CN109447989A - Defect detecting device and method based on motor copper bar burr growth district - Google Patents
Defect detecting device and method based on motor copper bar burr growth district Download PDFInfo
- Publication number
- CN109447989A CN109447989A CN201910014867.5A CN201910014867A CN109447989A CN 109447989 A CN109447989 A CN 109447989A CN 201910014867 A CN201910014867 A CN 201910014867A CN 109447989 A CN109447989 A CN 109447989A
- Authority
- CN
- China
- Prior art keywords
- copper bar
- image
- region
- burr
- formula
- 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.)
- Pending
Links
- 229910052802 copper Inorganic materials 0.000 title claims abstract description 77
- 239000010949 copper Substances 0.000 title claims abstract description 77
- RYGMFSIKBFXOCR-UHFFFAOYSA-N Copper Chemical compound [Cu] RYGMFSIKBFXOCR-UHFFFAOYSA-N 0.000 title claims abstract description 76
- 230000007547 defect Effects 0.000 title claims abstract description 36
- 238000000034 method Methods 0.000 title claims abstract description 31
- 238000001514 detection method Methods 0.000 claims abstract description 25
- 238000003708 edge detection Methods 0.000 claims abstract description 13
- 238000000605 extraction Methods 0.000 claims abstract description 7
- 238000013519 translation Methods 0.000 claims abstract description 6
- 238000001914 filtration Methods 0.000 claims abstract description 5
- 230000011218 segmentation Effects 0.000 claims description 8
- 238000012937 correction Methods 0.000 claims description 7
- 238000004519 manufacturing process Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 5
- 238000012545 processing Methods 0.000 claims description 5
- 238000005286 illumination Methods 0.000 claims description 4
- 230000000877 morphologic effect Effects 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000005260 corrosion Methods 0.000 claims description 3
- 230000007797 corrosion Effects 0.000 claims description 3
- 230000010339 dilation Effects 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 239000000284 extract Substances 0.000 claims description 3
- 210000001217 buttock Anatomy 0.000 claims description 2
- 238000004891 communication Methods 0.000 claims description 2
- 238000009792 diffusion process Methods 0.000 claims description 2
- 230000009466 transformation Effects 0.000 claims description 2
- 230000000007 visual effect Effects 0.000 claims description 2
- 230000007812 deficiency Effects 0.000 abstract 1
- 238000005299 abrasion Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000002372 labelling Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000005674 electromagnetic induction Effects 0.000 description 1
- 230000017525 heat dissipation Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 238000011897 real-time detection Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 238000003786 synthesis reaction 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/20—Image enhancement or restoration by the use of local operators
- G06T5/30—Erosion or dilatation, e.g. thinning
-
- G06T5/94—
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
- G06T2207/20032—Median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
Abstract
The present invention in order to solve in current industrial application frequently with image background differential technique present in deficiency, a kind of defect detecting device and method based on motor copper bar burr growth district is provided, for the various types burr generated in motor copper bar hole slot process, in the way of improved median filtering and mask operation obtains copper bar standard picture to be detected, and copper bar Image Edge-Detection region is constructed by Morphology Algorithm, finally carry out feature extraction, each region of tag image simultaneously makes classifier, threshold value is defined, burr is detected.Detection device includes CCD industrial camera, image pick-up card, LED light source, electronic control translation stage, precision stage, articles holding table, hood, manually controls instrument, position sensor and computer.
Description
Technical field
The present invention relates to a kind of defect detecting devices and method based on motor copper bar burr growth district, belong to quality inspection
Survey field.
Background technique
Motor is a kind of a kind of calutron that the foundation law of electromagnetic induction is realized electric energy conversion or transmitted, and by rotor and is determined
Sub- two parts composition.Rotor position applied to the large-size machine under industrial environment, which is set up, counts into special flute profile, then will be electric
Machine copper bar wraps in one layer of insulating film embedded groove, then end ring and copper bar is welded into one, and forms rotor structure.In order to control motor
Temperature obtains good heat dissipation, and the copper bar as rotor wire is generally required in some holes of intermediate cut, but is also generated simultaneously
Burr has buried very big hidden danger for motor.
