CN108765378A - The machine vision detection method of lower workpiece profile overlap protrusion is guided based on G code - Google Patents
The machine vision detection method of lower workpiece profile overlap protrusion is guided based on G code Download PDFInfo
- Publication number
- CN108765378A CN108765378A CN201810426396.4A CN201810426396A CN108765378A CN 108765378 A CN108765378 A CN 108765378A CN 201810426396 A CN201810426396 A CN 201810426396A CN 108765378 A CN108765378 A CN 108765378A
- Authority
- CN
- China
- Prior art keywords
- overlap
- protrusion
- workpiece
- point
- code
- 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
- 238000001514 detection method Methods 0.000 title claims abstract description 48
- 238000000034 method Methods 0.000 claims abstract description 32
- 238000000605 extraction Methods 0.000 claims abstract description 17
- 238000003708 edge detection Methods 0.000 claims abstract description 10
- 239000000284 extract Substances 0.000 claims abstract description 8
- 230000009466 transformation Effects 0.000 claims abstract description 5
- 230000007547 defect Effects 0.000 claims description 9
- 238000005259 measurement Methods 0.000 claims description 7
- 230000008878 coupling Effects 0.000 claims description 4
- 238000010168 coupling process Methods 0.000 claims description 4
- 238000005859 coupling reaction Methods 0.000 claims description 4
- 230000011218 segmentation Effects 0.000 claims description 4
- 125000004122 cyclic group Chemical group 0.000 claims description 3
- 239000000203 mixture Substances 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 2
- 238000012360 testing method Methods 0.000 claims description 2
- 238000006243 chemical reaction Methods 0.000 claims 1
- 238000003672 processing method Methods 0.000 abstract description 2
- 238000004458 analytical method Methods 0.000 abstract 1
- 238000001914 filtration Methods 0.000 abstract 1
- 238000009826 distribution Methods 0.000 description 6
- 238000013450 outlier detection Methods 0.000 description 5
- 230000002159 abnormal effect Effects 0.000 description 4
- 238000011160 research Methods 0.000 description 4
- 241000417436 Arcotheres Species 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000005520 cutting process Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000009931 harmful effect Effects 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/0006—Industrial image inspection using a design-rule based approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T3/00—Geometric image transformation in the plane of the image
- G06T3/60—Rotation of a whole image or part thereof
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
Abstract
The present invention relates to a kind of machine vision detection method of the workpiece profile overlap protrusion based on G code guiding, specific object is detection workpiece profile overlap protrusion, and step is:Analysis ratiocination goes out the relationship converted between workpiece coordinate and image coordinate;The use of G code profile is guiding, in conjunction with coordinate transformation relation, extracts interesting image regions, take the area-of-interest of extraction the processing method of medium filtering, image binaryzation and edge detection, obtain workpiece profile edge;On the basis of standard G code profile, identification judges edge protuberance, using Minimum Enclosing Rectangle method, surrounds contour edge protrusion, realizes the vision-based detection of contour edge protrusion.Workpiece profile overlap protrusion machine vision detection method provided by the invention, can efficiently identify out starting point, terminal and the elevation information of workpiece profile overlap protruding features section.
Description
Technical field
The present invention relates to a kind of machine vision detection method, the machine vision of especially a kind of workpiece profile overlap protrusion is examined
Survey method
Background technology
Profile overlap is inevitably resulted from machine components manufacturing process comprising be plastically deformed in cutting process
Material, the exception of forge piece is raised, the squeezed out defective material etc. of welding.The presence of profile overlap to the presentation quality of workpiece plus
Work precision, assembly precision etc. will produce harmful effect, to reduce the stability and reliability of mechanical system, therefore remove
Overlap protrusion is particularly important in the processing of machine components and manufacturing process, however it is to detect first to go the premise of deburring
With positioning overlap.The method of detection overlap mainly utilizes vernier caliper and tack miking overlap, or naked eyes at present
Overlap is roughly positioned, however overlap protrusion has diversity and randomness so that artificial detection is difficult to be accurately positioned, and detects
Efficiency is low.Therefore, for the research for the automated arm for improving overlap protrusion accuracy of detection and detection efficiency, by more and more
Concern.
Workpiece profile abnormal point is the basic component units of workpiece profile overlap protrusion, the abnormal point about overlap protrusion
It surveys, domestic and international experts and scholars have done this more research, and research is broadly divided into two classes.The first kind is parametric method, this method master
Want linear skeleton pattern or for the skeleton pattern that can be fitted to simple nonlinear combination.Second class is nonparametric
Method, non-linear profile model of this method mainly for random complexity.The above-mentioned research to contour curve outlier detection is all
Abnormal point detecting method based on statistical distribution assumes distribution or probabilistic model to the data set to be detected, and then uses not
Harmony, which is examined, carries out outlier detection.But in the outlier detection of practical overlap protrusion, abnormal point has random diversity,
The distribution or probabilistic model that above-mentioned detection method is assumed, therefore, this outlier detection based on statistical distribution can not be met
Method is applied in workpiece profile overlap outlier detection with certain limitation.
