CN110084787B - Synthetic leather gluing online monitoring method based on machine vision - Google Patents
Synthetic leather gluing online monitoring method based on machine vision Download PDFInfo
- Publication number
- CN110084787B CN110084787B CN201910295581.9A CN201910295581A CN110084787B CN 110084787 B CN110084787 B CN 110084787B CN 201910295581 A CN201910295581 A CN 201910295581A CN 110084787 B CN110084787 B CN 110084787B
- Authority
- CN
- China
- Prior art keywords
- image
- synthetic leather
- pixels
- contour
- leather
- 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.)
- Active
Links
- 239000002649 leather substitute Substances 0.000 title claims abstract description 71
- 238000000034 method Methods 0.000 title claims abstract description 34
- 238000004026 adhesive bonding Methods 0.000 title claims abstract description 25
- 238000012544 monitoring process Methods 0.000 title claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 11
- 238000012216 screening Methods 0.000 claims abstract description 9
- 230000007547 defect Effects 0.000 claims description 17
- 238000007689 inspection Methods 0.000 claims description 12
- 238000001514 detection method Methods 0.000 claims description 11
- 239000004744 fabric Substances 0.000 claims description 11
- 238000002372 labelling Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 5
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000009957 hemming Methods 0.000 claims description 4
- 238000004891 communication Methods 0.000 claims description 2
- 239000010985 leather Substances 0.000 abstract description 10
- 239000003292 glue Substances 0.000 abstract description 3
- 230000002159 abnormal effect Effects 0.000 abstract description 2
- 239000000463 material Substances 0.000 abstract description 2
- 239000000203 mixture Substances 0.000 abstract description 2
- 239000004745 nonwoven fabric Substances 0.000 abstract description 2
- 230000035699 permeability Effects 0.000 abstract description 2
- 239000004033 plastic Substances 0.000 abstract description 2
- 229920003023 plastic Polymers 0.000 abstract description 2
- 229920002635 polyurethane Polymers 0.000 abstract description 2
- 239000004814 polyurethane Substances 0.000 abstract description 2
- 239000010410 layer Substances 0.000 abstract 3
- 238000004519 manufacturing process Methods 0.000 abstract 1
- 239000002344 surface layer Substances 0.000 abstract 1
- 239000011248 coating agent Substances 0.000 description 3
- 238000000576 coating method Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000005315 distribution function Methods 0.000 description 2
- 238000009434 installation Methods 0.000 description 2
- 239000000853 adhesive Substances 0.000 description 1
- 230000001070 adhesive effect Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 238000005286 illumination Methods 0.000 description 1
- 230000004630 mental health Effects 0.000 description 1
- 239000002245 particle Substances 0.000 description 1
- 239000007787 solid Substances 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 231100000331 toxic Toxicity 0.000 description 1
- 230000002588 toxic effect Effects 0.000 description 1
Images
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
- G06T7/001—Industrial image inspection using an image reference approach
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/77—Retouching; Inpainting; Scratch removal
-
- 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
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/30—Computing systems specially adapted for manufacturing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Quality & Reliability (AREA)
- Treatment Of Fiber Materials (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
A synthetic leather gluing online monitoring method based on machine vision comprises the following steps: 1. collecting synthetic leather images by using a line scanning camera; 2. calculating the height h of the synthetic leather Image; 3. screening the Image through a set threshold value t, and obtaining a region R with the threshold value larger than t; 4. filling the region R to become a new region R1Then R is1Then is a synthetic leather image area; synthetic leather is a plastic product which simulates the composition and structure of natural leather and can be used as a substitute material of the natural leather, and is usually prepared by taking impregnated non-woven fabric as a net layer and a microporous polyurethane layer as a grain surface layer. The front and back surfaces of the leather are very similar to leather, have certain air permeability and are closer to natural leather than common artificial leather. The method is widely used for manufacturing shoes, boots, bags, balls and the like, and comprises a gluing process in the processing process of the synthetic leather, wherein a layer of glue solution is uniformly coated on the front side of the synthetic leather through gluing equipment, and the gluing process can cause some abnormity, so that the gluing is abnormal or corresponding flaws are generated.
Description
Technical Field
The invention relates to the technical field of quality monitoring, in particular to a synthetic leather gluing online monitoring method based on machine vision.
Background
Synthetic leather is a plastic product that mimics the composition and structure of natural leather and can be used as a substitute material for natural leather. Usually, the impregnated non-woven fabric is used as a net layer, and the microporous polyurethane layer is used as a grain layer. The front and back surfaces of the leather are very similar to leather, have certain air permeability and are closer to natural leather than common artificial leather. It is widely used for making shoes, boots, bags, balls, etc.
