether CCD binocular vision video camera moves to scope directly over workpiece blank center, if so, then carry out step (3e); If not, then step (3d) operation is carried out;
(3g), processing site optimized image is obtained by step (3e) CCD binocular vision video camera, left order, right order two pictures respectively, respectively image filtering, enhancing, identification, demarcation, distortion process are carried out to two pictures, thus obtain the mapping relations of two image pixel coordinates and robot coordinate system;
(4), according to the mapping relations obtained in step (3g) it be projected to respectively in above-mentioned left order and right order two images, workpiece blank outline and cutting profile curve are carried out position relationship and mate by industrial computer automatically;
(5), the depth information of planar point on corresponding cutting profile curve can be solved according to the parallax of the discrete point on same cutting profile curve in two images, namely solve the three-dimensional coordinate of characteristic point under robot coordinate system in cutting profile curve;
(6), according to the kinematical equation of robotics, by the cut point coordinate under robot coordinate system and attitude Converse solved go out the joint angle in each joint of manipulator;
(7), the joint angle corresponding to adjacent two cut points carry out linear interpolation, writes program that robot drives can identify and import manipulator control cabinet to carry out cutting operation.
The beneficial effect that the present invention reaches:
Intelligence degree of the present invention is higher, easy and simple to handle.System obtains cutting profile information by scanner scanning drawing and automatically mates with the visual pattern through image procossing and obtains cutting path.Cut coordinate points and artificial method of demarcating acquisition cutting path compared to the artificial extraction of like product, intelligence degree of the present invention is higher and lower to the skill level requirement of operating personnel.
The cutting robot of manual demarcation and artificial teaching relies on eye-observation to ensure the precision of cut point, and often precision is not high.The present invention utilizes high-precision CCD binocular vision camera, the vision mobile platform automatically regulated and machine vision light source, ensures to photograph optimum Cutting photo site, to reach cutting accuracy within 1mm.And system includes calibrating platform, there is the function of automatic Calibration, eliminate the artificial vision calibration process participated in, substantially increase intelligence degree.
Compared to the cutting robot needing manual programming or teaching one by one, the present invention according to cutting path and attitude automated programming, can save the time of programming and teaching, has improve production efficiency.
Accompanying drawing explanation
Fig. 1 is the structural representation of present system;
Fig. 2 is method flow diagram of the present invention;
Fig. 3 is the picture that the translucent drawing paper of scanner institute obtains;
Fig. 4 is that scanned picture is through image procossing gained picture;
Fig. 5 is the pixel schematic diagram of scanned picture through image procossing gained profile heavy line;
Fig. 6 is the schematic diagram of unit picture element and field thereof 8 pixels;
Fig. 7 is pixel profile schematic diagram inside and outside image procossing gained heavy line;
Fig. 8 is the box Cutting platform that surface is provided with calibration point;
Fig. 9 is the zoning plan of calibration point.
Detailed description of the invention
Below in conjunction with accompanying drawing, the invention will be further described.Following examples only for technical scheme of the present invention is clearly described, and can not limit the scope of the invention with this.
As shown in Figure 1, a kind of Intelligent Laser trimming system based on binocular vision, comprise industrial computer 1, scanner 2, box Cutting platform 5, manipulator 10, servomotor 9, vision mobile platform 7, illuminator 8 and CCD binocular vision video camera 6 are installed in box Cutting platform 5; Industrial computer 1 controls by servomotor 9 the CCD binocular vision video camera 6 be arranged on vision mobile platform 7 and moves, manipulator 10 is connected with industrial computer 1 by arranging a manipulator control cabinet 3, and vision mobile platform 7 is connected with industrial computer 1 by arranging a vision mobile platform switch board 4;
Illuminator 8 adopts machine vision light source, brightness-adjustable and crevice projection angle, is contained in box Cutting platform 5 top, ensures that CCD binocular vision video camera 6 photographs the processing site figure of clear no-reflection;
Industrial computer 1 is connected with scanner 2, and industrial computer 1 obtains operation drawing information by scanner 2, profile information and dimension information in abstraction process drawing;
Industrial computer 1 obtains and shows the video of CCD binocular vision video camera 6 shooting, in order to adjust machine vision light source to obtain processing site image clearly; The profile of image scene and operation drawing matches acquisition cutting path, automatically controls driving device hand and completes cutting operation.
