CN109225941A - Tapping internal thread situation detects and sorting system and method automatically - Google Patents
Tapping internal thread situation detects and sorting system and method automatically Download PDFInfo
- Publication number
- CN109225941A CN109225941A CN201811175978.6A CN201811175978A CN109225941A CN 109225941 A CN109225941 A CN 109225941A CN 201811175978 A CN201811175978 A CN 201811175978A CN 109225941 A CN109225941 A CN 109225941A
- Authority
- CN
- China
- Prior art keywords
- region
- screw hole
- workpiece
- light source
- image
- 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
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
- B07C5/342—Sorting according to other particular properties according to optical properties, e.g. colour
- B07C5/3422—Sorting according to other particular properties according to optical properties, e.g. colour using video scanning devices, e.g. TV-cameras
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/02—Measures preceding sorting, e.g. arranging articles in a stream orientating
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C2501/00—Sorting according to a characteristic or feature of the articles or material to be sorted
- B07C2501/0009—Sorting of fasteners, e.g. screws, nuts, bolts
Landscapes
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a kind of tapping internal thread situations to detect automatically and sorting system and method, including workpiece pipeline, photoelectric sensor, annular light source, light source controller, industrial camera, industrial personal computer, multi-axis robot；The workpiece pipeline includes detection station and discharge station, and the annular light source is connected with light source controller；Its external trigger interface of the industrial camera is connected with photoelectric sensor output, triggers the acquisition of image, and be transmitted to industrial personal computer by camera interface；Photoelectric sensor is connected with industrial camera and light source controller；The industrial personal computer is connected with industrial camera, multi-freedom robot controller, receives the digital image information of industrial camera, and the physical coordinates that location information is converted to multi-freedom robot movement are sent to multi-freedom robot controller and sort；The present invention realizes the real-time quick detection to workpiece tapping internal thread situation using machine vision technique, and the detection efficiency and the degree of automation of tapping internal thread situation greatly improved.
Description
Technical field
The invention patent belongs to tapping internal thread situation automatic measurement technique field, and in particular to a kind of to use machine vision
The tapping internal thread situation that technology is combined with multi-freedom robot detects automatically and sorting equipment and method.
Technical background
Whorl work piece be it is most important in industrial production and life be fastenedly connected and transmission component, thread connection application is wide
It is general, and for internal screw thread detection, manufacturing enterprise mostly uses greatly the method manually estimated to complete at present, artificial detection very great Cheng
It is influenced on degree vulnerable to factors such as subjective factor, individual differences, since work-piece throughput is larger, this method does not require nothing more than worker and carries out height
The labour of intensity, and the product of defect missing inspection can bring security risk, therefore, to the efficient, quasi- of workpiece tapping internal thread situation
True intellectualized detection is the basic guarantee for realizing automated production.
Through retrieving, patent CN106226316, CN202823923U and CN107289847 are disclosed based on machine vision
Screw thread automatic checkout system, but its design is directed to external screw thread context of detection；And in terms of interior thread measurement, patent
CN106644447 realizes the inspection of tapping internal thread situation by parameters such as torsion, stroke, the times of detection rotary part
It surveys, patent CN204115651 proposes a kind of pneumatic internal thread detection device for automotive hub, and the patent is with mechanical force
Method for sensing feeds back current thread tapping situation, and the implementation of this method requires the accurate positioning to threaded hole first, is unable to satisfy
The changeable situation of threaded hole；Patent CN201610236353 proposes gas medium laser-Doppler formula internal screw thread detector, should
Method structure of the detecting device complexity, higher cost, are difficult to realize and promote on a large scale.
Summary of the invention
1, the purpose of the present invention
For the above-mentioned deficiency of existing tapping internal thread situation automatic detection device, the technical problems to be solved by the invention
It is to propose that a kind of tapping internal thread situation combined using machine vision technique with multi-freedom robot is detected automatically and divided
Pick device and method.