The processing process of the synthetic leather comprises a gluing process, wherein a layer of glue solution is uniformly coated on the front side of the synthetic leather through gluing equipment. The gluing process may have the following anomalies, which may lead to abnormal gluing or corresponding defects:
1. synthetic leather is turned up, so that the edges are not glued;
2. solid particles are arranged on the roll shaft, so that the synthetic leather after being glued has long scratches;
3. uneven gluing or coating omission in partial areas and the like caused by hardware;
these anomalies all need to be dealt with in time, otherwise the quality of the whole roll of synthetic leather may be affected, so the gluing process needs to be monitored in real time.
Disclosure of Invention
The invention aims to provide a synthetic leather gluing online monitoring method based on machine vision so as to solve the problems in the background technology.
In order to realize the purpose, the synthetic leather gluing online monitoring method based on machine vision comprises the following steps:
1) collecting: collecting a synthetic leather Image by using a line scanning camera;
2) initial inspection: carrying out primary inspection processing on the synthetic leather Image;
3) detecting turned edges: comparing the obtained data with the image parameters for the first time, and detecting whether the synthetic leather is curled or not;
4) if the curling is detected, the system quits the algorithm detection of the graph;
5) and (4) supplementary inspection: if no curling is detected, the computer continues to perform the supplementary inspection on the synthetic leather area through an edge angle algorithm;
6) and (3) confirming defects: and comparing the parameters obtained by the supplementary inspection with the set defect screening parameters for the second time, and detecting the defects of the synthetic leather.
Preferably, the 2) preliminary examination processing includes the steps of:
2.1) converting the Image into a gray level Image GrayImage;
2.2) calculating the height h of the synthetic leather image GrayImage;
2.3) carrying out binarization on GrayImage through a set threshold value t to obtain a binary image BinImage1；
2.4) BinImage by connected region labeling Algorithm1Marking is carried out, each individual connected region forms an identified block, and all connected regions R in the image are obtainedS；
2.5) traversing all connected regions RSCounting the number of pixels of each block, screening the area with the largest number of pixels, marking as R, filling R by using a flooding filling method to obtain a synthetic leather image area R1；
2.6) extracting the region R1The outer contour C of (a);
2.7) according to the wheelContour algorithm divides C and extracts left contour CLAnd the right contour CR。
Preferably, step 2.4) specifically comprises the following steps:
2.4.1) traversal of the image BinImage1Building an identification number for each pixel until the gray value of the pixel point is 255, and marking the area;
2.4.2) traversing 8 adjacent pixels of the pixel, and putting the pixel with the gray value of 255 into the identification area;
2.4.3) traversing the pixels in the identification number, and executing the step 2.4.2) until the gray values of 8 adjacent pixels are all 0;
2.4.4) continue traversing the BinImage1And (4) building an identification number for each pixel until the gray value of the pixel point is 255 and the pixel point is not in the identified area, and marking the area.
2.4.5) performing step 2.4.2), 2.4.3);
2.4.6) repeating the steps 2.4.4) and 2.4.5) until the traversal is completed and all the identification blocks are obtained, and then all the connected regions R in the image can be obtaineds；
Preferably, the region R is extracted1The specific operation of the outer contour C of (a) is: traversing each pixel of the image, and determining eight adjacent pixel positions of the pixel, if any, of the eight adjacent pixels, at least one of which is adjacent to R1Simultaneously at least 1 pixel in R1And putting the pixels into a stack, wherein the combination of the pixels in the stack is the outline C.
Preferably, the contour algorithm is: it should be noted by those skilled in the art that the outer contour C is composed of four sub-contours, i.e., an upper, a lower, a left, and a right sub-contour, and since the image is in a continuous state, the upper contour is a straight line parallel to the x coordinate and the y coordinate of 0, the lower contour is a straight line parallel to the x coordinate and the y coordinate of h (image height), and the pixels on the outer contour C are traversed to obtain the minimum value x of the x coordinate among all the pixels1Maximum value x2Traversing pixels on the outer contour C, if the y coordinate is larger than 0 and smaller than h, judging the x coordinate: greater than x1-1, and less than x1+(x2-x1) The/2 pixels are placed in the stack L, larger than x1+(x2-x1) A/2 and less than x2+1 pixels are placed in the stack R, the set of pixels in the stack L being the left contour CLThe set of pixels in the stack R is the right contour CR。
Preferably, the first alignment comprises: respectively count CLAnd CRNumber of pixels lLAnd lRAnd compared with the height h of the image, e.g. lLAnd lRRespectively more than 1.5 × h, the edge of the synthetic leather is normal, otherwise the synthetic leather is curled (the edge of the normal cloth is zigzag, so the contour length of the cloth is more than 1.5 times of the fitting length, but after the curling, the edge of the cloth is almost a straight line, and the contour length of the cloth is less than 1.5 times of the fitting length).