Above-mentioned box Cutting platform 5 three side sealing closes, and disturbs for preventing external environment condition light.
Based on an Intelligent Laser method for cutting edge for binocular vision, utilize above-mentioned system, its step is as follows:
(1), start industrial computer, start two threads simultaneously.
(2), thread 1: the cutting action drawing put in advance is on the scanner scanned, after the operation picture that scanning obtains is carried out Image semantic classification, extract cutting profile curve and dimension information again, contour curve is discrete and arrange, obtain one group of orderly machining feature point coordinates.
Contour curve is discrete and algorithm that is arrangement is as follows:
2a, arrange scanner ppi resolution ratio, scanning obtains 1:1 colour picture, as shown in Figure 3.(picture is divided into two parts, red drawing scale value part, black cutting wheel profile)
2b, be partitioned into red ratio value according to rgb value, identify drawing scale.
2c, red ratio value part RGB is set to (255255255) namely becomes white, and pixel RGB values is converted to gray value.
2d, image enhaucament: by image intensity value lower than 100 block of pixels gray value be set to 0; Image intensity value is set to 255 higher than the block of pixels gray value of 100.
2e, through previous step image enhaucament, in image, only deposit the block of pixels that gray value is 0 and 255.In theory, profile bold portion gray value 0 represents black, and remainder gray value 255 represents white.But in fact there is the white pixel block that gray value is 255 in profile solid line inside.All white pixel block in scan image, centered by white pixel block, to its neighborhood four direction scanning extension 10 block of pixels.If four direction all exists the block of pixels that gray value is 0, then this white centers block of pixels gray value is set to 0 (this step is that profile solid line inner white noise is become black).
2f, scanned picture are because of reasons generation noise jamming (as shown in Fig. 3 noise region) such as quality of scanning, signal disturbing, drawing spot folds.Noise jamming can affect Iamge Segmentation and identification, and therefore image has to pass through filtering.Fourier changes, and image is changed to frequency domain from transform of spatial domain, high frequency positional representation noise or image edge location.From high frequency position by gray value be 0 pixel carry out region growth, make image noise region or TP bold portion according to the judgement of counting of each region.The point gray value in noise region is set to 255.Filter effect as shown in Figure 4.
2g, image now except the gray value of bold portion be except 0, remainder gray value is all 255.The solid line lines extracted through upper step have thickness, not easily extract the coordinate points of profile, as shown in Figure 5.
All unit picture elements in traversing graph picture, extract gray value and are 0 and around eight pixels exist gray value is that the unit picture element of 255 is as profile point (as shown in Figure 6).Judge whether there are two other profile point in profile point neighborhood four pixels extracted.If be then profile point, then this point is rejected profile point if not.These unit picture element set of now extracting are inside and outside two outline lines of heavy line, and two outline lines are formed by connecting between two by unit picture element, as shown in Figure 7.
2h, from two outline lines, choose arbitrarily a unit picture element respectively as starting point.In certain direction the point two outline lines is carried out respectively sorting (in step 2g, illustrating that these two outline lines are formed by connecting between two by unit picture element from original position, on outline line, a unit picture element has and only has two block of pixels to be adjacent, so pixel sequentially can sort from original position on outline line in a direction).
2i, (4, interval point solves its left and right partial derivative to obtain the approximate derivative value of each pixel of internal and external contour line respectively.If left and right local derviation numerical value is close, then average as approximate derivative value; If left and right local derviation numerical bias is excessive, then this point is distortion point, there is not derivative value, using left and right local derviation as its approximate derivative value).Order chooses outline pixel, contrasts the approximate derivative value of the point of Internal periphery near it, select wherein derivative value immediate one as match point (if distortion point then to choose on neighbouring inner outline corresponding distortion point.4, the front and back point approximate derivative value of distortion point is difficult to the actual derivative value of reflection, therefore do not participate in derivative value coupling, but 4, the front and back point of distortion point corresponding to Internal periphery matches in order).Match point pixel coordinate (x, y), (x1, y1) to be sued for peace respectively the mid point that is averaged its line worth discrete pixels point coordinates (X, Y) as heavy line outline line.
2j, pixel coordinate to be converted under robot coordinate system (1/ppi*25.4*X/D, 1/ppi*25.4*Y/D).(resolution ratio ppi represents the number of pixels that per inch has.The size dimension of 1/ppi*25.4 gained numeric representation unit picture element, and unit is mm.D is drawing scale, and (X, Y) is pixel coordinate system lower whorl profile discrete point coordinate.)