2, used technical solution of the invention
It is detected automatically the invention discloses a kind of tapping internal thread situation and sorting system, including workpiece pipeline, photoelectricity
Sensor, annular light source, light source controller, industrial camera, industrial personal computer, multi-axis robot；
The workpiece pipeline includes detection station and discharge station；
The annular light source is connected with light source controller, has a diameter larger than the diameter of screw thread, is mounted on detection station detection
It on line bracket, is illuminated with light field front low angle, screw hole via regions is made to form dark field, and internal screw thread internal wall features are enhanced；
Photoelectric sensor is mounted in detection station, is connected with industrial camera and light source controller, and position is located at annular
Light source is vertical with production line traffic direction diametrically, reaches the diametrical position instantaneous trigger work with the annular light source in screw thread
The shooting of industry camera, triggering light source starting, and light-source brightness is adjusted to by suitable intensity by potentiometer；
What the industrial camera had certain tilt angle is fixed on detection line bracket, in central region and annular light source
The heart is concentric, and light source covers examined workpiece entirety, obtains complete screw hole and screw thread imaging features；Its external trigger interface and photoelectric transfer
Sensor output is connected, and triggers the acquisition of image, and be transmitted to industrial personal computer by camera interface；
The industrial personal computer is connected with industrial camera, multi-freedom robot controller, receives the digital picture of industrial camera
Information, by location information convert to multi-freedom robot move physical coordinates be sent to multi-freedom robot controller into
Row sorting；
The industrial personal computer is primary processor, is connected with industrial camera by camera interface, is realized to the real-time of digital picture
Acquisition and analysis, and examined workpiece tapping internal thread situation and Work position information are converted into multi-freedom robot movement
Physical coordinates, multi-freedom robot controller is sent to by data-interface；
The multi-freedom robot carries out sort operation for realizing to different quality workpiece, and front end clamping jaw can basis
Workpiece different workpieces are replaced, and are fastenedly connected by standard flange and robot body, using digital image processing method
Acquired image is handled and analyzed, to realize the identification to examined workpiece tapping internal thread situation.
Further, by Digital Image Transmission to industrial personal computer, industrial personal computer detects automatically includes the industrial camera with sorting
Average smooth, image pre-segmentation, screw hole extracted region, threaded portion extraction, feature extraction and calculating, classifier design and instruction
Practice, the intelligent sorting of workpiece:
The average smooth refers to using the average template for using 3*3 to acquired internal screw thread imageInto
The correlation filtering of row average smooth is handled, for filtering out the noise spot on internal screw workpiece surface in image；
Screw hole region to be checked is split by described image pre-segmentation using single threshold dividing method with background area, and is adopted
It is extracted with eight neighborhood connected region, by calculating circularity, area and the radius of each connected region, according to Feature Selection, is realized
The pre-segmentation of screw thread bore region；
The screw hole extracted region refers to that the mass center using region-growing method, with above-mentioned pre-segmentation screw hole region is
Seed point, using the equal value difference and region area in the gray value of screw hole region contour pixel and screw hole region as Growing law, i.e.,
It is that the area grayscale average value and the difference before being included into after growth are less than set threshold value that screw hole region adjacent pixel, which is included into condition,
And the screw hole region area after growing is less than given threshold, when being included into condition there is no pixel satisfaction, region growing is calculated
Method terminates, and realizes the accurate segmentation in screw hole region；
The threaded portion extraction refers to centered on the screw hole region mass center obtained after above-mentioned region growing, obtains current
The minimum circumscribed circle in screw hole region, and algebra subtraction is carried out to circumscribed circle region and screw hole region, threaded area can be realized
Regional partition；
The feature extraction and calculating, which refer to, calculates each screw hole region and the color of threaded portion, consistency, circularity, square
The parameters such as shape degree, concavity, spherical property, standard deviation and entropy, and use coarse central algorithm processing region shape and texture descriptor with
The most simple classified description subset for finding out suitable threaded hole and the identification of missed tapping hole realizes that Feature Dimension Reduction and recognition rule extract；
The classifier design and training refer to using the rule coverage degree of rough set and confidence level as heuristic information, to redundancy
Fuzzy rule is simplified, and the threshold value that the Rules control of acquisition is arranged is 0.75, and the threshold value of coverage is 0.1, final to obtain
The strongest sorting criterion of classification capacity differentiates examined workpiece tapping internal thread situation；
The intelligent sorting of the workpiece refers to through workpiece quality information and physical location information acquired in machine vision,
The physical coordinates moved to multi-freedom robot are converted, realize the sorting to examined workpiece.
Further, the front end clamping jaw of the multi-freedom robot 7 is detachable, passes through standard flange and robot sheet
Body is fastenedly connected.