Preferably, the edge angle algorithm is specifically as follows: c is to beLAnd CRRespectively fitted into straight lines LLAnd LRAnd calculate their angleL,angleRE.g. angleL-an absolute value of 90 ° greater than 15 °, or angleRAn absolute value of-90 ° greater than 15 °, curling occurs (the edge of the curl facing non-perpendicular to the direction of advance of the web).
Preferably, the supplementary inspection comprises the following steps:
5.1) carrying out Fourier transform on the GrayImage to obtain a frequency domain image ImageF of the GrayImage;
5.2) setting the first Gaussian Filter Filter1And filtering the frequency domain image ImageF to obtain a filtered image ImageF1；
5.3) setting the second Gaussian Filter Filter2And filtering the frequency domain image ImageF to obtain a filtered image ImageF2；
5.4) image ImageF1、ImageF2Conversion into a spatial domain Image by inverse Fourier transform1、Image2；
5.5) Image1、Image2Subtracting to obtain Image3；
5.6) By a set threshold value t1For Image3Processing to obtain a binary image BinImage2；
5.7) BinImage with connected region labeling Algorithm2Marking to obtain each communication region RS1；
5.8) traverse RS1And respectively calculating the height h of the minimum circumscribed rectangletAnd width wt。
Preferably, the second alignment comprises: will be of height htAnd width wtComparing with a set defect screening height threshold h and a width threshold w, such as htGreater than h or wtIf the number of the areas in the stacks is larger than 0, the system is judged to detect the flaws.
Preferably, a roller shaft is arranged at the lower position of the synthetic leather, an encoder is assembled on the roller shaft, a linear scanning camera is fixedly installed at the upper position of the synthetic leather, a linear scanning lens with a proper focal length is installed on the linear scanning camera, a linear scanning DOME light source and a linear scanning coaxial light source line are arranged at the upper position of the synthetic leather, and the center of a slit at the top end of the linear scanning coaxial light source, the center of a slit of the linear scanning DOME light source and the center of the linear scanning camera are coplanar during installation, so that the camera can collect images of the synthetic leather.
In conclusion, the beneficial effects of the invention are as follows:
1. the method adopts a machine vision system to replace manual work to complete field monitoring, and reduces the influence of some toxic glue solution on the physical and mental health of monitoring workers.
2. Compared with manual detection, the method has the characteristics of high accuracy, high efficiency and good stability.
3. The method consists of line scanning DOME and line scanning coaxial light, not only can highlight the flaws of the synthetic leather, but also can avoid the influence of the reflection of the synthetic leather after gluing on the image quality.
4. The method adopts a novel hemming algorithm, so that the hemming detection is simpler and more convenient.
5. The method adopts the combination of the Gaussian band-pass filter and the Fourier transform, can convert the image into a frequency domain for processing, and enables the defect detection to be more accurate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the invention, and it is also possible for a person skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a circuit block diagram of a synthetic leather gluing online monitoring method based on machine vision;
FIG. 2 is a schematic view of the installation orientation of the apparatus of the present invention;
FIG. 3 is a right side view of FIG. 2;
FIG. 4 is an enlarged schematic view of a line-scanning coaxial light source;
fig. 5 is an enlarged schematic diagram of a line scan DOME light source.
Detailed Description
Embodiments of the invention are described in further detail below with reference to the accompanying drawings, it being noted that the examples are merely illustrative of the invention and should not be considered as limiting, and that all features disclosed in the examples of the invention, or steps in all methods or processes disclosed, may be combined in any way, except for mutually exclusive features and/or steps.