The detection device and method for mostly using image background difference in industrial application at present carry out burr to motor copper bar
Detection obtains background image that is, using the smoothness of local gray level and change of gradient information adjustment different zones in image
And pretreated image is subjected to difference operation with it, obtain defect image.But algorithm operation in entire image, and meeting
Indistinguishably the defects of scratch of device to hole head position, abrasion, is detected, and weakens burr detection accuracy.In addition, algorithm is general
Property it is poor, high-precision detection effectively can not be carried out to motor copper bar various types burr, to a large amount of motor copper bar images
In actually detected process, erroneous detection easily occurs, brings biggish trouble to work on the spot.
Summary of the invention
For the above-mentioned prior art, the present invention provides a kind of defect detecting device based on motor copper bar burr growth district
And method, to solve the problems of above-mentioned background differential technique.
A kind of defect detecting device based on motor copper bar burr growth district, including CCD industrial camera, Image Acquisition
Card, electronic control translation stage, precision stage, articles holding table, hood, manually controls instrument, position sensor and computer at LED light source；
The position sensor is set to workpiece position to be detected, triggers the work of CCD industrial camera by generating pulse signal；Institute
The LED light source stated diffuses illumination using the low angle of ring structure, is set to below CCD industrial camera；CCD industrial camera
Lower section is provided with precision stage, the copper bar workpiece with hole slot is placed on the articles holding table of precision stage and by automatically controlled
Translation stage controls speed and the track of articles holding table；Image pick-up card and CCD industrial camera are connected by buttock line and and computer
Communication.
The CCD industrial camera is disposed vertically right above workpiece position to be detected, and CCD industrial camera uses
Hood shielding environment shadow is rung, and camera lens visual field is covered in the articles holding table center of precision stage, can be once to 2 copper bars
Workpiece hole slot is shot.
The collected copper bar image of image pick-up card is carried out analysis and algorithm operation by the computer, and passes through classification
Device picks the various types burr in motor copper bar hole slot.
A kind of defect inspection method based on motor copper bar burr growth district, comprising the following steps:
Step 1 obtains copper bar standard picture to be detected.
Step 2 constructs copper bar Image Edge-Detection region.
Step 3 feature extraction and veining defect determine.
Further, the step 1 obtains copper bar standard picture to be detected, the specific steps are as follows:
Acquisition motor copper bar image simultaneously pre-processes it, the pixel region for being 3 × 3 in window size, with point (x, y)
Centered on point, choose 8 kinds of representative edges judge whether the point is edge:
Threshold value T is set up, crosses to obtain 4 adjacent pixels, the grey scale pixel value and adjacent pixel gray value with center
Difference be D, as D < T, G from plus 1, as a < G < b, then can determine the pixel be marginal point, wherein a, b are constant, G table
Show adjacent pixel number similar with center pixel gray scale；If a < G < b, the i.e. point are marginal point, without any transformation,
Directly this grey scale pixel value f (x, y) is exported, it will after otherwise being filtered by formula 1Output:
In formula, SxyIt indicates centered on point (x, y), the window subgraph set of coordinates that size is 3 × 3, g (s, t) represents neck
The grey scale pixel value of domain coordinate,Indicate gray value of this after median filter process.
Use 8 mode of communicating tag image regions and using area as threshold value find out the boundary rectangle comprising maximum region and its
Parameter tn constructs the mask of one completely black (gray value 0), obtains centre using maximum boundary rectangle parameter and be left white region (i.e.
In label rectangular area 1) gray value is set as, and gained mask and copper bar image are carried out exposure mask operation, eliminates copper bar workpiece back
The influence of scape.
Ibid, the mask of one width of construction complete white (gray value 1), obtain centre stay black rectangular area and with exposure mask operation institute
It obtains result and carries out summation operation, eliminate the influence of copper bar workpiece plate face, obtain copper bar standard picture I_mask_white to be detected.
Further, the step 2 constructs copper bar Image Edge-Detection region, the specific steps are as follows:
Gaussian filter is created first, to I_mask_white image filtering；Threshold segmentation, w are carried out using OTSU method0With
w1The respectively ratio of prospect points and the shared image of background points, u0And u1The respectively average gray value of foreground and background.Note
T is the segmentation threshold of prospect and background, and the overall average gray scale of image is u=w0×u0+w1×u1From minimal gray
It is worth maximum gradation value traversal t, when t makes formula 2 maximum, t at this time is optimal segmentation threshold:
G=w0×(u0-u)2+w1×(u1-u)2 (2)
F (x, y) is the workpiece image being partitioned into formula, and t is above-mentioned optimal segmenting threshold.