It is detected automatically the invention discloses a kind of tapping internal thread situation and method for sorting, it is characterised in that including walking as follows
It is rapid:
(1), under off-line state, industrial camera is demarcated, corrective lens distortion, and calculates each location of pixels in image
Pixel resolution, phase modifier aperture and focal length, the image for obtaining it is the most clear；
(2), under off-line state, hand and eye calibrating is carried out to industrial camera and multi-freedom robot, establish industrial camera and
Mapping relations between robot coordinate system；
(3), examined workpiece is placed in workpiece pipeline；
(4), when reaching detection station, photoelectric sensor output triggering annular light source illumination and industrial camera Image Acquisition；
(5), master controller obtains and handles current imaging scheme realtime graphic collected；
(6), by image analysis, current examined workpiece tapping internal thread situation is identified, and position the image of examined workpiece
Position, specifically:
(61), average smooth, using the average template for using 3*3 to acquired imageAveragely put down
Sliding correlation filtering processing, for filtering out the noise spot on internal screw workpiece surface in image；
(62), screw hole region to be checked is split by image pre-segmentation using single threshold dividing method with background area, and
It is extracted using eight neighborhood connected region, it is real according to Feature Selection by calculating circularity, area and the radius of each connected region
The pre-segmentation of existing screw thread bore region；
(63), screw hole extracted region is kind with the mass center of above-mentioned pre-segmentation screw hole region using region-growing method
It is sub-, using the equal value difference and region area in the gray value of screw hole region contour pixel and screw hole region as Growing law, i.e. spiral shell
Bore region adjacent pixel be included into condition be growth after area grayscale average value be included into before difference be less than set threshold value, and
Screw hole region area after growth is less than given threshold, when being included into condition there is no pixel satisfaction, algorithm of region growing
Terminate, realizes the accurate segmentation in screw hole region；
(64), threaded portion extracts, and centered on the screw hole region mass center obtained after above-mentioned region growing, obtains current spiral shell
The minimum circumscribed circle of bore region, and algebra subtraction is carried out to circumscribed circle region and screw hole region, threaded portion can be realized
Segmentation；
(65), feature extraction and calculating calculate each screw hole region and the color of threaded portion, consistency, circularity, rectangle
The parameters such as degree, concavity, spherical property, standard deviation and entropy, and use coarse central algorithm processing region shape and texture descriptor to look for
It is suitble to the most simple classified description subset of threaded hole and the identification of missed tapping hole out, realizes that Feature Dimension Reduction and recognition rule extract；
(66), classifier design and training, using the rule coverage degree of rough set and confidence level as heuristic information, to redundancy mould
Paste rule is simplified, and the threshold value that the Rules control of acquisition is arranged is 0.75, and the threshold value of coverage is 0.1, is finally divided
The strongest sorting criterion of class ability differentiates examined workpiece tapping internal thread situation；
(67), the intelligent sorting of workpiece is turned by workpiece quality information and physical location information acquired in machine vision
The physical coordinates of multi-freedom robot movement are shifted to, realize the sorting to examined workpiece；
(7), current examined workpiece quality and location information are converted to multi-freedom robot using hand and eye calibrating result
Three-dimensional motion coordinate；
(8), examined workpiece is clamped and is sorted, be placed in predeterminated position.
3, beneficial effects of the present invention
(1) present invention realizes the real-time quick detection to workpiece tapping internal thread situation using machine vision technique, gram
The detection efficiency and the degree of automation of tapping internal thread situation greatly improved in the drawbacks of having taken traditional detection method；
(2) internal screw thread characteristics of image is carried out quantification treatment by the present invention, improves the accuracy of workpiece quality detection to be checked,
The average time-consuming about 0.256s of single workpiece sensing of the invention, the recall rate of missed tapping workpiece is 95.88%, tapping workpiece
Recall rate be 100%, online detection requirements can be met；
(3) present invention mutually melts machine vision imaging technology, machine vision automatic detection algorithm with multi-freedom robot
It closes, to realize the automated sorting of workpiece to be checked.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of the invention patent；
Fig. 2 is the workpiece imaging effect figure that the invention patent imaging scheme obtains；
Fig. 3 is the electrical connection diagram of the invention patent；
Fig. 4 is the invention patent image processing flow figure；
Fig. 5 is the invention patent workpiece tapping internal thread situation detection effect figure；
In figure: 1 workpiece pipeline；2 photoelectric sensors；3 annular light sources；4 light source controllers；5 industrial cameras；6 industrial personal computers；
7 multi-axis robots；8 detection stations；9 discharge stations；The imaging of 10 inner bolt holes；The imaging of 11 internal screw thread inner walls；The interior spiral shell of 12 non-tapping
Hole；The inner bolt hole of 13 tapping.