(3), thread 2: the automatic acquisition task performing processing site optimized image.
(a), first adjust illuminator, guarantee that CCD binocular vision video camera obtains image scene clearly.
B (), box Cutting platform surface is provided with the special identifier calibration point being supplied to image automatic Calibration in advance, each calibration point represents a coordinate position under robot coordinate system.
(c), obtain processing site single-frame images by CCD binocular vision video camera, comprise left order, right order two pictures, Image semantic classification is carried out to left order image, calibration point in the left order image of automatic identification, use Tsai two-step method to carry out automatic Calibration and distortion process to left order image, obtain the mapping relations of left order image and robot coordinate system.
As shown in Figure 8, box Cutting platform surface is provided with calibration point.Each calibration point home position represents a coordinate points under mechanical arm coordinate system.As shown in Figure 9, each calibration point is distinguished by zones of different color alignment.If white is 0, yellow is 1, and blueness is 2, and redness is 3, encodes as shown in table 1.
| ABCDEFGHI | Number | ABCDEFGHI | Number | ABCDEFGHI | Number | ABCDEFGHI | Number |
| 120000000 | 0 | 102000030 | 10 | 130000000 | 20 | 103000020 | 30 |
| 102000000 | 1 | 102000003 | 11 | 103000000 | 21 | 103000002 | 31 |
| 100200000 | 2 | 100203000 | 12 | 100300000 | 22 | 100302000 | 32 |
| 100020000 | 3 | 100200300 | 13 | 100030000 | 23 | 100300200 | 33 |
| 120003000 | 4 | 100200030 | 14 | 130002000 | 24 | 100300020 | 34 |
| 120000300 | 5 | 100200003 | 15 | 130000200 | 25 | ||
| 120000030 | 6 | 100023000 | 16 | 130000020 | 26 | ||
| 120000003 | 7 | 100020300 | 17 | 130000002 | 27 | ||
| 102003000 | 8 | 100020030 | 18 | 103002000 | 28 | ||
| 102000300 | 9 | 100020003 | 19 | 103000200 | 29 |
Table 1
(c1), scanned picture, extract each calibration point A district yellow circle, using composition each A district circle pixel coordinate ask arithmetic mean of instantaneous value as center of circle pixel coordinate (u, v).
(c2), centered by the center of circle, A district, increase sweep radius gradually, judge that A district is with exterior pixel RGB color value.If color is white, blue or white, red, continue increased radius value, until there is other color, radius increases and terminates, and has so scanned BCDE district.
(c3) if scan in vain, blue, then asked by the pixel coordinate of blue region arithmetic mean of instantaneous value as color lump center of gravity (u1, v1), differentiate that color lump center of gravity is at the center of circle (u, v) orientation (if u1>u, v1>v then blueness be in E district; U1>u, v1<v then blueness are in C district; U1<u, v1<v then blueness are in B district; U1<u, v1>v then blueness are in D district), BCDE district color coding can be determined.If scan in vain, red dichromatism, then ask arithmetic mean of instantaneous value as color lump center of gravity the pixel coordinate of red area, differentiate that color lump center of gravity is in the orientation of the center of circle (u, v), can determine BCDE district color coding.
(c4), continue to increase sweep radius, judge that BCDE is with exterior domain RGB color value.If color is white, blue or white, red, continue increased radius value, until be white entirely beyond sweep radius, radius increases and terminates, and has so scanned FGHI district.
(c5) if scan in vain, blue, then ask arithmetic mean of instantaneous value as color lump center of gravity (u1, v1) pixel coordinate of blue region, differentiate that color lump center of gravity is in the orientation of the center of circle (u, v), can determine BCDE district color coding.If scan in vain, red dichromatism, then ask arithmetic mean of instantaneous value as color lump center of gravity the pixel coordinate of red area, differentiate that color lump center of gravity is in the orientation of the center of circle (u, v), can determine BCDE district color coding.
(c6), through above step obtain ABCDEFGHI field color coding, obtain the demarcation number that calibration point is corresponding.Number namely determine to demarcate and obtain coordinate (X, Y) under mechanical arm coordinate system corresponding to calibration point.