The present invention is object self property using the existing G code of workpiece, using point and the neighbor relationships of straight line, to workpiece into
Row profile overlap protrusion is positioned and is measured.This method infers the relationship converted between workpiece coordinate and image coordinate first, with G
Code profile is that guiding extracts interesting image regions in conjunction with coordinate transformation relation, and intermediate value is taken to the area-of-interest of extraction
It filters, the method for image binaryzation and edge detection, obtains workpiece profile edge.Finally on the basis of standard G code profile, know
Do not judge edge protuberance, is used in combination minimum enclosed rectangle to surround contour edge protrusion, realizes the vision-based detection of contour edge protrusion.For
The detection of workpiece profile overlap protrusion provides a kind of detection method of machine vision.
Invention content
The technical problem to be solved in the present invention and the technical assignment of proposition are to be improved to prior art and perfect,
A kind of machine vision detection method of the workpiece profile overlap protrusion guided based on G code is provided, this method can be to random more
The workpiece flashes protrusion of sample distribution is identified and detects, and can be detected, possess higher under the arbitrary placement state of workpiece
Detection efficiency and precision.
To achieve the above object, the present invention takes following technical scheme:
A kind of machine vision detection method being guided lower workpiece profile overlap protrusion based on G code, is flown using workpiece profile
Side protrusion vision detection system, the system include industrial camera, light source and camera lens, motion control card, servo-driver, servo electricity
Machine, ball-screw, nut, retarder, detection platform, industrial camera along X-axis with the platform that Y-axis moves by can connect by X-axis
With two axis servo-actuating devices of Y-axis composition, servo nut on every axis and ball-screw in two axis servo-actuating devices are led to
It crosses shaft coupling and connects servo motor with retarder, servo motor connects servo-driver by encoder, and servo-driver passes through
Line connection terminal plate connects host computer and motion control card;Industrial camera, camera lens and light source are placed in detection platform
Workpiece for measurement top, industrial camera are electrically connected host computer, include the following steps:
Step 1. is analyzed and determines the work that the workpiece coordinate system of workpiece to be detected is obtained with industrial camera in detection platform
Transformational relation between the image coordinate system of part picture, in the workpiece coordinate system and image detection characterized by G code based on
Image coordinate, which ties up in identification burr protrusion, has closely contact, analyzes and infer platform coordinate system and workpiece coordinate system
The transformational relation of transformational relation and platform coordinate system and image coordinate system obtains workpiece coordinate system with above two transformational relation
With the transformational relation of image coordinate system;
Step 2. carries out region of interesting extraction to the collected picture of industrial camera, and is directed to region of interest area image,
The method for taking image binaryzation and edge detection obtains the contour edge for needing to detect;
Step 3. carries out overlap bump defects based on the profile information in G code, to area-of-interest profile and judges identification,
Using Minimum Enclosing Rectangle method, contour edge protrusion is surrounded, realizes the vision-based detection of contour edge protrusion, is completed to workpiece profile
The positioning and measurement of overlap protrusion obtain the starting point and elevation information of the characteristic segments of overlap protrusion.
The corresponding coordinate of G code is mapped in image coordinate by the transformation relation obtained according to step 1 in step 2, according to
Line style is line segment or circular arc, carries out region of interesting extraction;The ROI extractions of line segment type:According to the starting point P1 of line segment, terminal P2
With line segment length P1P2, drafting one is using line segment P1P2 as the rectangle of neutrality line, the k of a height of overlap highest empirical value of rectangle.
The ROI extractions of circular arc type:Circular arc starting point P3 and terminal P4 is connected, is drawn using line segment P3P4 as the rectangle of neutrality line, a height of experience
Value K, you can obtain area-of-interest and the method that area-of-interest takes image binaryzation and edge detection is obtained needing to examine
The contour edge of survey.
Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, for being judged as the G code of line segment
Section, the line segment starting point coordinate preserved after parsing are set as P1 (x1, y1), P2 (x2, y2), and the detection algorithm of overlap protrusion is:
1) according to two point P1, P2 on line segment, the general expression linear equation of straight line L where releasing line segment;
2) number at the edge midpoints point set V1 containing overlap is calculated;
3) i-th of point Pi is taken in point set V, and calculates the i-th point of distance Dist for arriving straight line L;
4) judge that Dist and standard overlap protrusion judge the relationship of distance D deviations, if Dist>D deviations then protect point Pi
There are in a new point set V2;
5) minimum rectangle that straight line L is parallel to using one side surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates square
The width and height of shape, unit pixel;
6) image coordinate is converted by world coordinates according to camera calibration parameter, while calculates rectangle in world coordinates
Practical wide and high, unit mm, can be obtained by the practical wide high of overlap protrusion and the position occurred at this time.
Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, the knowledge for the G code section of circular arc
Other determination method:
1) according to terminal starting point two point coordinates P3, P4 and radius R, the general expression an arc equation of circle O where releasing circular arc;
2) central angle for judging circular arc then carries out circular arc segmentation if it is greater than 15 °, and 15 ° of mark is not more than according to central angle
Standard is segmented, and can the edge point set containing overlap protrusion be divided into M sections;
3) cyclic variable j=0 is set, Pj sections are taken in M sections of point set;
4) i-th of point Pi is taken in point set Vj, and calculates the i-th point of distance Dist for arriving center of circle O;
5) judge that Dist and standard overlap protrusion judge distance D deviations, if Dist>Point Pi is then stored in one by D deviations
In a new point set V2, otherwise do not handle;
6) usable floor area minimum rectangle surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates the width and height of rectangle, single
Position is pixel;
7) image coordinate is converted by world coordinates according to camera calibration parameter, while calculates rectangle in world coordinates
Practical wide and high, unit mm can be obtained by the position that the raised practical width of Pj sections of overlaps is high and occurs at this time.
Beneficial effects of the present invention:
1. the profile information that this method is provided using G code provides the criterion of complete profile as standard form,
It is more accurate compared with artificial detection, possess higher accuracy.Meet the testing requirements of random distribution overlap workpiece operating mode.
2. comparing general image processing method, the extraction of the area-of-interest guided based on G code is innovatively increased,
The image for being indifferent to region is eliminated, the time of image procossing is greatly reduced, improves image procossing and the identification of overlap protrusion
Efficiency.
Description of the drawings
Fig. 1 is the system platform structure principle chart that this method is implemented;
Fig. 2 is the overlap protrusion detection method flow chart that this method is implemented;
Fig. 3 is the camera motion scheme schematic diagram that this method is implemented;
Fig. 4 is the region of interesting extraction illustraton of model that this method is implemented;
Wherein:A is line segment benchmark, and b is circular arc benchmark;
Fig. 5 is the image processing effect figure that this method is implemented;
Wherein:A is artwork, and b is the ROI extraction figures of G code guiding, and c is binary picture, and d is edge detection graph, and e is winged
Edge point collection minimum rectangle surrounds figure.
Specific implementation method
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with the accompanying drawings and case study on implementation is to this hair
It is bright to make further clear, complete description:
The present invention establishes workpiece profile overlap protrusion vision detection system as shown in Figure 1, and the system is by three parts
Composition:Image capture module, motion-control module and host module.Wherein image capture module includes industrial camera 4, light source 6
With camera lens 5;Motion-control module includes motion control card 17, servo-driver, servo motor and executing agency;Host module master
To include host computer 1 and power-supply device.Executing agency includes ball-screw 10, nut 11, retarder 8.Industrial camera 4 is logical
The nut 11 in X-axis and the nut 11 on ball-screw 10, Y-axis and ball-screw can be connected along X-axis with the platform that Y-axis moves by crossing
10, nut 11 and ball-screw 10 in X-axis connect X-servo motor 12, the nut 11 in Y-axis by shaft coupling 9 and retarder 8
And ball-screw 10 connects Y servo motors 13 by shaft coupling 9 and retarder 8, X-servo motor 12 connects X-axis by encoder 7
Servo-driver 14, Y servo motors 13 connect Y-axis servo-driver 15 by encoder 7, and X-axis servo-driver 14 and Y-axis are watched
It takes driver 15 and host computer 1 and motion control card 17 is connected by line connection terminal plate 16.Industrial camera 4,5 and of camera lens
Light source 6 is placed in 3 top of workpiece for measurement in detection platform 2, and industrial camera 4 is electrically connected host computer 1.