Specific embodiment
Detailed technology scheme of the present invention is introduced below in conjunction with attached drawing, detection device of the invention can be applied to spiral shell in workpiece
The automatic detection of line tapping situation.
Embodiment 1
As shown in Figure 1, being detected automatically the invention discloses a kind of tapping internal thread situation and sorting system, including workpiece is defeated
Line sending 1, photoelectric sensor 2, annular light source 3, light source controller 4, industrial camera 5, industrial personal computer 6, multi-axis robot 7；
The workpiece pipeline 1 includes detection station 8 and discharge station 9；
The annular light source 3 is connected with light source controller, has a diameter larger than the diameter of screw thread, is mounted on the inspection of detection station 8
It on survey line bracket, is illuminated with light field front low angle, screw hole via regions is made to form dark field, and internal screw thread internal wall features are increased
By force；
Photoelectric sensor 2 is mounted in detection station 8, is connected with industrial camera 5 and light source controller 4, position is located at
Annular light source 3 is vertical with production line traffic direction diametrically, reaches the diametrical position moment with the annular light source in screw thread
Industrial camera shooting, triggering light source starting are triggered, and light-source brightness is adjusted to by suitable intensity by potentiometer；
The industrial camera 5 is fixed on detection line bracket, central region and annular light source 3 with certain tilt angle
Center is concentric, and light source covers examined workpiece entirety, obtains complete screw hole and screw thread imaging features；Its external trigger interface and photoelectricity
The output of sensor 2 is connected, and triggers the acquisition of image, and be transmitted to industrial personal computer 6 by camera interface；
The industrial personal computer 6 is connected with industrial camera 5, multi-freedom robot controller 7, receives the number of industrial camera 5
Location information is converted to the physical coordinates that multi-freedom robot moves and is sent to multi-freedom robot control by image information
Device 7 is sorted；
The industrial personal computer is primary processor, is connected with industrial camera by camera interface, is realized to the real-time of digital picture
Acquisition and analysis, and examined workpiece tapping internal thread situation and Work position information are converted into multi-freedom robot movement
Physical coordinates, multi-freedom robot controller is sent to by data-interface；
The multi-freedom robot carries out sort operation for realizing to different quality workpiece, and front end clamping jaw can basis
Workpiece different workpieces are replaced, and are fastenedly connected by standard flange and robot body, using digital image processing method
Acquired image is handled and analyzed, to realize the identification to examined workpiece tapping internal thread situation.
Embodiment 2
On the basis of embodiment 1, as shown in Fig. 2, a kind of workpiece imaging effect figure is selected by using above scheme
With annular light source 3, using the low angle direct illumination of light field front, industrial camera 5 is fixed on detection with certain tilt angle
Line bracket, and annular light source 3, industrial camera 5 and 8 center of detection station are concentric makes inner bolt hole imaging 10 form dark fields, and interior spiral shell
It is enhanced that 11 features are imaged in line inner wall, to reach the screw hole of measured workpiece and screw thread imaging features show, and inhibits other not
Important feature.
As shown in figure 3, a kind of tapping internal thread situation detects automatically and the electrical connection of sorting equipment, industrial camera 5 with
The external trigger interface of light source controller 4 is connected to photoelectric sensor；Annular light source 3 and light source controller 4 pass through special interface
It is connected；Industrial camera 5 is connect with industrial personal computer 6 by camera interface；Industrial personal computer 6 is connect with multi-freedom robot 7 by data
Mouth connection.
Workpiece to be checked is detected automatically with method for sorting as shown in figure 4, mainly including average smooth, image pre-segmentation, screw hole
Extracted region, threaded portion extract, the intelligent sorting of feature extraction and calculating, classifier design and training, workpiece；
The average smooth refers to using the average template for using 3*3 to acquired imageIt is averaged
Smooth correlation filtering processing, for filtering out the noise spot of surface of workpiece in image.