The embodiment provided by the invention comprises the following steps: a synthetic leather gluing online monitoring method based on machine vision comprises the following steps:
1. collecting the synthetic leather Image by using a line scanning camera,
2. the Image is converted into a grayscale Image gray Image,
3. calculating the height h of the synthetic leather image GrayImage,
4. binarizing the GrayImage with the set threshold value t being 10, setting the pixel gray value with the gray scale larger than t as 255 and setting the gray scale of the pixel with the gray scale not larger than t as 0, obtaining a binary image BinImage,
5. by usingConnected region labeling algorithm pair BinImage1Marking pixels with the gray value of 255 to enable each individual connected region to form an identified block, specifically operating as follows:
5.1) traversing the BinImage of the image1Building an identification number for each pixel until the gray value of the pixel point is 255, and marking the area;
5.2) traversing 8 adjacent pixels of the pixel, and putting the pixel with the gray value of 255 into the identification area;
5.3) traversing the pixels in the identification number, and executing the step 5.2) until the gray values of 8 adjacent pixels are 0;
5.4) continuously traversing each pixel of the BinImage1 until the gray value of the pixel point is 255 and the pixel point is not in the identified area, and creating an identification number to label the area;
5.5) performing step 5.2), 5.3);
5.6) repeating the steps 5.4) and 5.5) until the traversal is completed, and obtaining all the identification blocks, all the connected regions R in the image can be obtaineds；
6. Traverse all connected regions RsCounting the number of pixels of each block, screening the area with the largest number of pixels, marking as R, filling R by a flooding filling method to obtain R1，R1Namely the synthetic leather image area,
7. extracting region R1The outer contour C of (a) operates as follows: and traversing each pixel of the image, and judging eight adjacent pixel positions of the pixel. E.g. having at least one of its eight adjacent pixels in R1Simultaneously at least 1 pixel in R1And putting the pixel into a stack, wherein the combination of the pixels in the stack is the outer contour C,
8. the outer contour C is composed of four sub-contours, namely an upper sub-contour, a lower sub-contour, a left sub-contour and a right sub-contour, because the image is in a continuous state, the upper contour is a straight line which is parallel to the x coordinate and the y coordinate and is 0, the lower contour is a straight line which is parallel to the x coordinate and the y coordinate and is h (image height),
9. traversing the pixels on the outer contour C to obtain the minimum value x of the x coordinate in all the pixels1Maximum value x2，
10. Traversing pixels on the outer contour C, if the y coordinate is larger than 0 and smaller than h, judging the x coordinate: greater than x1-1, and less than x1+(x2-x1) The/2 pixels are placed in the stack L, larger than x1+(x2-x1) A/2 and less than x2+1 pixels are placed in the stack R, the set of pixels in the stack L being the left contour CLThe set of pixels in the stack R is the right contour CRSeparately counting CLAnd CRThe number of pixels (c) is respectively marked as lLAnd lR. Will lLAnd lRCompared with the height h of the image, e.g. lLAnd lRRespectively more than 1.5 x h, the edge of the synthetic leather is normal, otherwise, the synthetic leather is curled, (the edge of the normal cloth is jagged, so the contour length of the cloth is more than 1.5 times of the fitting length, but after the curling, the edge of the cloth is almost a straight line, and the contour length of the cloth is less than 1.5 times of the fitting length)
11. Most of the hemming defects can be judged through the above, but partial missing detection still exists, and the detection needs to be supplemented, wherein the judgment is carried out through the angle of the edge, and the specific algorithm is as follows: c is to beLAnd CRRespectively fitted into straight lines LLAnd LRAnd calculate their angleL,angleRE.g. angleL-an absolute value of 90 ° greater than 15 °, or angleRAn absolute value of-90 ° greater than 15 °, curling occurs (edge after curling is not perpendicular to the direction of advance of the cloth)
12. If the curling edge is detected, the system exits the algorithm detection of the figure,
13. if no curling is detected, the system continues to detect flaws in the synthetic leather area, and in the image, the edge area with violent conversion is an area with violent gray level change in the image, and the corresponding frequency value is higher. In the synthetic leather image, the flaw point edge is always a high-frequency signal, whether the flaw point edge is the flaw point edge can be judged through the pixel frequency value,
14. firstly, Fourier transform is carried out on GrayImage to obtain a frequency domain image ImageF of the GrayImage, (the description is that if f is an energy-limited analog signal, the Fourier transform represents the frequency spectrum of f, and the Fourier transform transforms the gray distribution function of the image into the frequency distribution function of the image)
15. Next, set the 1 st Gaussian Filter Filter1The kernel size is 3 x 3, the sigma value is 10, and the image is filtered to obtain a filtered image ImageF1The specific algorithm is as follows:
15.1) two-dimensional Gaussian function of
Where (x, y) is the point coordinate and σ is the standard deviation
15.2) discretizing the Gaussian function to obtain a Gaussian function value as the coefficient of the template, 3 × 3 Gaussian filter
Form panel
I.e. a Gaussian template of
15.3) frequency domain image ImageF Filter1Convolving the Gaussian template to obtain ImageF1
16. Setting the 2 nd Gaussian Filter Filter2The Gaussian distribution parameter kernel size is 3 x 3, sigma is 3, filtering is carried out on the image to obtain a filtered image ImageF2Specific operating parameters the above-described step 15,
17. thirdly, the frequency domain image ImageF is processed1、ImageF2Conversion into a spatial domain Image by inverse Fourier transform1，Image2，
18. Image is recorded1，Image2Subtracting to obtain Image3，Image3Most low-frequency signals are filtered out, high-frequency signals are highlighted, flaw detection is facilitated,
19. by a set threshold value t1For Image3Gray scale greater than t1The gray value of the pixel is 255, and the gray value of the other pixel is 0, so that a binary image is obtainedBinImage2，
20. BinImage is labeled by connected region labeling Algorithm (see step 5)2Marking the pixels with the gray value of 255 to obtain each connected region Rs1，
21. Traverse Rs1And respectively calculating the height h of the minimum circumscribed rectangletAnd width wtAnd comparing with a set defect screening height threshold h and width threshold w, such as htGreater than h or wtGreater than w, the region is placed in a stack,
22. the area blocks in the stack are defect areas, and if the number of the areas in the stack is more than 0, the system is judged to detect defects.