The machine vision detection method of the workpiece profile overlap protrusion based on G code of the present invention is flown using workpiece profile
Side protrusion vision detection system, includes the following steps:
Step 1. is analyzed and determines the work that the workpiece coordinate system of workpiece to be detected is obtained with industrial camera in detection platform
Transformational relation between the image coordinate system of part picture, in the workpiece coordinate system and image detection characterized by G code based on
Image coordinate, which ties up in identification burr protrusion, has closely contact, analyzes and infer platform coordinate system and workpiece coordinate system
The transformational relation of transformational relation and platform coordinate system and image coordinate system obtains workpiece coordinate system with above two transformational relation
With the transformational relation of image coordinate system;
Step 2. carries out region of interesting extraction to the collected picture of industrial camera, and is directed to region of interest area image,
The method for taking image binaryzation and edge detection obtains the contour edge for needing to detect;
Step 3. carries out overlap bump defects based on the profile information in G code, to area-of-interest profile and judges identification,
Using Minimum Enclosing Rectangle method, contour edge protrusion is surrounded, realizes the vision-based detection of contour edge protrusion, is completed to workpiece profile
The positioning and measurement of overlap protrusion obtain the starting point and elevation information of the characteristic segments of overlap protrusion.
The corresponding coordinate of G code is mapped in image coordinate by the transformation relation obtained according to step 1 in step 2, according to
Line style is line segment or circular arc, carries out region of interesting extraction;The ROI extractions of line segment type:According to the starting point P1 of line segment, terminal P2
With line segment length P1P2, drafting one is using line segment P1P2 as the rectangle of neutrality line, the k of a height of overlap highest empirical value of rectangle.
The ROI extractions of circular arc type:Circular arc starting point P3 and terminal P4 is connected, is drawn using line segment P3P4 as the rectangle of neutrality line, a height of experience
Value K, you can obtain area-of-interest and the method that area-of-interest takes image binaryzation and edge detection is obtained needing to examine
The contour edge of survey.
Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, for being judged as the G code of line segment
Section, the line segment starting point coordinate preserved after parsing are set as P1 (x1, y1), P2 (x2, y2), and the detection algorithm of overlap protrusion is:
1) according to two point P1, P2 on line segment, the general expression of straight line L is where releasing line segment:Ax+By+C=0;
2) number at the edge midpoints point set V1 containing overlap is calculated;
3) i-th of point Pi is taken in point set V, and calculates the i-th point of distance Dist for arriving straight line L;
4) judge that Dist and standard overlap protrusion judge the relationship of distance D deviations, if Dist>D deviations then protect point Pi
There are in a new point set V2;
5) minimum rectangle that straight line L is parallel to using one side surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates square
The width and height of shape, unit pixel;
6) image coordinate is converted by world coordinates according to camera calibration parameter, while calculates rectangle in world coordinates
Practical wide and high, unit mm, can be obtained by the practical wide high of overlap protrusion and the position occurred at this time.
Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, the knowledge for the G code section of circular arc
Other determination method:
1) according to terminal starting point two point coordinates P3, P4 and radius R, the general expression of circle O where releasing circular arc is:(x-a)2+
(y-b)2=R2;
2) central angle for judging circular arc then carries out circular arc segmentation if it is greater than 15 °, and 15 ° of mark is not more than according to central angle
Standard is segmented, and can the edge point set containing overlap protrusion be divided into M sections;
3) cyclic variable j=0 is set, Pj sections are taken in M sections of point set;
4) i-th of point Pi is taken in point set Vj, and calculates the i-th point of distance Dist for arriving center of circle O;
5) judge that Dist and standard overlap protrusion judge the relationship of distance D deviations, if Dist>D deviations then protect point Pi
There are in a new point set V2, otherwise do not handle;
6) usable floor area minimum rectangle surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates the width and height of rectangle, single
Position is pixel;
7) image coordinate is converted by world coordinates according to camera calibration parameter, while calculates rectangle in world coordinates
Practical wide and high, unit mm can be obtained by the position that the raised practical width of Pj sections of overlaps is high and occurs at this time.
The detection method flow of the present invention is as shown in Figures 2 and 3, and host computer parses G code first, generates industrial phase
Machine motion path;X1, Y1 axis back to zero of applied host machine module control platform, the position coordinates of industrial camera are (0,0) at this time;?
Position S1 is moved to using host module control industrial camera, at this time industrial camera central point O3 and workpiece coordinate origin O2 weights
It closes, suspends 20ms, record the position coordinates of industrial camera at this time, be platform coordinate P1 (X1, Y1), take pictures;According to dotted line
The direction host module control industrial camera of arrow moves to position S2, at this time the midpoint weight of image center point O3 and line segment O2G
It closes, suspends 20ms, record the position coordinates of industrial camera at this time, be platform coordinate P2 (X2, Y2), take pictures;According to dotted line
The direction of arrow moves to the point midway after the parsing of G code section, pause using host module control industrial camera successively
20ms records the position coordinates of industrial camera, is platform coordinate P3 (X3, Y3), takes pictures, and it is corresponding to acquire G code successively
Each profile point picture.According to the transformational relation of workpiece coordinate and image coordinate, the area-of-interest of G code guiding is extracted, such as
Fig. 4 a, shown in b.Then the method for taking image binaryzation and edge detection obtains needing to detect contour edge.Finally, it applies
Based on the point of G code profile information and the neighbor relationships of straight line, positioning and measurement to workpiece profile overlap protrusion are completed, is obtained
The starting point and elevation information of the characteristic segments of overlap protrusion.Image procossing and effect such as Fig. 5 a, b, c, d, shown in e, are identified winged
Side protrusion position, is outlined with rectangle frame.