Screw hole region to be checked is split by described image pre-segmentation using single threshold dividing method with background area, and is adopted
It is extracted with eight neighborhood connected region, by calculating circularity, area and the radius of each connected region, according to Feature Selection, is realized
The pre-segmentation of screw thread bore region；
The screw hole extracted region refers to that the mass center using region-growing method, with above-mentioned pre-segmentation screw hole region is
Seed point, using the equal value difference and region area in the gray value of screw hole region contour pixel and screw hole region as Growing law, i.e.,
It is that the area grayscale average value and the difference before being included into after growth are less than set threshold value that screw hole region adjacent pixel, which is included into condition,
And the screw hole region area after growing is less than given threshold, when being included into condition there is no pixel satisfaction, region growing is calculated
Method terminates, to obtain complete screw hole imaging region 10；
The threaded portion extraction refers to centered on the screw hole region mass center obtained after above-mentioned region growing, obtains current
The minimum circumscribed circle in screw hole region, and algebra subtraction is carried out to circumscribed circle region and screw hole region, to obtain complete
Screw thread imaging region 11.
The feature extraction and calculating refer to the color for calculating each inner bolt hole region 10 and threaded region 11, consistency,
The parameters such as circularity, rectangular degree, concavity, spherical property, standard deviation and entropy, and use coarse central algorithm processing region shape and texture
The sub most simple classified description subset to find out suitable threaded hole and the identification of missed tapping hole of description, realizes Feature Dimension Reduction and recognition rule
It extracts；
The classifier design and training refer to using the rule coverage degree of rough set and confidence level as heuristic information, to redundancy
Fuzzy rule is simplified, and the threshold value that the Rules control of acquisition is arranged is 0.75, and the threshold value of coverage is 0.1, final to obtain
The strongest sorting criterion of classification capacity differentiates examined workpiece tapping internal thread situation；
As shown in figure 5, a kind of workpiece tapping internal thread situation detection effect figure, the inner bolt hole of non-tapping is marked with white circle
12, the screw hole not marked then is the inner bolt hole 13 of tapping, and machine vision algorithm differentiates that effect is consistent with improvement of visual effect, to test
Above-mentioned algorithm has been demonstrate,proved with good detection effect.
The intelligent sorting of the workpiece refers to through workpiece quality information and physical location information acquired in machine vision,
The physical coordinates moved to multi-freedom robot are converted, realize the sorting to examined workpiece.
Claims (4)
1. a kind of tapping internal thread situation detects automatically and sorting system, it is characterised in that: including workpiece pipeline, photoelectric sensing
Device, annular light source, light source controller, industrial camera, industrial personal computer, multi-axis robot；
The workpiece pipeline includes detection station and discharge station；
The annular light source is connected with light source controller, has a diameter larger than the diameter of screw thread, is mounted on detection station detection line branch
It on frame, is illuminated with light field front low angle, screw hole via regions is made to form dark field, and internal screw thread internal wall features are enhanced；
Photoelectric sensor is mounted in detection station, is connected with industrial camera and light source controller, and position is located at annular light source
It is vertical with production line traffic direction diametrically, screw thread reach and the annular light source diametrical position instantaneous trigger industry phase
Machine shooting, triggering light source starting, and light-source brightness is adjusted to by suitable intensity by potentiometer；
The industrial camera is fixed on detection line bracket with certain tilt angle, and central region and annular light source center are same
The heart, light source cover examined workpiece entirety, obtain complete screw hole and screw thread imaging features；Its external trigger interface and photoelectric sensor
Output is connected, and triggers the acquisition of image, and be transmitted to industrial personal computer by camera interface；
The industrial personal computer is connected with industrial camera, multi-freedom robot controller, receives the digital image information of industrial camera,
The physical coordinates that location information is converted to multi-freedom robot movement are sent to multi-freedom robot controller and are divided
It picks；
The industrial personal computer is primary processor, is connected with industrial camera by camera interface, realizes the real-time acquisition to digital picture
With analysis, and by examined workpiece tapping internal thread situation and Work position information be converted into multi-freedom robot movement object
Coordinate is managed, multi-freedom robot controller is sent to by data-interface；
The multi-freedom robot carries out sort operation for realizing to different quality workpiece, and front end clamping jaw can be according to workpiece
Different workpieces are replaced, and are fastenedly connected by standard flange and robot body, using digital image processing method to institute
Acquisition image is handled and is analyzed, to realize the identification to examined workpiece tapping internal thread situation.