Beneficially, a roller 5 for providing power for online monitoring is arranged at a position below the synthetic leather 6.
Advantageously, the roller 5 is fitted with an encoder which also operates as the roller 5 rotates with the motor, for controlling the camera to capture images.
Beneficially, the linear scanning camera 1 is fixedly installed at a position above the synthetic leather 6, and the linear scanning lens 2 with a proper focal length is installed on the linear scanning camera 1.
Beneficially, the line scanning DOME light source 4 is arranged above the synthetic leather, and the line scanning DOME light source 4 adopts a sphere integral design, so that full-angle illumination can be provided for a strong light-reflecting object and a curved object; the special coating in the device ensures soft and uniform light, highlights the fine characteristics of the surface of the measured object, and is suitable for detecting the flaws of the highly reflective synthetic leather adhesive coating.
Beneficially, the linear scanning coaxial light source 3 is fixedly arranged at the position above the synthetic leather, a slit exists at the top end of the scanning DOME light source 3, so that the scanning DOME light source lacks vertically downward light, uniformity of images is affected, the linear scanning coaxial light source 3 can provide a vertically downward light source, and the light source uniformity can be guaranteed by combining with the linear scanning DOME light source 4.
Beneficially, the center of the slit at the top end of the line scanning coaxial light source 3, the center of the slit of the line scanning DOME light source 4 and the center of the line scanning camera 1 are coplanar, so that the camera can acquire synthetic leather images.
The specific operation method of the invention is as follows:
1. starting a line scanning DOME light source and a line scanning coaxial light source line and adjusting the brightness of the two light sources to ensure that the synthetic leather image is uniform and has obvious flaws on a camera acquisition point;
2. the starting motor drives the roll shaft to rotate, the roll shaft drives the synthetic leather on the roll shaft to slowly move, the encoders on the roll shafts of the colleagues work simultaneously, and the camera is controlled to collect images;
3. detecting the turned edge of the synthetic leather, namely other flaws, by an image processing algorithm;
4. if the curling is detected, the system alarms;
5. if other defects are detected, the position is recorded and the result is fed back to the computer in the monitoring room.
The above description is only an embodiment of the invention, but the scope of the invention is not limited thereto, and any changes or substitutions that are not thought of through the inventive work should be included in the scope of the invention. Therefore, the protection scope of the invention should be subject to the protection scope defined by the claims.
Claims (6)
1. A synthetic leather gluing online monitoring method based on machine vision is characterized by comprising the following steps: the method comprises the following steps:
1) collecting: collecting a synthetic leather Image by using a line scanning camera;
2) initial inspection: carrying out primary inspection processing on the synthetic leather Image;
3) detecting turned edges: performing first comparison on the data obtained by the initial detection processing and the image parameters;
4) if the curling is detected, the system quits the algorithm detection;
5) and (4) supplementary inspection: if no curling is detected, the computer continues to perform the supplementary inspection on the synthetic leather area through an edge angle algorithm;
6) and (3) confirming defects: performing secondary comparison on the parameters obtained by the supplementary inspection and the set defect screening parameters to detect the defects of the synthetic leather;
the initial detection processing in the step 2) comprises the following steps:
2.1) converting the Image into a gray level Image GrayImage;
2.2) calculating the height h of the synthetic leather image GrayImage;
2.3) carrying out binarization on GrayImage through a set threshold value t to obtain a binary image BinImage1；
2.4) BinImage by connected region labeling Algorithm1Marking is carried out, each individual connected region forms an identified block, and all connected regions R in the image are obtainedS；
2.5) traversing all connected regions RSCounting the number of pixels of each block, screening the area with the largest number of pixels, marking as R, filling R by using a flooding filling method to obtain a synthetic leather image area R1；
2.6) extracting the region R1The outer contour C of (a);
2.7) segmenting C according to a contour algorithm and extracting a left contour CLAnd the right contour CR；