The above embodiments merely illustrate the technical concept and features of the present invention, and its object is to allow person skilled in the art
Scholar cans understand the content of the present invention and implement it accordingly, and it is not intended to limit the scope of the present invention.It is all according to the present invention
Equivalent change or modification made by Spirit Essence, should be covered by the protection scope of the present invention.
Claims (4)
1. a kind of machine vision detection method guiding lower workpiece profile overlap protrusion based on G code, using workpiece profile overlap
Raised vision detection system, the system include industrial camera, light source and camera lens, motion control card, servo-driver, servo electricity
Machine, ball-screw, nut, retarder, detection platform, industrial camera along X-axis with the platform that Y-axis moves by can connect by X-axis
With two axis servo-actuating devices of Y-axis composition, servo nut on every axis and ball-screw in two axis servo-actuating devices are led to
It crosses shaft coupling and connects servo motor with retarder, servo motor connects servo-driver by encoder, and servo-driver passes through
Line connection terminal plate connects host computer and motion control card;Industrial camera, camera lens and light source are placed in detection platform
Workpiece for measurement top, industrial camera are electrically connected host computer, which is characterized in that include the following steps:
Step 1. is analyzed and determines the workpiece figure that the workpiece coordinate system of workpiece to be detected is obtained with industrial camera in detection platform
Transformational relation between the image coordinate system of piece, in the workpiece coordinate system and image detection characterized by G code based on image
Coordinate system has closely contact in identifying burr protrusion, analyzes and infer the conversion of platform coordinate system and workpiece coordinate system
The transformational relation of relationship and platform coordinate system and image coordinate system obtains workpiece coordinate system and figure with above two transformational relation
As the transformational relation of coordinate system;
Step 2. carries out region of interesting extraction to the collected picture of industrial camera, and is directed to region of interest area image, takes
The method of image binaryzation and edge detection obtains the contour edge for needing to detect;
Step 3. carries out overlap bump defects based on the profile information in G code, to area-of-interest profile and judges identification, uses
Minimum Enclosing Rectangle method surrounds contour edge protrusion, realizes the vision-based detection of contour edge protrusion, completes to workpiece profile overlap
The positioning and measurement of protrusion, obtain the starting point and elevation information of the characteristic segments of overlap protrusion.
2. the machine vision detection method according to claim 1 that lower workpiece profile overlap protrusion is guided based on G code,
It is characterized in that:The corresponding coordinate of G code is mapped in image coordinate, root by the transformation relation obtained according to step 1 in step 2
It is line segment or circular arc according to line style, carries out region of interesting extraction;The ROI extractions of line segment type:According to starting point P1, the terminal of line segment
P2 and line segment length P1P2, draw one using line segment P1P2 as the rectangle of neutrality line, a height of overlap highest empirical value of rectangle
K, the ROI extractions of circular arc type:Circular arc starting point P3 and terminal P4 is connected, is drawn using line segment P3P4 as the rectangle of neutrality line, Gao Weijing
Test value K, you can obtain area-of-interest and needed to the method that area-of-interest takes image binaryzation and edge detection
The contour edge of detection.
3. the machine vision detection method according to claim 1 that lower workpiece profile overlap protrusion is guided based on G code,
It is characterized in that:Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, for being judged as the G generations of line segment
Code section, the line segment starting point coordinate preserved after parsing are set as P1(X1, y1),P2(X2, y2), the detection algorithm of overlap protrusion is:
1)According to two point P1, P2 on line segment, the general expression linear equation of straight line L where releasing line segment;
2) number at the edge midpoints point set V1 containing overlap is calculated;
3) i-th of point Pi is taken in point set V, and calculates the i-th point of distance Dist for arriving straight line L;
4) judge that Dist and standard overlap protrusion judge the relationship of distance D deviations, if Dist>D deviations are then stored in point Pi
In one new point set V2;
5) minimum rectangle that straight line L is parallel to using one side surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates rectangle
It is wide and high, unit pixel;
6) image coordinate is converted by world coordinates according to camera calibration parameter, while calculates reality of the rectangle in world coordinates
Border is wide and high, unit mm, can be obtained by the practical wide high of overlap protrusion and the position occurred at this time.