(c7) from the calibration point obtained, extracting 16 calibration points, (these 16 points are generally peripheral 12, nexine 4 as the input parameter of Tsai two-step method.But because workpiece can stop calibration point, thus 16 points extract principle for peripheral demarcate to count to demarcate more than nexine count), solve camera interior and exterior parameter, obtain being tied to by pixel coordinate the transformation matrix that robot coordinate system changes.Each pixel coordinate (u, v), under transformation matrix is converted to robot coordinate system (X, Y), namely obtains pixel coordinate and robot coordinate mapping relations.
(d), identify and the geometric center of workpiece blank in left order image solve the relative position relation of the geometric center of CCD binocular vision video camera center and workpiece blank according to the calibration result of step (c).
(d1), by the image scene that collects from RGB model conversion to HSI model (tone, saturation degree, intensity).
(d2), workpiece color is gathered, by H, S, I threshold range of program determination color of object.This threshold range is using the judgment basis increased as region and preserve into color of object storehouse.
(d3), by certain step scan photo site, using pixel in judgment basis of H, S, I as Seed Points.
(d4) centered by Seed Points, judge whether neighborhood 8 pixels meet H, S, I judgment basis.If so, then as new Seed Points, and be incorporated in growth district, put area flag position.The pixel being set to flag bit can not be repeated to judge, to improve efficiency of algorithm.If around 8 pixels are without new seed point, then region increases end.
(d5), centered by new seed point, step (d4) is repeated.
(d6), region increases end.Field working conditions color is comparatively complicated, also there is noise interference, but the region area of these interference is less simultaneously, is rejected target area less for area, and leaving the maximum region of area is then target area.
(d7), image often row in target area head and the tail two pixels of contiguous pixels be wire-frame image vegetarian refreshments.The contour pixel of target area is extracted line by line according to this judgment basis.
(d8), the pixel coordinate that defines each contour pixel is (Xn, Yn), ask its contour pixel arithmetic mean of instantaneous value (
) as the center point coordinate of profile.
(d9), the central pixel point position of camera center position and image, according to camera centre coordinate (x0 under calibration result and known mechanical arm coordinate system, and profile barycentric coodinates (X y0), Y), the x that can obtain practical work piece center and camera center is to, y to deviation distance X-x0, Y-y0.
(e), relative position relation according to step (d), servomotor drive is driven to be mounted with the vision mobile platform motion of CCD binocular vision video camera, make directly over CCD binocular vision video camera central motion to workpiece blank center, obtain processing site optimized image.
F (), CCD binocular vision video camera mechanism system are semi-closed loop system, need to carry out repetition step (c), judge whether CCD binocular vision video camera moves to scope directly over workpiece blank center, if so, then carries out step (e); If not, then step (d) operation is carried out.
(g), obtain processing site optimized image by step (e) CCD binocular vision video camera, left order, right order two pictures respectively, respectively image filtering, enhancing, identification, demarcation, distortion process are carried out to two pictures, thus obtain the mapping relations of two image pixel coordinates and robot coordinate system.
(4), be projected in above-mentioned left order and right order two images according to the mapping relations obtained in step (g) respectively by it, workpiece blank outline and cutting profile curve are carried out position relationship and mate by industrial computer automatically.
(5), the depth information of planar point on corresponding cutting profile curve can be solved according to the parallax of the discrete point on same cutting profile curve in two images, namely solve the three-dimensional coordinate of characteristic point under robot coordinate system in cutting profile curve.
Example: on known cutting curve, discrete point coordinate under mechanical arm coordinate system is (x, y), this is (u1 in the respective pixel position of left images, v1), (u2, v2), (method for solving is quoted and is edited China Machine Press from " vision measurement technology " Chi Jiannan can to solve this depth value according to the difference of (u1, v1), (u2, v2).
(6), according to the kinematical equation of robotics, by the cut point coordinate under robot coordinate system and attitude Converse solved go out the joint angle in each joint of manipulator.Method for solving is shown in " robotics " second edition Cai Zixing edits publishing house of Tsing-Hua University, and this is robot kinematics's common-used formula.
(7), the joint angle corresponding to adjacent two cut points carry out linear interpolation, writes program that robot drives can identify and import manipulator control cabinet to carry out cutting operation.
The above is only the preferred embodiment of the present invention; it should be pointed out that for those skilled in the art, under the prerequisite not departing from the technology of the present invention principle; can also make some improvement and distortion, these improve and distortion also should be considered as protection scope of the present invention.