The contour algorithm is as follows: the outer contour C is composed of four sub-contours, namely an upper sub-contour, a lower sub-contour, a left sub-contour and a right sub-contour, because the image is in a continuous state, the upper sub-contour is a straight line which is parallel to the x coordinate and the y coordinate is 0, the lower sub-contour is a straight line which is parallel to the x coordinate and the y coordinate is h, the h is the height of the image, pixels on the outer contour C are traversed, and the minimum value x of the x coordinate in all the pixels is obtained1Maximum value x2Traversing pixels on the outer contour C, if the y coordinate is larger than 0 and smaller than h, judging the x coordinate: greater than x1-1, and less than x1+(x2-x1) The/2 pixels are placed in the stack L, larger than x1+(x2-x1) A/2 and less than x2+1 pixels are placed in the stack R, the set of pixels in the stack L being the left contour CLThe set of pixels in the stack R is the right contour CR；
Since the normal cloth edge is jagged, its contour length should be greater than 1.5 times the fitting length, but after hemming occurs, its edge is almost straight, and its contour length should be less than 1.5 times the fitting length, so the first alignment includes: respectively count CLAnd CRNumber of pixels lLAnd lRAnd compared with the height h of the image, e.g. lLAnd lRIf the sizes of the synthetic leather are respectively larger than 1.5 multiplied by h, the edges of the synthetic leather are normal, otherwise, the synthetic leather is curled;
because the edge of the curled edge is not perpendicular to the cloth advancing direction, the edge angle algorithm is as follows: c is to beLAnd CRRespectively fitted into straight lines LLAnd LRAnd calculate their angleLAnd angleRE.g. angleL-an absolute value of 90 ° greater than 15 °, or angleRAn absolute value of-90 ° greater than 15 °, curling occurs.
2. The machine vision-based synthetic leather gluing online monitoring method of claim 1, characterized in that: the step 2.4) specifically comprises the following steps:
2.4.1) traversal of the image BinImage1Building an identification number for each pixel until the gray value of the pixel point is 255, and marking the area;
2.4.2) traversing 8 adjacent pixels of the pixel, and putting the pixel with the gray value of 255 into the identification area;
2.4.3) traversing the pixels in the identification number, and executing the step 2.4.2) until the gray values of 8 adjacent pixels are all 0;
2.4.4) continue traversing the BinImage1Each pixel is marked with an identification number until the gray value of the pixel point is 255 and the pixel point is not in the identified area;
2.4.5) performing step 2.4.2), 2.4.3);
2.4.6) repeating the steps 2.4.4) and 2.4.5) until the traversal is completed and all the identification blocks are obtained, and then all the connected regions R in the image can be obtainedS；
3. The machine vision-based synthetic leather gluing online monitoring method of claim 1, characterized in that: extracting region R1The specific operation of the outer contour C of (a) is: traversing each pixel of the image and judging eight adjacent pixels of the pixelPositions, e.g. eight adjacent pixels, at least one adjacent in R1Simultaneously at least 1 pixel in R1And putting the pixels into a stack, wherein the combination of the pixels in the stack is the outline C.
4. The machine vision-based synthetic leather gluing online monitoring method of claim 1, characterized in that: the supplementary inspection comprises the following steps:
5.1) carrying out Fourier transform on the GrayImage to obtain a frequency domain image ImageF of the GrayImage;
5.2) setting the first Gaussian Filter Filter1And filtering the frequency domain image ImageF to obtain a filtered image ImageF1；
5.3) setting the second Gaussian Filter Filter2And filtering the frequency domain image ImageF to obtain a filtered image ImageF2；
5.4) image ImageF1、ImageF2Conversion into a spatial domain Image by inverse Fourier transform1、Image2；
5.5) Image1、Image2Subtracting to obtain Image3；
5.6) passing the set threshold t1For Image3Processing to obtain a binary image BinImage2；
5.7) BinImage with connected region labeling Algorithm2Marking to obtain each communication area RS 1;
5.8) traversing RS1 and respectively calculating the height h of the minimum circumscribed rectangletAnd width wt。
5. The machine vision-based synthetic leather gluing online monitoring method of claim 1, characterized in that: the second alignment comprises: will be of height htAnd width wtComparing with a set defect screening height threshold h 'and width threshold w', such as htGreater than h' or wtGreater than w', the region is placed in a stack,the area blocks in the stack are defect areas, and if the number of the areas in the stack is more than 0, the system is judged to detect defects.