2. tapping internal thread situation according to claim 1 detects automatically and sorting system, it is characterised in that the industry
For camera by Digital Image Transmission to industrial personal computer, it includes average smooth, image pre-segmentation, screw hole area that industrial personal computer, which is detected automatically with sorting,
Domain is extracted, threaded portion extracts, the intelligent sorting of feature extraction and calculating, classifier design and training, workpiece:
The average smooth refers to using the average template for using 3*3 to acquired internal screw thread imageIt carries out flat
Smooth correlation filtering processing, for filtering out the noise spot on internal screw workpiece surface in image；
Screw hole region to be checked is split by described image pre-segmentation using single threshold dividing method with background area, and uses eight
Neighborhood connected region is extracted, and by calculating circularity, area and the radius of each connected region, according to Feature Selection, realizes screw thread
The pre-segmentation of bore region；
The screw hole extracted region refers to using region-growing method, using the mass center of above-mentioned pre-segmentation screw hole region as seed
Point, using the equal value difference and region area in the gray value of screw hole region contour pixel and screw hole region as Growing law, i.e. screw hole
It is that the area grayscale average value and the difference before being included into after growth are less than set threshold value, and life that region adjacent pixel, which is included into condition,
Screw hole region area after length is less than given threshold, when being included into condition there is no pixel satisfaction, algorithm of region growing knot
Beam realizes the accurate segmentation in screw hole region；
The threaded portion extraction refers to centered on the screw hole region mass center obtained after above-mentioned region growing, obtains current screw hole
The minimum circumscribed circle in region, and algebra subtraction is carried out to circumscribed circle region and screw hole region, threaded portion point can be realized
It cuts；
The feature extraction and calculating, which refer to, calculates each screw hole region and the color of threaded portion, consistency, circularity, rectangle
The parameters such as degree, concavity, spherical property, standard deviation and entropy, and use coarse central algorithm processing region shape and texture descriptor to look for
It is suitble to the most simple classified description subset of threaded hole and the identification of missed tapping hole out, realizes that Feature Dimension Reduction and recognition rule extract；
The classifier design and training refer to using the rule coverage degree of rough set and confidence level as heuristic information, fuzzy to redundancy
Rule is simplified, and the threshold value that the Rules control of acquisition is arranged is 0.75, and the threshold value of coverage is 0.1, is finally classified
The strongest sorting criterion of ability differentiates examined workpiece tapping internal thread situation；
The intelligent sorting of the workpiece refers to through workpiece quality information and physical location information acquired in machine vision, conversion
The physical coordinates moved to multi-freedom robot, realize the sorting to examined workpiece.
3. tapping internal thread situation according to claim 1 detects automatically and sorting system, it is characterised in that: it is described mostly from
It is detachable by the front end clamping jaw of degree robot 7, it is fastenedly connected by standard flange and robot body.
4. a kind of tapping internal thread situation detects automatically and method for sorting, it is characterised in that include the following steps:
(1), under off-line state, industrial camera is demarcated, corrective lens distortion, and calculates the picture of each location of pixels in image
Plain resolution ratio, phase modifier aperture and focal length, the image for obtaining it are the most clear；
(2), under off-line state, hand and eye calibrating is carried out to industrial camera and multi-freedom robot, establishes industrial camera and machine
Mapping relations between people's coordinate system；
(3), examined workpiece is placed in workpiece pipeline；
(4), when reaching detection station, photoelectric sensor output triggering annular light source illumination and industrial camera Image Acquisition；
(5), master controller obtains and handles current imaging scheme realtime graphic collected；
(6), by image analysis, current examined workpiece tapping internal thread situation is identified, and position the picture position of examined workpiece,
Specifically:
(61), average smooth, using the average template for using 3*3 to acquired imageCarry out average smooth