Claims (3)
1. the Intelligent Laser trimming system based on binocular vision, it is characterized in that: comprise industrial computer, scanner, box Cutting platform, manipulator, servomotor, vision mobile platform, illuminator and CCD binocular vision video camera are installed in described box Cutting platform; Described industrial computer is arranged on the CCD binocular vision camera motion on vision mobile platform by Serve Motor Control, described manipulator is connected with industrial computer by arranging a manipulator control cabinet, and described vision mobile platform is connected with industrial computer by arranging a vision mobile platform switch board;
Described illuminator adopts machine vision light source, and brightness-adjustable and crevice projection angle, be contained in box Cutting platform top, ensures that CCD binocular vision video camera photographs the processing site figure of clear no-reflection;
Described industrial computer is connected with scanner, and industrial computer obtains operation drawing information by scanner, profile information and dimension information in abstraction process drawing;
The video that described industrial computer obtains and display CCD binocular vision video camera is taken, in order to adjust machine vision light source to obtain processing site image clearly; The profile of image scene and operation drawing matches acquisition cutting path, automatically controls driving device hand and completes cutting operation.
2. a kind of Intelligent Laser trimming system based on binocular vision according to claim 1, is characterized in that: described box Cutting platform three side sealing closes, and disturbs for preventing external environment condition light.
3. based on an Intelligent Laser method for cutting edge for binocular vision, it is characterized in that the system utilized described in claim 1, its step is as follows:
(1), start industrial computer, start two threads simultaneously;
(2), thread 1: the cutting action drawing put in advance is on the scanner scanned, after the operation picture that scanning obtains is carried out Image semantic classification, extract cutting profile curve and dimension information again, contour curve is discrete and arrange, obtain one group of orderly machining feature point coordinates;
(3), thread 2: the automatic acquisition task performing processing site optimized image;
(3a), first adjust illuminator, guarantee that CCD binocular vision video camera obtains image scene clearly;
(3b), box Cutting platform surface be provided with the special identifier calibration point being supplied to image automatic Calibration in advance, each calibration point represents a coordinate position under robot coordinate system;
(3c), processing site single-frame images is obtained by CCD binocular vision video camera, comprise left order, right order two pictures, Image semantic classification is carried out to left order image, calibration point in the left order image of automatic identification, use Tsai two-step method to carry out automatic Calibration and distortion process to left order image, obtain the mapping relations of left order image and robot coordinate system;
(3d), identify and the geometric center of workpiece blank in left order image solve the relative position relation of the geometric center of CCD binocular vision video camera center and workpiece blank according to the calibration result of step (3c);
(3e), according to the relative position relation of step (3d), servomotor drive is driven to be mounted with the vision mobile platform motion of CCD binocular vision video camera, make directly over CCD binocular vision video camera central motion to workpiece blank center, obtain processing site optimized image;
(3f), CCD binocular vision video camera mechanism system is semi-closed loop system, needs to carry out repetition step (3c), judges whether CCD binocular vision video camera moves to scope directly over workpiece blank center, if so, then carry out step (3e); If not, then step (3d) operation is carried out;
(3g), processing site optimized image is obtained by step (3e) CCD binocular vision video camera, left order, right order two pictures respectively, respectively image filtering, enhancing, identification, demarcation, distortion process are carried out to two pictures, thus obtain the mapping relations of two image pixel coordinates and robot coordinate system;