6. The machine vision-based synthetic leather gluing online monitoring method of claim 1, characterized in that: the device comprises a roller shaft, an encoder, a linear scanning camera, a linear scanning lens, a linear scanning DOME light source, a linear scanning coaxial light source, a linear scanning camera center and a camera collecting device, wherein the roller shaft is arranged at the lower position of the synthetic leather, the encoder is assembled on the roller shaft, the linear scanning camera is fixedly installed at the upper position of the synthetic leather, the linear scanning camera is provided with the linear scanning lens with a proper focal length, the linear scanning DOME light source and the linear scanning coaxial light source are arranged at the upper position of the synthetic leather, the center of a slit at the top end of the linear scanning.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110139661.2A CN112950541B (en) | 2019-04-12 | 2019-04-12 | Online monitoring method for gluing synthetic leather based on machine vision |
| CN201910295581.9A CN110084787B (en) | 2019-04-12 | 2019-04-12 | Synthetic leather gluing online monitoring method based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910295581.9A CN110084787B (en) | 2019-04-12 | 2019-04-12 | Synthetic leather gluing online monitoring method based on machine vision |
Related Child Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110139661.2A Division CN112950541B (en) | 2019-04-12 | 2019-04-12 | Online monitoring method for gluing synthetic leather based on machine vision |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110084787A CN110084787A (en) | 2019-08-02 |
| CN110084787B true CN110084787B (en) | 2021-03-23 |
Family
ID=67415008
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110139661.2A Active CN112950541B (en) | 2019-04-12 | 2019-04-12 | Online monitoring method for gluing synthetic leather based on machine vision |
| CN201910295581.9A Active CN110084787B (en) | 2019-04-12 | 2019-04-12 | Synthetic leather gluing online monitoring method based on machine vision |
Family Applications Before (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110139661.2A Active CN112950541B (en) | 2019-04-12 | 2019-04-12 | Online monitoring method for gluing synthetic leather based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN112950541B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113567447A (en) * | 2019-08-07 | 2021-10-29 | 浙江大学台州研究院 | Synthetic leather hemming online detection method |
| CN110927172B (en) * | 2019-12-10 | 2020-08-25 | 南京航空航天大学 | Online detection device and method for missing coating of sealant of integral fuel tank of airplane |
| CN114486903B (en) * | 2021-12-06 | 2024-05-14 | 浙江大学台州研究院 | Gray-scale self-adaptive coiled material detection system, device and algorithm |
| CN116046791B (en) * | 2023-03-30 | 2023-06-09 | 深圳市元硕自动化科技有限公司 | Method and device for detecting defect of adhesive dispensing |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102818809A (en) * | 2012-09-05 | 2012-12-12 | 杭州瑞利测控技术有限公司 | Gray cloth defect on-line detecting system based on machine vision and achieving method |
| CN103234976A (en) * | 2013-04-03 | 2013-08-07 | 江南大学 | Warp knitting machine cloth flaw on-line visual inspection method based on Gabor transformation |
| CN104268505A (en) * | 2014-09-12 | 2015-01-07 | 河海大学常州校区 | Automatic cloth defect point detection and recognition device and method based on machine vision |
| CN204174460U (en) * | 2014-10-21 | 2015-02-25 | 宿州学院 | High-Speed Automatic cloth inspecting machine |
| CN205484106U (en) * | 2016-01-11 | 2016-08-17 | 深圳市麦克斯泰有限公司 | Flaw off -line monitoring device of cloth |
| CN106898563A (en) * | 2015-12-18 | 2017-06-27 | 中芯国际集成电路制造(上海)有限公司 | Product Acceptance Review system and Product Acceptance Review method |
| CN107220649A (en) * | 2017-05-27 | 2017-09-29 | 江苏理工学院 | A kind of plain color cloth defects detection and sorting technique |
| CN107643295A (en) * | 2017-08-24 | 2018-01-30 | 中国地质大学（武汉） | A kind of method and system of the cloth defect on-line checking based on machine vision |
| CN109366830A (en) * | 2018-11-23 | 2019-02-22 | 安徽宏实紫晶光电研究所有限公司 | A kind of automatic aligning detection device and detection method for laminating machine |
Family Cites Families (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| ATE190722T1 (en) * | 1995-05-10 | 2000-04-15 | Mahlo Gmbh & Co Kg | METHOD AND DEVICE FOR DETECTING DEFECTS IN TEXTILE WEAVES AND THE LIKE |
| DE10361018C9 (en) * | 2003-12-23 | 2021-03-04 | QUISS Qualitäts-Inspektionssysteme und Service GmbH | Method for recognizing a structure to be applied to a substrate with a plurality of cameras and a device therefor |
| KR100772023B1 (en) * | 2005-03-31 | 2007-10-31 | 고등기술연구원연구조합 | Edge Arrangement Control System For Rolling Textile fabrics |