4. the machine vision detection method according to claim 1 that lower workpiece profile overlap protrusion is guided based on G code,
It is characterized in that:Overlap bump defects are carried out to area-of-interest profile in step 3 and judge identification, for the G code section of circular arc
Identification decision method:
1)According to terminal starting point two point coordinates P3, P4 and radius R, the general expression an arc equation of circle O where releasing circular arc;
2)The central angle for judging circular arc then carries out circular arc segmentation if it is greater than 15 °, according to central angle no more than 15 ° standard into
Row segmentation, can be divided into M sections by the edge point set containing overlap protrusion;
3) cyclic variable j=0 is set, Pj sections are taken in M sections of point set;
4)I-th of point Pi is taken in point set Vj, and calculates the i-th point of distance Dist for arriving center of circle O;
5) relationship of Dist and D deviations is judged, if Dist>D deviations are then stored in point Pi in one new point set V2, no
It does not handle then;
6) usable floor area minimum rectangle surrounds point set V2, extracts 4 angular coordinates of rectangle, calculates the width and height of rectangle, unit is
pixel;
7)It converts image coordinate to world coordinates according to camera calibration parameter, while calculating reality of the rectangle in world coordinates
Wide and high, unit mm can be obtained by practical position width height and occurred of Pj sections of overlap protrusions at this time.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810426396.4A CN108765378B (en) | 2018-05-07 | 2018-05-07 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810426396.4A CN108765378B (en) | 2018-05-07 | 2018-05-07 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108765378A true CN108765378A (en) | 2018-11-06 |
CN108765378B CN108765378B (en) | 2021-07-09 |
Family
ID=64010017
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810426396.4A Active CN108765378B (en) | 2018-05-07 | 2018-05-07 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108765378B (en) |
Cited By (16)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109978874A (en) * | 2019-04-02 | 2019-07-05 | 湖南大学 | A kind of rail surface defects vision inspection apparatus and recognition methods |
CN110163853A (en) * | 2019-05-14 | 2019-08-23 | 广东奥普特科技股份有限公司 | A kind of detection method of edge defect |
CN110263204A (en) * | 2019-06-05 | 2019-09-20 | 广州文冲船厂有限责任公司 | A kind of component contour coding method, device and equipment |
CN110455182A (en) * | 2019-07-23 | 2019-11-15 | 中广核检测技术有限公司 | A method of control rod guide card abrasion loss is measured based on image recognition technology |
CN111915581A (en) * | 2020-07-27 | 2020-11-10 | 青岛大学 | Method for detecting defects of smooth metal cambered surface |
CN112017232A (en) * | 2020-08-31 | 2020-12-01 | 浙江水晶光电科技股份有限公司 | Method, device and equipment for positioning circular pattern in image |
CN112215891A (en) * | 2020-07-13 | 2021-01-12 | 浙江大学山东工业技术研究院 | Visual positioning method and system for glue injection hole and pin hole of aluminum profile door and window |
CN112233063A (en) * | 2020-09-14 | 2021-01-15 | 东南大学 | Circle center positioning method for large-size round object |
CN112329501A (en) * | 2020-01-08 | 2021-02-05 | 沈阳和研科技有限公司 | Method for detecting workpiece shape by dicing saw |
CN112489009A (en) * | 2020-11-27 | 2021-03-12 | 芜湖哈特机器人产业技术研究院有限公司 | Pump body mouth ring pose detection method based on visual image |
CN108765378B (en) * | 2018-05-07 | 2021-07-09 | 上海理工大学 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
CN114022483A (en) * | 2022-01-08 | 2022-02-08 | 南通欣斯特机械制造有限公司 | Injection molding flash area identification method based on edge characteristics |
CN114441499A (en) * | 2022-04-11 | 2022-05-06 | 天津美腾科技股份有限公司 | Grade detection method and device, identification equipment, ore pulp grade instrument and storage medium |
WO2022222467A1 (en) * | 2021-04-22 | 2022-10-27 | 苏州华兴源创科技股份有限公司 | Open circular ring workpiece appearance defect detection method and system, and computer storage medium |
CN115308222A (en) * | 2022-07-11 | 2022-11-08 | 江苏汤谷智能科技有限公司 | System and method for identifying bad chip appearance based on machine vision |
CN115753791A (en) * | 2022-11-10 | 2023-03-07 | 哈尔滨耐是智能科技有限公司 | Defect detection method, device and system based on machine vision |
Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0292644A1 (en) * | 1987-05-21 | 1988-11-30 | VDO Adolf Schindling AG | Method of programming a digital control device |
US20060091217A1 (en) * | 2004-11-03 | 2006-05-04 | The Code Corporation | Graphical code reader that is configured for efficient decoder management |
CN103733223A (en) * | 2011-08-04 | 2014-04-16 | 三菱电机株式会社 | Method and system for determining defect of surface of model of object |
CN104933220A (en) * | 2015-05-11 | 2015-09-23 | 东莞市凌英模具塑胶有限公司 | High precision manufacturing method and injection mold for plastic injection mold for complex curved surface |
CN106020120A (en) * | 2016-07-29 | 2016-10-12 | 芜湖哈特机器人产业技术研究院有限公司 | Method for generating G code by using image based on ios system |
CN106709911A (en) * | 2016-12-26 | 2017-05-24 | 西北工业大学 | High-speed rail fastener detection and counting method and system based on machine vision |
CN107533769A (en) * | 2015-03-16 | 2018-01-02 | 彩滋公司 | Prepare the automatic computing engine Computer Aided Design in cutting heat bonding film |
Family Cites Families (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108765378B (en) * | 2018-05-07 | 2021-07-09 | 上海理工大学 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
-
2018
- 2018-05-07 CN CN201810426396.4A patent/CN108765378B/en active Active
Patent Citations (7)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP0292644A1 (en) * | 1987-05-21 | 1988-11-30 | VDO Adolf Schindling AG | Method of programming a digital control device |
US20060091217A1 (en) * | 2004-11-03 | 2006-05-04 | The Code Corporation | Graphical code reader that is configured for efficient decoder management |
CN103733223A (en) * | 2011-08-04 | 2014-04-16 | 三菱电机株式会社 | Method and system for determining defect of surface of model of object |
CN107533769A (en) * | 2015-03-16 | 2018-01-02 | 彩滋公司 | Prepare the automatic computing engine Computer Aided Design in cutting heat bonding film |
CN104933220A (en) * | 2015-05-11 | 2015-09-23 | 东莞市凌英模具塑胶有限公司 | High precision manufacturing method and injection mold for plastic injection mold for complex curved surface |
CN106020120A (en) * | 2016-07-29 | 2016-10-12 | 芜湖哈特机器人产业技术研究院有限公司 | Method for generating G code by using image based on ios system |
CN106709911A (en) * | 2016-12-26 | 2017-05-24 | 西北工业大学 | High-speed rail fastener detection and counting method and system based on machine vision |
Non-Patent Citations (2)
Title |
---|
YOSAFAT SURYA MURIJANTO等: ""MACHINE VISION IMPLEMENTATION IN RAPID PCB PROTOTYPING IMPLEMENTASI MACHINE VISION PADA PEMBUATAN DUPLIKASI PCB"", 《JOURNAL OF MECHATRONICS, ELECTRICAL POWER, AND VEHICULAR TECHNOLOGY》 * |
冯勋壮等: ""铸件自动循迹浮动磨边的研究"", 《现代制造技术与装备》 * |
Cited By (24)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN108765378B (en) * | 2018-05-07 | 2021-07-09 | 上海理工大学 | Machine vision detection method for workpiece contour flash bulge under guidance of G code |
CN109978874B (en) * | 2019-04-02 | 2023-03-14 | 湖南大学 | Visual detection device and identification method for surface defects of steel rail |
CN109978874A (en) * | 2019-04-02 | 2019-07-05 | 湖南大学 | A kind of rail surface defects vision inspection apparatus and recognition methods |
CN110163853A (en) * | 2019-05-14 | 2019-08-23 | 广东奥普特科技股份有限公司 | A kind of detection method of edge defect |
CN110163853B (en) * | 2019-05-14 | 2021-05-25 | 广东奥普特科技股份有限公司 | Edge defect detection method |
CN110263204A (en) * | 2019-06-05 | 2019-09-20 | 广州文冲船厂有限责任公司 | A kind of component contour coding method, device and equipment |
CN110455182A (en) * | 2019-07-23 | 2019-11-15 | 中广核检测技术有限公司 | A method of control rod guide card abrasion loss is measured based on image recognition technology |
CN112329501A (en) * | 2020-01-08 | 2021-02-05 | 沈阳和研科技有限公司 | Method for detecting workpiece shape by dicing saw |
CN112215891A (en) * | 2020-07-13 | 2021-01-12 | 浙江大学山东工业技术研究院 | Visual positioning method and system for glue injection hole and pin hole of aluminum profile door and window |
CN111915581A (en) * | 2020-07-27 | 2020-11-10 | 青岛大学 | Method for detecting defects of smooth metal cambered surface |
CN112017232A (en) * | 2020-08-31 | 2020-12-01 | 浙江水晶光电科技股份有限公司 | Method, device and equipment for positioning circular pattern in image |
CN112017232B (en) * | 2020-08-31 | 2024-03-15 | 浙江水晶光电科技股份有限公司 | Positioning method, device and equipment for circular patterns in image |
CN112233063A (en) * | 2020-09-14 | 2021-01-15 | 东南大学 | Circle center positioning method for large-size round object |
CN112233063B (en) * | 2020-09-14 | 2024-02-13 | 东南大学 | Circle center positioning method for large-size round object |