(4), according to the mapping relations obtained in step (3g) it be projected to respectively in above-mentioned left order and right order two images, workpiece blank outline and cutting profile curve are carried out position relationship and mate by industrial computer automatically;
(5), the depth information of planar point on corresponding cutting profile curve can be solved according to the parallax of the discrete point on same cutting profile curve in two images, namely solve the three-dimensional coordinate of characteristic point under robot coordinate system in cutting profile curve;
(6), according to the kinematical equation of robotics, by the cut point coordinate under robot coordinate system and attitude Converse solved go out the joint angle in each joint of manipulator;
(7), the joint angle corresponding to adjacent two cut points carry out linear interpolation, writes program that robot drives can identify and import manipulator control cabinet to carry out cutting operation.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510509465.4A CN105149794B (en) | 2015-08-18 | 2015-08-18 | A kind of Intelligent Laser method for cutting edge based on binocular vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201510509465.4A CN105149794B (en) | 2015-08-18 | 2015-08-18 | A kind of Intelligent Laser method for cutting edge based on binocular vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105149794A true CN105149794A (en) | 2015-12-16 |
| CN105149794B CN105149794B (en) | 2016-09-28 |
Family
ID=54790993
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201510509465.4A Expired - Fee Related CN105149794B (en) | 2015-08-18 | 2015-08-18 | A kind of Intelligent Laser method for cutting edge based on binocular vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105149794B (en) |
Cited By (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105225225A (en) * | 2015-08-31 | 2016-01-06 | 臻雅科技温州有限公司 | A kind of leather system for automatic marker making method and apparatus based on machine vision |
| CN105666274A (en) * | 2016-02-03 | 2016-06-15 | 华中科技大学 | Dinner plate edging method based on vision control |
| CN105825175A (en) * | 2016-03-11 | 2016-08-03 | 上海弼智仿生高科技有限公司 | Generation method for industrial vision detection scheme and industrial vision detection method |
| CN106670652A (en) * | 2016-12-29 | 2017-05-17 | 苏州逸美德科技有限公司 | Laser coaxial processing device and method |
| CN106964907A (en) * | 2017-04-25 | 2017-07-21 | 重庆邮电大学 | A kind of method and apparatus of laser cutting |
| CN107931865A (en) * | 2016-10-12 | 2018-04-20 | 中国科学院沈阳自动化研究所 | Large Enclosure inner wall dirt automatic online laser cleaner and method |
| CN108279584A (en) * | 2018-01-19 | 2018-07-13 | 海宁市晨丰橡塑有限公司 | A kind of hardware turning equipment |
| CN108414325A (en) * | 2018-06-05 | 2018-08-17 | 中南大学湘雅三医院 | A kind of organization chip coremaking system and core-making method based on image recognition positioning |
| CN108961168A (en) * | 2018-07-27 | 2018-12-07 | 奔腾激光(温州)有限公司 | Laser processing field graph display method |
| CN109501084A (en) * | 2017-09-14 | 2019-03-22 | 北京天源科创风电技术有限责任公司 | Overlap diced system and method for wind generator set blade |
| CN110148124A (en) * | 2019-05-21 | 2019-08-20 | 中山大学 | Throat recognition methods, device, system, storage medium and equipment |
| CN110449749A (en) * | 2019-07-19 | 2019-11-15 | 东莞理工学院 | A kind of laser cutting scanning system |
| CN110580003A (en) * | 2018-06-11 | 2019-12-17 | 大族激光科技产业集团股份有限公司 | Position control method, device and equipment of servo motion system and storage medium |
| CN111145254A (en) * | 2019-12-13 | 2020-05-12 | 上海新时达机器人有限公司 | Door valve blank positioning method based on binocular vision |
| CN111397535A (en) * | 2020-04-29 | 2020-07-10 | 苏州龙抬头智能科技有限公司 | Dynamic calibration method based on linear scanning laser and conveyor belt operating platform device |
| CN111736528A (en) * | 2020-07-07 | 2020-10-02 | 华中科技大学 | Laser cutting automatic programming system based on vision deviation rectification |
| CN113607419A (en) * | 2021-08-02 | 2021-11-05 | 广东工业大学 | Engine cylinder block electronic tag detection device and method |
| CN113634876A (en) * | 2021-10-18 | 2021-11-12 | 武汉逸飞激光股份有限公司 | Auxiliary calibration method and device for machine vision, electronic equipment and storage medium |