Morphological scale-space completion notch is carried out to the image after Threshold segmentation, then simultaneously by connected component labeling image-region
Largest connected region labeling is found out to obtain the angle, θ in largest connected region, if θ=0 is directly entered in next step, is otherwise carried out
Image rotation correction makes its angle, θ=0.
Closed operation is made to the image F (x, y) after overcorrection with structural element S and opening operation obtains template image X:
It is accorded in formula for closed operation,It is accorded with for opening operation,For dilation operation symbol, Θ is erosion operation symbol, structural element S
Diameter is 8, and element is all 1, is collar plate shape structure.
Obtained template image X is corroded to obtain IM image with the structural element S that diameter is 8, as shown in Equation 6:
X Θ S=∩ { X-s/s ∈ S } (6)
The larger-diameter collar plate shape structural element corrosion IM of setting one obtains IM_erode image, and cuts IM_ with IM
Erode carries out image operation, obtains copper bar Image Edge-Detection region.
Further, step 3 feature extraction and veining defect determine, the specific steps are as follows:
By in step 1 after rotational correction copper bar standard picture I_mask_white to be detected and step 2 obtained in
Copper bar Image Edge-Detection region carries out the multiplication of pixel point, extracts burr feature.
With each regions of 8 mode of communicating tag images and classifier is made, defines threshold value, various types burr is carried out
Algorithm determines, and the various types burr areas minimum circumscribed rectangle frame of classifier identification is selected, and original after correction
Its coordinate information is shown on image, is stored in system, is completed the detection of veining defect.
Compared with prior art, the invention has the following beneficial technical effects:
Structure is simple, fast response time, and can complete the real-time detection of industry spot big data quantity: this programme is based on motor copper
Burr growth district is arranged, is masked the background and the covering of workpiece plate face in copper bar image incoherent non-using exposure mask algorithm
Detection zone effectively improves the execution efficiency of algorithm, optimizes picture quality and detection accuracy.
Detection accuracy is high: this programme utilizes the region characteristic of motor copper bar burr growth, by constructing copper bar image border
Detection zone, only device to hole groove edge part carries out burr feature extraction, substantially increases the detection precision of system burr, Neng Gouyou
Effect avoids the scratch of hole head position, abrasion and the interference of noise.
It is with strong points, higher robustness is all had to the various types burr occurred in motor copper bar process: this
The size and morphological feature for the four seed type burrs that scheme is generated for motor copper bar hole slot position, by a large amount of defect images
It carries out the analysis of section object attribute synthesis and comparison production classifier defines threshold value, realize burr detection.
Detailed description of the invention
Fig. 1 is a kind of defect detecting device figure based on motor copper bar burr growth district of the present invention.
Fig. 2 is a kind of defects detection algorithm flow chart based on motor copper bar burr growth district of the present invention.
Fig. 3 is the copper bar standard picture to be detected of the method for the present invention processing.
Fig. 4 is the copper bar Image Edge-Detection administrative division map of the method for the present invention construction.
Fig. 5 is four kinds of veining defect type maps of research object of embodiment of the present invention copper bar workpiece.
In figure: 1- electronic control translation stage, 2-LED light source, 3-CCD industrial camera, 4- image pick-up card, 5- computer, 6-
Copper bar workpiece, 7- precision stage, 8- articles holding table, 9- position sensor, 10- hood, 11- manually control instrument.
Specific embodiment
The present invention is further explained in the light of specific embodiments:
Such as Fig. 2, defects detection algorithm of this embodiment scheme based on motor copper bar burr growth district the following steps are included:
Step 1 obtains copper bar standard picture to be detected.
Step 2 constructs copper bar Image Edge-Detection region.
Step 3 feature extraction and veining defect determine.
The step 1 obtains copper bar standard picture to be detected, as shown in figure 3, detailed process is as follows:
Referring to Fig. 1, system led light source 2 is using annular low angle diffusion illumination, when copper bar workpiece 6 is in precision stage 7
Position to be detected is moved on articles holding table 8, position sensor 9 responds and triggers the work of CCD industrial camera 3, acquires copper bar figure
As and the image processing module that is transmitted in computer 5 it is pre-processed.