Correlation filtering processing, for filtering out the noise spot on internal screw workpiece surface in image；
(62), screw hole region to be checked is split with background area using single threshold dividing method, and used by image pre-segmentation
Eight neighborhood connected region is extracted, and by calculating circularity, area and the radius of each connected region, according to Feature Selection, realizes spiral shell
The pre-segmentation of pit area；
(63), screw hole extracted region, using region-growing method, using the mass center of above-mentioned pre-segmentation screw hole region as seed
Point, using the equal value difference and region area in the gray value of screw hole region contour pixel and screw hole region as Growing law, i.e. screw hole
It is that the area grayscale average value and the difference before being included into after growth are less than set threshold value, and life that region adjacent pixel, which is included into condition,
Screw hole region area after length is less than given threshold, when being included into condition there is no pixel satisfaction, algorithm of region growing knot
Beam realizes the accurate segmentation in screw hole region；
(64), threaded portion extracts, and centered on the screw hole region mass center obtained after above-mentioned region growing, obtains current screw hole area
The minimum circumscribed circle in domain, and algebra subtraction is carried out to circumscribed circle region and screw hole region, threaded portion segmentation can be realized；
(65), feature extraction and calculating, calculate the color in each screw hole region and threaded portion, consistency, circularity, rectangular degree,
The parameters such as concavity, spherical property, standard deviation and entropy, and use coarse central algorithm processing region shape and texture descriptor suitable to find out
The most simple classified description subset for closing threaded hole and the identification of missed tapping hole realizes that Feature Dimension Reduction and recognition rule extract；
(66), classifier design and training, using the rule coverage degree of rough set and confidence level as heuristic information, to the fuzzy rule of redundancy
Then simplified, the threshold value that the Rules control of acquisition is arranged is 0.75, and the threshold value of coverage is 0.1, final to obtain classification energy
The strongest sorting criterion of power differentiates examined workpiece tapping internal thread situation；
(67), the intelligent sorting of workpiece, by workpiece quality information and physical location information acquired in machine vision, conversion is extremely
The physical coordinates of multi-freedom robot movement, realize the sorting to examined workpiece；
(7), current examined workpiece quality and location information are converted to multi-freedom robot three-dimensional using hand and eye calibrating result
The coordinates of motion；
(8), examined workpiece is clamped and is sorted, be placed in predeterminated position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811175978.6A CN109225941B (en) | 2018-10-10 | 2018-10-10 | Automatic detection and sorting system and method for internal thread tapping condition |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811175978.6A CN109225941B (en) | 2018-10-10 | 2018-10-10 | Automatic detection and sorting system and method for internal thread tapping condition |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109225941A true CN109225941A (en) | 2019-01-18 |
| CN109225941B CN109225941B (en) | 2021-03-19 |
Family
ID=65055907
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811175978.6A Active CN109225941B (en) | 2018-10-10 | 2018-10-10 | Automatic detection and sorting system and method for internal thread tapping condition |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109225941B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144426A (en) * | 2019-12-28 | 2020-05-12 | 广东拓斯达科技股份有限公司 | Sorting method, device, equipment and storage medium |
| CN112858312A (en) * | 2021-01-20 | 2021-05-28 | 上海工程技术大学 | Pen cap silicon bead installation visual detection device and method |
| CN113351503A (en) * | 2021-06-23 | 2021-09-07 | 上海宏金设备工程有限公司 | Scaffold detection and sorting system and method |
| CN115880365A (en) * | 2023-03-01 | 2023-03-31 | 超音速人工智能科技股份有限公司 | Double-station automatic screw screwing detection method, system and device |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5995407A (en) * | 1982-11-24 | 1984-06-01 | Nagoya Rashi Seisakusho:Kk | Detector for defective bolt |
| KR100899361B1 (en) * | 2008-12-10 | 2009-05-26 | (주)진합 | Screw thread test device |
| CN105004275A (en) * | 2015-07-21 | 2015-10-28 | 苏州佳祺仕信息科技有限公司 | Threaded hole three-CCD camera detection mechanism |
| CN105066890A (en) * | 2015-07-21 | 2015-11-18 | 苏州佳祺仕信息科技有限公司 | Threaded hole detection device |
| CN106226316A (en) * | 2016-08-31 | 2016-12-14 | 江苏大学 | A kind of single camera wide visual field vision thread detecting device and detection method thereof |