| CN113878592A (en) * | 2021-10-14 | 2022-01-04 | 三一建筑机器人(西安)研究院有限公司 | Workpiece cutting method and device based on laser position finding and cutting robot |
| TWI755189B (en) * | 2020-12-07 | 2022-02-11 | 財團法人工業技術研究院 | Deburring trajectory recognition mehtod and system thereof |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101363188A (en) * | 2008-07-29 | 2009-02-11 | 瑞安市博业激光应用技术有限公司 | Cutting method along boundary of laser cutting equipment |
| EP2256576A1 (en) * | 2009-05-29 | 2010-12-01 | Datacard Corporation | System for the calibration of operating parameters of a laser engraver |
| CN202555971U (en) * | 2012-02-29 | 2012-11-28 | 武汉三工光电设备制造有限公司 | Charge coupled device (CCD) full-automatic identification and position system |
| CN202726319U (en) * | 2012-08-21 | 2013-02-13 | 武汉先河激光技术有限公司 | Silver paste laser etching machine |
| KR20130032425A (en) * | 2011-09-23 | 2013-04-02 | (주)에이디에스 테크놀로지 | Laser welder for recognizing image of laser welding edge using camera and method for aligning the same |
| CN103495807A (en) * | 2013-09-27 | 2014-01-08 | 重庆大学 | Multi-robot fiber laser cutting system |
| CN104551411A (en) * | 2014-11-18 | 2015-04-29 | 南京大学 | Calibration method of laser galvanometer processing system under guidance of binocular stereoscopic vision |
-
2015
- 2015-08-18 CN CN201510509465.4A patent/CN105149794B/en not_active Expired - Fee Related
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101363188A (en) * | 2008-07-29 | 2009-02-11 | 瑞安市博业激光应用技术有限公司 | Cutting method along boundary of laser cutting equipment |
| EP2256576A1 (en) * | 2009-05-29 | 2010-12-01 | Datacard Corporation | System for the calibration of operating parameters of a laser engraver |
| KR20130032425A (en) * | 2011-09-23 | 2013-04-02 | (주)에이디에스 테크놀로지 | Laser welder for recognizing image of laser welding edge using camera and method for aligning the same |
| CN202555971U (en) * | 2012-02-29 | 2012-11-28 | 武汉三工光电设备制造有限公司 | Charge coupled device (CCD) full-automatic identification and position system |
| CN202726319U (en) * | 2012-08-21 | 2013-02-13 | 武汉先河激光技术有限公司 | Silver paste laser etching machine |
| CN103495807A (en) * | 2013-09-27 | 2014-01-08 | 重庆大学 | Multi-robot fiber laser cutting system |
| CN104551411A (en) * | 2014-11-18 | 2015-04-29 | 南京大学 | Calibration method of laser galvanometer processing system under guidance of binocular stereoscopic vision |
Cited By (26)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105225225B (en) * | 2015-08-31 | 2017-12-22 | 温州城电智能科技有限公司 | A kind of leather system for automatic marker making method and apparatus based on machine vision |
| CN105225225A (en) * | 2015-08-31 | 2016-01-06 | 臻雅科技温州有限公司 | A kind of leather system for automatic marker making method and apparatus based on machine vision |
| CN105666274A (en) * | 2016-02-03 | 2016-06-15 | 华中科技大学 | Dinner plate edging method based on vision control |
| CN105666274B (en) * | 2016-02-03 | 2018-03-09 | 华中科技大学 | A kind of service plate edging method of view-based access control model control |
| CN105825175A (en) * | 2016-03-11 | 2016-08-03 | 上海弼智仿生高科技有限公司 | Generation method for industrial vision detection scheme and industrial vision detection method |
| CN107931865A (en) * | 2016-10-12 | 2018-04-20 | 中国科学院沈阳自动化研究所 | Large Enclosure inner wall dirt automatic online laser cleaner and method |
| CN106670652A (en) * | 2016-12-29 | 2017-05-17 | 苏州逸美德科技有限公司 | Laser coaxial processing device and method |
| CN106964907A (en) * | 2017-04-25 | 2017-07-21 | 重庆邮电大学 | A kind of method and apparatus of laser cutting |
| CN109501084A (en) * | 2017-09-14 | 2019-03-22 | 北京天源科创风电技术有限责任公司 | Overlap diced system and method for wind generator set blade |
| CN108279584A (en) * | 2018-01-19 | 2018-07-13 | 海宁市晨丰橡塑有限公司 | A kind of hardware turning equipment |
| CN108414325A (en) * | 2018-06-05 | 2018-08-17 | 中南大学湘雅三医院 | A kind of organization chip coremaking system and core-making method based on image recognition positioning |
| CN110580003A (en) * | 2018-06-11 | 2019-12-17 | 大族激光科技产业集团股份有限公司 | Position control method, device and equipment of servo motion system and storage medium |
| CN108961168B (en) * | 2018-07-27 | 2021-08-06 | 奔腾激光(温州)有限公司 | Laser processing field graph display method |
| CN108961168A (en) * | 2018-07-27 | 2018-12-07 | 奔腾激光(温州)有限公司 | Laser processing field graph display method |