Using improved median filtering mode, first chooses 8 kinds of representative edges and judge whether pixel is edge, be side
Edge is directly entered next step, otherwise, carries out following 3 × 3 wicket median filter process:
In formula, SxyIt indicates centered on point (x, y), the window subgraph set of coordinates that size is 3 × 3, g (s, t) represents neck
The grey scale pixel value of domain coordinate.Indicate gray value of this after median filter process.
Largest connected region and its parameter tn are found out using 8 mode of communicating tag image regions and by threshold value of area, is constructed
Size and the consistent completely black mask of original image obtain centre using tn and are left white region (i.e. gray value setting in label rectangular area
For 1).Gained mask and copper bar image are subjected to exposure mask operation.
Ibid, the complete white mask of one width of construction, and pixel point summation operation is carried out with above-mentioned acquired results, it obtains to be checked
Survey copper bar standard picture I_mask_white.
The step 2 constructs copper bar Image Edge-Detection region, as shown in figure 4, detailed process is as follows:
Gaussian filter is created first, and gaussian filtering is carried out to I_mask_white image:
(x, y) is point coordinate in formula, and δ is standard deviation；One 3 × 3 Gaussian filter template is generated, with the center of template
Position is sampled for coordinate origin, the coordinate of each position is brought into Gaussian function, calculation formula is as follows:
H in formulai,jFor the element value of position coordinates, i, j are position coordinates, and δ is standard deviation.
Then Threshold segmentation is carried out using OTSU method, as follows:
F (x, y) is the workpiece image being partitioned into formula, and t is optimal segmenting threshold.
Closing operation of mathematical morphology processing completion notch is carried out to the image after Threshold segmentation and calculates maximum area connected region
Angle, θ be directly entered next step if θ=0, otherwise rotate image and corrected: calculating angle, θ using regionprops
In Orientation function；Rotating angle is (90- | θ |).
Morphological scale-space obtains template image X, as follows:
It is accorded in formula for closed operation,It is accorded with for opening operation,For dilation operation symbol, Θ is erosion operation symbol, and S is that diameter is
8, element is all 1 collar plate shape structural element.
Corrode the collar plate shape structural element S that obtained template image X diameter is 8 to obtain IM image:
X Θ S=∩ { X-s/s ∈ S } (7)
The collar plate shape structural element corrosion IM that a larger diameter (35) is arranged obtains IM_erode image, and is cut with IM
IM_erode carries out image operation, obtains copper bar Image Edge-Detection region.
Step 3 feature extraction and veining defect determine that detailed process is as follows:
By after rotational correction copper bar standard picture I_mask_white to be detected and copper bar Image Edge-Detection region into
Row pixel point is multiplied, and extracts burr feature.
Motor copper bar hole slot burrs on edges is as shown in figure 5, there are four types of type: I class burr, II class burr, Group III burr and
IV class burr；Each region of gained defect image is marked with 8 mode of communicating and utilizes the fineness ratio based on area and perimeter
Thinness ratio defines threshold value production classifier and determines I class and II class burr:
K=(4 × π × S)/C2 (8)
K indicates that fineness scale parameter, S indicate that region area, C indicate area circumference in formula；
In formulaIndicate defect image in each region set of pixels and, m be a constant,Indicate veining defect type I
With the detection of Type II.
Using with region there is the elliptical long axis length Axis Length of identical standard second-order moment around mean to define threshold value system
Make classifier to determine Group III burr:
AL indicates the elliptical long axis length for having identical standard second-order moment around mean with region in formula, and p is a constant,Table
Show the detection of veining defect type-iii.
Setting up a reference quantity EX indicates degree of expansion of the region in minimum boundary rectangle, and defines threshold value production classification
Device determines IV class burr:
EX=F/S1 (11)
F indicates area pixel number, S in formula1Indicate the minimum circumscribed rectangle area of enclosing region；
In formulaIndicate the detection of veining defect type IV, EX indicates region degree of expansion, and r is a constant.
Finally the four seed type burr areas minimum circumscribed rectangle frame of classifier identification is selected, and original after correction
Its coordinate information is shown on image, is stored in system, is completed the detection of veining defect.