| CN106409711A (en) * | 2016-09-12 | 2017-02-15 | 佛山市南海区广工大数控装备协同创新研究院 | Solar silicon wafer defect detecting system and method |
| CN206235577U (en) * | 2016-11-03 | 2017-06-09 | 上海戈冉泊精模科技有限公司 | A kind of nut thread detection means |
| CN106925529A (en) * | 2017-03-21 | 2017-07-07 | 柳州欧维姆机械股份有限公司 | With the method that internal thread hemicone tooth form and profile automatic checkout system and internal threads hemicone tooth form are detected with profile |
-
2018
- 2018-10-10 CN CN201811175978.6A patent/CN109225941B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPS5995407A (en) * | 1982-11-24 | 1984-06-01 | Nagoya Rashi Seisakusho:Kk | Detector for defective bolt |
| KR100899361B1 (en) * | 2008-12-10 | 2009-05-26 | (주)진합 | Screw thread test device |
| CN105004275A (en) * | 2015-07-21 | 2015-10-28 | 苏州佳祺仕信息科技有限公司 | Threaded hole three-CCD camera detection mechanism |
| CN105066890A (en) * | 2015-07-21 | 2015-11-18 | 苏州佳祺仕信息科技有限公司 | Threaded hole detection device |
| CN106226316A (en) * | 2016-08-31 | 2016-12-14 | 江苏大学 | A kind of single camera wide visual field vision thread detecting device and detection method thereof |
| CN106409711A (en) * | 2016-09-12 | 2017-02-15 | 佛山市南海区广工大数控装备协同创新研究院 | Solar silicon wafer defect detecting system and method |
| CN206235577U (en) * | 2016-11-03 | 2017-06-09 | 上海戈冉泊精模科技有限公司 | A kind of nut thread detection means |
| CN106925529A (en) * | 2017-03-21 | 2017-07-07 | 柳州欧维姆机械股份有限公司 | With the method that internal thread hemicone tooth form and profile automatic checkout system and internal threads hemicone tooth form are detected with profile |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111144426A (en) * | 2019-12-28 | 2020-05-12 | 广东拓斯达科技股份有限公司 | Sorting method, device, equipment and storage medium |
| CN112858312A (en) * | 2021-01-20 | 2021-05-28 | 上海工程技术大学 | Pen cap silicon bead installation visual detection device and method |
| CN113351503A (en) * | 2021-06-23 | 2021-09-07 | 上海宏金设备工程有限公司 | Scaffold detection and sorting system and method |
| CN113351503B (en) * | 2021-06-23 | 2023-02-24 | 上海宏金设备工程有限公司 | Scaffold detection and sorting system and method |
| CN115880365A (en) * | 2023-03-01 | 2023-03-31 | 超音速人工智能科技股份有限公司 | Double-station automatic screw screwing detection method, system and device |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109225941B (en) | 2021-03-19 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109225941A (en) | Tapping internal thread situation detects and sorting system and method automatically | |
| CN107804514B (en) | Toothbrush sorting method based on image recognition | |
| CN116879308A (en) | Industrial machine vision system image processing method | |
| CN107154039B (en) | Rubber tube online defect detection method | |
| CN110216080A (en) | Quality monitoring system of PCB processing production line based on image contrast | |
| CN110108711A (en) | The vision detection system of annulus side wall defect | |
| CN103454285A (en) | Transmission chain quality detection system based on machine vision | |
| CN111266315A (en) | Ore material online sorting system and method based on visual analysis | |
| CN114241298A (en) | Tower crane environment target detection method and system based on laser radar and image fusion | |
| CN106824816A (en) | A kind of PE based on machine vision bottles of detection and method for sorting | |
| CN110108712A (en) | Multifunctional visual sense defect detecting system | |
| CN102221559A (en) | Online automatic detection method of fabric defects based on machine vision and device thereof | |
| CN112184648A (en) | Piston surface defect detection method and system based on deep learning | |
| CN106709529B (en) | Visual detection method for photovoltaic cell color difference classification | |
| CN109848073A (en) | A kind of apparatus and method for sorting coal and gangue | |
| Hsu et al. | Development of a faster classification system for metal parts using machine vision under different lighting environments | |
| CN1643543A (en) | Method for linking edges in stereo images into chains | |
| CN110763600A (en) | Real-time online detection device for suspended particles | |
| CN114155301A (en) | Robot target positioning and grabbing method based on Mask R-CNN and binocular camera | |
| CN113706498A (en) | Visual detection device for lithium battery pole piece | |
| CN108109154A (en) | A kind of new positioning of workpiece and data capture method | |
| CN116486287A (en) | Target detection method and system based on environment self-adaptive robot vision system | |
| CN113822810A (en) | Method for positioning workpiece in three-dimensional space based on machine vision | |
| CN106097323B (en) | Engine cylinder block casting positioning method based on machine vision | |
| CN114387262A (en) | Nut positioning detection method, device and system based on machine vision |
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 |