| CN201897572U (en) * | 2010-03-02 | 2011-07-13 | 新奥光伏能源有限公司 | Detecting system for glass broken edge |
| CN106033534B (en) * | 2015-03-18 | 2020-01-31 | 成都理想境界科技有限公司 | Electronic paper marking method based on straight line detection |
| FR3053792B1 (en) * | 2016-07-06 | 2023-07-14 | Tiama | METHOD, DEVICE AND INSPECTION LINE FOR DETERMINING A BURR AT THE LOCATION OF AN INTERNAL EDGE OF A RING SURFACE |
-
2019
- 2019-04-12 CN CN202110139661.2A patent/CN112950541B/en active Active
- 2019-04-12 CN CN201910295581.9A patent/CN110084787B/en active Active
Patent Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102818809A (en) * | 2012-09-05 | 2012-12-12 | 杭州瑞利测控技术有限公司 | Gray cloth defect on-line detecting system based on machine vision and achieving method |
| CN103234976A (en) * | 2013-04-03 | 2013-08-07 | 江南大学 | Warp knitting machine cloth flaw on-line visual inspection method based on Gabor transformation |
| CN104268505A (en) * | 2014-09-12 | 2015-01-07 | 河海大学常州校区 | Automatic cloth defect point detection and recognition device and method based on machine vision |
| CN204174460U (en) * | 2014-10-21 | 2015-02-25 | 宿州学院 | High-Speed Automatic cloth inspecting machine |
| CN106898563A (en) * | 2015-12-18 | 2017-06-27 | 中芯国际集成电路制造(上海)有限公司 | Product Acceptance Review system and Product Acceptance Review method |
| CN205484106U (en) * | 2016-01-11 | 2016-08-17 | 深圳市麦克斯泰有限公司 | Flaw off -line monitoring device of cloth |
| CN107220649A (en) * | 2017-05-27 | 2017-09-29 | 江苏理工学院 | A kind of plain color cloth defects detection and sorting technique |
| CN107643295A (en) * | 2017-08-24 | 2018-01-30 | 中国地质大学（武汉） | A kind of method and system of the cloth defect on-line checking based on machine vision |
| CN109366830A (en) * | 2018-11-23 | 2019-02-22 | 安徽宏实紫晶光电研究所有限公司 | A kind of automatic aligning detection device and detection method for laminating machine |
Non-Patent Citations (2)
| Title |
|---|
| Skew Detection of Fabric Images Based on Edge Detection and Projection Profile Analysis;LIU Z F等;《Foundations of Intelligent Systems》;20121231;论文正文 * |
| 基于机器视觉和图像处理的色织物疵点自动检测研究;李文羽;《中国博士学位论文全文数据库工程科技Ⅰ辑》;20140515;论文正文摘要、第1-2章 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112950541A (en) | 2021-06-11 |
| CN110084787A (en) | 2019-08-02 |
| CN112950541B (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110084787B (en) | Synthetic leather gluing online monitoring method based on machine vision | |
| CN108460757B (en) | Mobile phone TFT-LCD screen Mura defect online automatic detection method | |
| CN108230324B (en) | Visual detection method for microdefect on surface of magnetic shoe | |
| CN108931531B (en) | Method, system and computer readable storage medium for automatically detecting fabric defects | |
| CN105158258B (en) | A kind of bamboo cane detection method of surface flaw based on computer vision | |
| Li et al. | Detection of surface crack defects on ferrite magnetic tile | |
| US7495758B2 (en) | Apparatus and methods for two-dimensional and three-dimensional inspection of a workpiece | |
| CN103913468B (en) | Many defects of vision checkout equipment and the method for large-scale LCD glass substrate on production line | |
| Boukouvalas et al. | Ceramic tile inspection for colour and structural defects | |
| KR101895873B1 (en) | Method and apparatus for fabric inspection | |
| CN104792794A (en) | Machine vision based optical film surface defect detecting method | |
| CN110044910B (en) | Automobile cup box part detection system and detection method | |
| CN108827181B (en) | Vision-based plate surface detection method | |
| CN109507192A (en) | A kind of magnetic core detection method of surface flaw based on machine vision | |
| CN113554080A (en) | Non-woven fabric defect detection and classification method and system based on machine vision | |
| CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
| CN104574353A (en) | Surface defect judgment method based on visual saliency | |
| CN115100206B (en) | Printing defect identification method for textile with periodic pattern | |
| CN113706490B (en) | Wafer defect detection method | |
| CN107328781A (en) | A kind of columnar product detection method of surface flaw and device based on machine vision | |
| JP7282790B2 (en) | Virtual camera array for inspecting manufactured webs | |
| WO2020119246A1 (en) | Highlight object surface defect detection system and method based on reflected light | |
| CN104048966B (en) | The detection of a kind of fabric defect based on big law and sorting technique | |
| CN111047556A (en) | Strip steel surface defect detection method and device | |
| CN113820319A (en) | Textile surface defect detection device and method |
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 |