The present embodiment does not need to be accurately positioned copper bar workpiece, even if appropriate rotation offset occurs for image, still can accurately detect
Various veining defects in copper bar hole slot.Algorithmic stability is efficient simultaneously, is compared to background subtraction, substantially increases system
Serious forgiveness and detection accuracy.By production scene actual motion, for recall rate close to 98%, it is 0% that rate is picked up in leakage, can be effectively applicable to
The various models of industry spot, the copper bar burr detection of various specifications.
Embodiment described above is only schematical, and present invention is described, and implementation model of the invention is not limited with this
It encloses.So if those skilled in the art are inspired by it, without departing from the spirit of the invention, without
Creative designs structure similar with the technical solution and embodiment, is within the scope of protection of the invention.
Claims (4)
1. a kind of defect detecting device and method based on motor copper bar burr growth district, which is characterized in that including CCD industry
Video camera (3), image pick-up card (4), LED light source (2), electronic control translation stage (1), precision stage (7), articles holding table (8), shading
Cover (10) manually controls instrument (11), position sensor (9) and computer (5)；The position sensor (9) is set to workpiece
Position to be detected, by generating pulse signal triggering CCD industrial camera (3) work；The LED light source (2) is using annular
The low angle of structure diffuses illumination, is set to below CCD industrial camera (3)；Essence is provided with below CCD industrial camera (3)
Copper bar workpiece (6) with hole slot is placed on the articles holding table (8) of precision stage (7) and by automatically controlled by close workbench (7)
Translation stage (1) controls speed and the track of articles holding table (8)；Image pick-up card (4) and CCD industrial camera (3) pass through buttock line phase
Lian Bingyu computer (5) communication；CCD industrial camera (3) is disposed vertically right above workpiece position to be detected, and CCD industry
Video camera (3) is rung using hood (10) shielding environment shadow, and camera lens visual field is covered in the articles holding table of precision stage (7)
(8) center can once shoot 2 copper bar workpiece (6) hole slots；The computer (5) adopts image pick-up card (4)
The copper bar image collected carry out analysis and algorithm operation, and by classifier by the various types burr in motor copper bar hole slot into
Row picks:
The following steps are included:
Step 1 obtains copper bar standard picture to be detected；
Step 2 constructs copper bar Image Edge-Detection region；
Step 3 feature extraction and veining defect determine.
2. the defect detecting device and method according to claim 1 based on motor copper bar burr growth district, feature
It is, in step 1, system led light source (2) is using annular low angle diffusion illumination, when copper bar workpiece (6) is in precision stage
(7) position to be detected is moved on articles holding table (8), position sensor (9) responds and triggers CCD industrial camera (3) work,
It acquires image processing module of the copper bar image transmitting into computer (5) to pre-process it, and is filtered using improved intermediate value
Wave mode is filtered with 3 × 3 wicket:
Threshold value T is set up, crosses to obtain 4 adjacent pixels, the difference of the grey scale pixel value and adjacent pixel gray value with center
For D, as D < T, G as a < G < b, then can determine that the pixel is marginal point from adding 1；Wherein a, b are constant, and G indicates phase
Adjacent pixel number similar with center pixel gray scale；If a < G < b, the i.e. point are marginal point, without any transformation, directly
This grey scale pixel value f (x, y) is exported, it will after otherwise being filtered by formula 1Output:
In formula, SxyIt indicates centered on point (x, y), the window subgraph set of coordinates that size is 3 × 3, g (s, t) represents field seat
Target grey scale pixel value,Indicate gray value of this after median filter process；
Then with 8 mode of communicating tag image regions, largest connected region and its parameter tn are found out by threshold value of area, construction is big
It is small with the consistent completely black mask of original image, obtain centre using tn and be left white region and gained mask and copper bar image are subjected to exposure mask
Operation, then be same as above, construction and the consistent complete white mask of original image obtain centre and stay black region, this gained mask and exposure mask are transported
The result of calculation carries out pixel point summation operation, obtains copper bar standard picture I_mask_white to be detected.