| CN110148124A (en) * | 2019-05-21 | 2019-08-20 | 中山大学 | Throat recognition methods, device, system, storage medium and equipment |
| CN110449749A (en) * | 2019-07-19 | 2019-11-15 | 东莞理工学院 | A kind of laser cutting scanning system |
| CN111145254A (en) * | 2019-12-13 | 2020-05-12 | 上海新时达机器人有限公司 | Door valve blank positioning method based on binocular vision |
| CN111145254B (en) * | 2019-12-13 | 2023-08-11 | 上海新时达机器人有限公司 | Door valve blank positioning method based on binocular vision |
| CN111397535A (en) * | 2020-04-29 | 2020-07-10 | 苏州龙抬头智能科技有限公司 | Dynamic calibration method based on linear scanning laser and conveyor belt operating platform device |
| CN111736528A (en) * | 2020-07-07 | 2020-10-02 | 华中科技大学 | Laser cutting automatic programming system based on vision deviation rectification |
| TWI755189B (en) * | 2020-12-07 | 2022-02-11 | 財團法人工業技術研究院 | Deburring trajectory recognition mehtod and system thereof |
| US11656597B2 (en) | 2020-12-07 | 2023-05-23 | Industrial Technology Research Institute | Method and system for recognizing deburring trajectory |
| CN113607419A (en) * | 2021-08-02 | 2021-11-05 | 广东工业大学 | Engine cylinder block electronic tag detection device and method |
| CN113878592A (en) * | 2021-10-14 | 2022-01-04 | 三一建筑机器人(西安)研究院有限公司 | Workpiece cutting method and device based on laser position finding and cutting robot |
| CN113634876B (en) * | 2021-10-18 | 2021-12-28 | 武汉逸飞激光股份有限公司 | Auxiliary calibration method and device for machine vision, electronic equipment and storage medium |
| CN113634876A (en) * | 2021-10-18 | 2021-11-12 | 武汉逸飞激光股份有限公司 | Auxiliary calibration method and device for machine vision, electronic equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105149794B (en) | 2016-09-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105149794B (en) | A kind of Intelligent Laser method for cutting edge based on binocular vision | |
| CN106041937B (en) | A kind of control method of the manipulator crawl control system based on binocular stereo vision | |
| CN106338521B (en) | Increasing material manufacturing surface and internal flaw and pattern composite detection method and device | |
| CN107945192B (en) | Tray carton pile type real-time detection method | |
| CN104842361B (en) | Robotic system with 3d box location functionality | |
| CN108592788B (en) | 3D intelligent camera system facing spraying production line and workpiece online measurement method | |
| CN102135416B (en) | Online image detecting system and method for bottle covers | |
| CN110580725A (en) | Box sorting method and system based on RGB-D camera | |
| CN110084243B (en) | File identification and positioning method based on two-dimensional code and monocular camera | |
| CN100461820C (en) | Image processing device and registration data generation method in image processing | |
| CN106251353A (en) | Weak texture workpiece and the recognition detection method and system of three-dimensional pose thereof | |
| CN102529019B (en) | Method for mould detection and protection as well as part detection and picking | |
| CN106951905A (en) | Apple identification and localization method on a kind of tree based on TOF camera | |
| CN105528789A (en) | Robot vision positioning method and device, and visual calibration method and device | |
| CN108064560A (en) | The automatic picker system of fruit and method based on Kinect depth of field cameras | |
| CN106384115A (en) | Mechanical arm joint angle detection method | |
| Xia et al. | Workpieces sorting system based on industrial robot of machine vision | |
| CN104614372B (en) | Detection method of solar silicon wafer | |
| CN107030690A (en) | A kind of mechanical arm barrier-avoiding method of view-based access control model | |
| CN108171713A (en) | The robot of view-based access control model quickly identifies and three-dimensional visual positioning method | |
| CN102529020B (en) | Visual-motor integrated device for mould detection and protection as well as part detection and picking | |
| CN101706447A (en) | Method for detecting CCD marking defects at spherical opening of high-pressure natural gas bottle | |
| CN108109172A (en) | A kind of robotic vision system and method based on new vision | |
| CN113689365A (en) | Target tracking and positioning method based on Azure Kinect | |
| CN109905692A (en) | A kind of machine new vision system and 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: 20160928 Termination date: 20210818 |