CN112489009A (en) * | 2020-11-27 | 2021-03-12 | 芜湖哈特机器人产业技术研究院有限公司 | Pump body mouth ring pose detection method based on visual image |
CN112489009B (en) * | 2020-11-27 | 2022-07-26 | 芜湖哈特机器人产业技术研究院有限公司 | Pump body mouth ring pose detection method based on visual image |
WO2022222467A1 (en) * | 2021-04-22 | 2022-10-27 | 苏州华兴源创科技股份有限公司 | Open circular ring workpiece appearance defect detection method and system, and computer storage medium |
CN114022483A (en) * | 2022-01-08 | 2022-02-08 | 南通欣斯特机械制造有限公司 | Injection molding flash area identification method based on edge characteristics |
CN114022483B (en) * | 2022-01-08 | 2022-03-25 | 南通欣斯特机械制造有限公司 | Injection molding flash area identification method based on edge characteristics |
CN114441499A (en) * | 2022-04-11 | 2022-05-06 | 天津美腾科技股份有限公司 | Grade detection method and device, identification equipment, ore pulp grade instrument and storage medium |
CN115308222A (en) * | 2022-07-11 | 2022-11-08 | 江苏汤谷智能科技有限公司 | System and method for identifying bad chip appearance based on machine vision |
CN115308222B (en) * | 2022-07-11 | 2024-02-09 | 江苏汤谷智能科技有限公司 | System and method for identifying poor chip appearance based on machine vision |
CN115753791A (en) * | 2022-11-10 | 2023-03-07 | 哈尔滨耐是智能科技有限公司 | Defect detection method, device and system based on machine vision |
CN115753791B (en) * | 2022-11-10 | 2024-03-01 | 哈尔滨耐是智能科技有限公司 | Defect detection method, device and system based on machine vision |
Also Published As
Publication number | Publication date |
---|---|
CN108765378B (en) | 2021-07-09 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108765378A (en) | The machine vision detection method of lower workpiece profile overlap protrusion is guided based on G code | |
CN111080693A (en) | Robot autonomous classification grabbing method based on YOLOv3 | |
JP3522280B2 (en) | Method and apparatus for a ball bond inspection system | |
CN106952250A (en) | A kind of metal plate and belt detection method of surface flaw and device based on Faster R CNN networks | |
JP4571763B2 (en) | Image processing apparatus and bonding apparatus | |
CN108344743A (en) | One kind being based on machine vision drug blister package defect inspection method and system | |
CN107393270A (en) | A kind of portable vision inspection device and method for electric meter detection | |
CN110400315A (en) | A kind of defect inspection method, apparatus and system | |
CN110991360B (en) | Robot inspection point position intelligent configuration method based on visual algorithm | |
TWI628415B (en) | Positioning and measuring system based on image scale | |
CN113340909B (en) | Glue line defect detection method based on machine vision | |
CN110533654A (en) | The method for detecting abnormality and device of components | |
CN103308524A (en) | PCB automatic optical inspection system | |
CN110186375A (en) | Intelligent high-speed rail white body assemble welding feature detection device and detection method | |
CN107703513A (en) | A kind of novel non-contact contact net relative position detection method based on image procossing | |
CN112964724A (en) | Multi-target multi-zone visual detection method and system | |
CN112233175A (en) | Chip positioning method based on YOLOv3-tiny algorithm and integrated positioning platform | |
Han et al. | SSGD: A smartphone screen glass dataset for defect detection | |
CN104766330B (en) | A kind of image processing method and electronic equipment | |
Li et al. | Autofeeding system for assembling the CBCs on automobile engine based on 3-D vision guidance | |
CN113869407A (en) | Monocular vision-based vehicle length measuring method and device | |
CN104677906A (en) | Image information detecting method | |
He et al. | Research on defect detection of the outer side of bottle cap based on high angle and multi-view vision system | |
CN110807416A (en) | Digital instrument intelligent recognition device and method suitable for mobile detection device | |
Sun et al. | Cascaded detection method for surface defects of lead frame based on high-resolution detection images |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant | ||
TR01 | Transfer of patent right |
Effective date of registration: 20231116 Address after: 239000 No. 857, Shanghai South Road, Chengdong Industrial Park, Chuzhou City, Anhui Province Patentee after: Puwanini Intelligent Equipment Co.,Ltd. Address before: 200093 No. 516, military road, Shanghai, Yangpu District Patentee before: University of Shanghai for Science and Technology |
|
TR01 | Transfer of patent right |