3. the defect detecting device and method according to claim 1 based on motor copper bar burr growth district, feature
It is, in step 2:
Gaussian filter is created first, and gaussian filtering is carried out to I_mask_white image:
(x, y) is point coordinate in formula, and δ is standard deviation；One 3 × 3 Gaussian filter template is generated, with the center of template
It is sampled, the coordinate of each position is brought into Gaussian function, calculation formula is as follows for coordinate origin:
H in formulai,jFor the element value of position coordinates, i, j are position coordinates, and δ is standard deviation；
Then Threshold segmentation is carried out using OTSU method, as follows:
F (x, y) is the workpiece image being partitioned into formula, and t is optimal segmenting threshold；
Closing operation of mathematical morphology processing completion notch is carried out to the image after Threshold segmentation and calculates the angle of maximum area connected region
It spends θ and is directly entered next step if θ=0, otherwise rotate image and corrected: calculating angle, θ using in regionprops
Orientation function；Rotating angle is (90- | θ |)；
Morphological scale-space obtains template image X, as follows:
It is accorded in formula for closed operation,It is accorded with for opening operation,For dilation operation symbol, Θ is erosion operation symbol, and S is that diameter is 8, element
It is all 1 collar plate shape structural element；
Corrode the collar plate shape structural element S that obtained template image X diameter is 8 to obtain IM image:
X Θ S=∩ { X-s/s ∈ S } (7)
The larger-diameter collar plate shape structural element corrosion IM of setting one obtains IM_erode image, and cuts IM_ with IM
Erode carries out image operation, obtains copper bar Image Edge-Detection region.
4. defect detecting device and method based on motor copper bar burr growth district according to claim 2 and 3, special
Sign is, in step 3, by the copper bar standard picture I_mask_white to be detected and copper bar Image Edge-Detection after rotational correction
Region carry out the multiplication of pixel point, extract burr feature and use each region of 8 mode of communicating tag images, production classifier,
Threshold value is defined, algorithm judgement is carried out to various types burr:
(1) threshold value production classifier is defined to I class and II using the fineness ratio thinness ratio based on area and perimeter
Class burr is determined:
K=(4 × π × S)/C2 (8)
K indicates that fineness scale parameter, S indicate that region area, C indicate area circumference in formula,
In formulaIndicate defect image in each region set of pixels and, m be a constant,Indicate veining defect type I and class
The detection of type II；
(2) using with region there is the elliptical long axis length Axis Length of identical standard second-order moment around mean to define threshold value system
Make classifier to determine Group III burr:
AL indicates the elliptical long axis length for having identical standard second-order moment around mean with region in formula, and p is a constant,Indicate hair
Pierce the detection of defect type III；
(3) setting up a reference quantity EX indicates degree of expansion of the region in minimum boundary rectangle, and defines threshold value production classification
Device determines IV class burr:
EX=F/S1 (11)
F indicates area pixel number, S in formula1Indicate the minimum circumscribed rectangle area of enclosing region,
In formulaIndicate the detection of veining defect type IV, EX indicates region degree of expansion, and r is a constant；
Finally the four seed type burr areas minimum circumscribed rectangle frame of classifier identification is selected, and the original image after correction
Upper its coordinate information of display is stored in system, completes the detection of veining defect.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910014867.5A CN109447989A (en) | 2019-01-08 | 2019-01-08 | Defect detecting device and method based on motor copper bar burr growth district |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910014867.5A CN109447989A (en) | 2019-01-08 | 2019-01-08 | Defect detecting device and method based on motor copper bar burr growth district |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109447989A true CN109447989A (en) | 2019-03-08 |
Family
ID=65540044
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910014867.5A Pending CN109447989A (en) | 2019-01-08 | 2019-01-08 | Defect detecting device and method based on motor copper bar burr growth district |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109447989A (en) |
Cited By (13)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110763700A (en) * | 2019-10-22 | 2020-02-07 | 深选智能科技(南京)有限公司 | Method and equipment for detecting defects of semiconductor component |
| CN111062959A (en) * | 2019-11-28 | 2020-04-24 | 重庆大学 | Extraction and characterization method for bottom edge burr cutting characteristics of aviation thin-wall micro-structural part |
| CN111189843A (en) * | 2020-01-10 | 2020-05-22 | 江苏森蓝智能系统有限公司 | Online detection method for embossing quality of copper bar |
| CN111815600A (en) * | 2020-07-04 | 2020-10-23 | 博科视(苏州)技术有限公司 | Visual sense-based annular magnetic steel appearance defect detection method |
| CN111833350A (en) * | 2020-08-26 | 2020-10-27 | 南京原觉信息科技有限公司 | Machine vision detection method and system |
| CN112432954A (en) * | 2020-12-09 | 2021-03-02 | 浙江理工大学 | Braided tube flaw detection method |
| CN112819745A (en) * | 2019-10-31 | 2021-05-18 | 合肥美亚光电技术股份有限公司 | Nut kernel center worm-eating defect detection method and device |
| CN112862694A (en) * | 2019-11-12 | 2021-05-28 | 合肥欣奕华智能机器有限公司 | Screen position correction method and device, computing equipment and storage medium |
| CN113379723A (en) * | 2021-06-29 | 2021-09-10 | 上海闻泰信息技术有限公司 | Irregular glue overflow port detection method, device, equipment and storage medium |
| CN114279357A (en) * | 2021-12-23 | 2022-04-05 | 杭州电子科技大学 | Die casting burr size measurement method and system based on machine vision |
| CN114813760A (en) * | 2022-06-27 | 2022-07-29 | 荣旗工业科技(苏州)股份有限公司 | Surface defect detection device and surface defect detection method |
| CN116587043A (en) * | 2023-07-18 | 2023-08-15 | 太仓德纳森机电工程有限公司 | Workpiece conveying system for industrial automatic production and processing |
| CN114279357B (en) * | 2021-12-23 | 2024-05-03 | 杭州电子科技大学 | Die casting burr size measurement method and system based on machine vision |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006171816A (en) * | 2004-12-10 | 2006-06-29 | Sony Corp | Image processor |
| US20140168418A1 (en) * | 2012-12-13 | 2014-06-19 | Kla-Tencor Corporation | Delta die intensity map measurement |
| WO2015120734A1 (en) * | 2014-02-17 | 2015-08-20 | 华南理工大学 | Special testing device and method for correcting welding track based on machine vision |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN106442556A (en) * | 2016-11-16 | 2017-02-22 | 哈尔滨理工大学 | Device and method for detecting surface defects of perforated plate workpiece |
-
2019
- 2019-01-08 CN CN201910014867.5A patent/CN109447989A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006171816A (en) * | 2004-12-10 | 2006-06-29 | Sony Corp | Image processor |
| US20140168418A1 (en) * | 2012-12-13 | 2014-06-19 | Kla-Tencor Corporation | Delta die intensity map measurement |
| WO2015120734A1 (en) * | 2014-02-17 | 2015-08-20 | 华南理工大学 | Special testing device and method for correcting welding track based on machine vision |
| CN106053479A (en) * | 2016-07-21 | 2016-10-26 | 湘潭大学 | System for visually detecting workpiece appearance defects based on image processing |
| CN106442556A (en) * | 2016-11-16 | 2017-02-22 | 哈尔滨理工大学 | Device and method for detecting surface defects of perforated plate workpiece |
Non-Patent Citations (3)
| Title |
|---|
| XIANGNING LU等: "A novel approach for flip chip solder joint inspection based on pulsed phase thermography", 《NDT & E INTERNATIONAL》 * |
| 杨亚: "嵌入式小模数齿轮缺陷视觉检测系统的研究", 《CNKI优秀硕士毕业论文全文库（工程科技I辑）》 * |
| 郭斌等: "柱塞主缸密封凹槽表面缺陷检测系统", 《中国测试》 * |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110763700A (en) * | 2019-10-22 | 2020-02-07 | 深选智能科技(南京)有限公司 | Method and equipment for detecting defects of semiconductor component |
| CN112819745A (en) * | 2019-10-31 | 2021-05-18 | 合肥美亚光电技术股份有限公司 | Nut kernel center worm-eating defect detection method and device |
| CN112862694A (en) * | 2019-11-12 | 2021-05-28 | 合肥欣奕华智能机器有限公司 | Screen position correction method and device, computing equipment and storage medium |
| CN111062959A (en) * | 2019-11-28 | 2020-04-24 | 重庆大学 | Extraction and characterization method for bottom edge burr cutting characteristics of aviation thin-wall micro-structural part |
| CN111189843A (en) * | 2020-01-10 | 2020-05-22 | 江苏森蓝智能系统有限公司 | Online detection method for embossing quality of copper bar |
| CN111815600A (en) * | 2020-07-04 | 2020-10-23 | 博科视(苏州)技术有限公司 | Visual sense-based annular magnetic steel appearance defect detection method |
| CN111815600B (en) * | 2020-07-04 | 2024-04-02 | 博科视(苏州)技术有限公司 | Visual-based annular magnetic steel appearance defect detection method |
| CN111833350B (en) * | 2020-08-26 | 2023-06-06 | 南京原觉信息科技有限公司 | Machine vision detection method and system |
| CN111833350A (en) * | 2020-08-26 | 2020-10-27 | 南京原觉信息科技有限公司 | Machine vision detection method and system |
| CN112432954A (en) * | 2020-12-09 | 2021-03-02 | 浙江理工大学 | Braided tube flaw detection method |
| CN113379723A (en) * | 2021-06-29 | 2021-09-10 | 上海闻泰信息技术有限公司 | Irregular glue overflow port detection method, device, equipment and storage medium |
| CN113379723B (en) * | 2021-06-29 | 2023-07-28 | 上海闻泰信息技术有限公司 | Irregular glue overflow port detection method, device, equipment and storage medium |
| CN114279357A (en) * | 2021-12-23 | 2022-04-05 | 杭州电子科技大学 | Die casting burr size measurement method and system based on machine vision |
| CN114279357B (en) * | 2021-12-23 | 2024-05-03 | 杭州电子科技大学 | Die casting burr size measurement method and system based on machine vision |
| CN114813760A (en) * | 2022-06-27 | 2022-07-29 | 荣旗工业科技(苏州)股份有限公司 | Surface defect detection device and surface defect detection method |
| CN116587043A (en) * | 2023-07-18 | 2023-08-15 | 太仓德纳森机电工程有限公司 | Workpiece conveying system for industrial automatic production and processing |
| CN116587043B (en) * | 2023-07-18 | 2023-09-15 | 太仓德纳森机电工程有限公司 | Workpiece conveying system for industrial automatic production and processing |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109447989A (en) | Defect detecting device and method based on motor copper bar burr growth district | |
| CN113724216B (en) | Method and system for detecting wave crest welding spot defects | |
| CN102915433B (en) | Character combination-based license plate positioning and identifying method | |
| CN105894036A (en) | Image feature template matching method being applied to detection of mobile phone screen defects | |
| CN108982512A (en) | A kind of circuit board detecting system and method based on machine vision | |
| CN108007388A (en) | A kind of turntable angle high precision online measuring method based on machine vision | |
| CN109409355A (en) | A kind of method and device of novel transformer nameplate identification | |
| CN113781486B (en) | IC card image defect detection method | |
| CN107067595B (en) | State identification method and device of indicator light and electronic equipment | |
| CN111598856A (en) | Chip surface defect automatic detection method and system based on defect-oriented multi-point positioning neural network | |
| CN109752392A (en) | A kind of pcb board defect type detection system and method | |
| CN113705564B (en) | Pointer type instrument identification reading method | |
| CN113435452A (en) | Electrical equipment nameplate text detection method based on improved CTPN algorithm | |
| CN111832388A (en) | Method and system for detecting and identifying traffic sign in vehicle running | |
| CN111652844A (en) | X-ray defect detection method and system based on digital image region growth | |
| CN111047646A (en) | Multi-target lens positioning method and system based on FPGA | |
| CN107367514A (en) | A kind of POS defect inspection method based on machine vision | |
| CN110610474A (en) | Solar panel defect real-time detection method based on infrared image | |
| CN115082504B (en) | Light spot identification method for solar photovoltaic panel | |
| CN110223339A (en) | One kind being based on machine vision thermal protector calibration point center positioning method | |
| Zhang et al. | Defect detection of mobile phone screen based on improved difference image method | |
| CN113689365B (en) | Target tracking and positioning method based on Azure Kinect | |
| CN115689990A (en) | Carpet lamp stray light defect intelligent detection algorithm and equipment based on machine vision | |
| CN115019306A (en) | Embedding box label batch identification method and system based on deep learning and machine vision | |
| Huang et al. | An automatic detection and recognition method for pointer-type meters in natural gas stations |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20190308 |