CN113063731A - Detection system and detection method for rotary disc type glass cover plate silk screen printing - Google Patents
Detection system and detection method for rotary disc type glass cover plate silk screen printing Download PDFInfo
- Publication number
- CN113063731A CN113063731A CN202110312387.4A CN202110312387A CN113063731A CN 113063731 A CN113063731 A CN 113063731A CN 202110312387 A CN202110312387 A CN 202110312387A CN 113063731 A CN113063731 A CN 113063731A
- Authority
- CN
- China
- Prior art keywords
- screen
- image
- unit
- silk
- template
- 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
Images
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/84—Systems specially adapted for particular applications
- G01N21/88—Investigating the presence of flaws or contamination
- G01N21/95—Investigating the presence of flaws or contamination characterised by the material or shape of the object to be examined
- G01N21/956—Inspecting patterns on the surface of objects
- G01N21/95607—Inspecting patterns on the surface of objects using a comparative method
Abstract
The invention discloses a detection system and a detection method for rotating disc type glass cover plate silk screen printing, wherein the system comprises the following steps: the camera module comprises three CCD cameras; the modeling module comprises a calibration unit, a first calculation unit and a storage unit, wherein the calibration unit is connected with the three CCD cameras, the first calculation unit and the storage unit, and the first calculation unit is connected with the storage unit; and the comparison module comprises a loading unit, a preprocessing unit, a second calculating unit, an extracting unit, a third calculating unit and a screening unit, wherein the loading unit is connected to the storage unit, the preprocessing unit is connected to the CCD camera, the loading unit, the second calculating unit and the extracting unit, and the third calculating unit is connected to the extracting unit and the screening unit. The invention solves the problem that the detection of tiny defects such as sand holes and the like has a large amount of misjudgments in the traditional line frame printing detection of the glass cover plate of the mobile phone.
Description
Technical Field
The invention relates to the technical field of silk-screen detection, in particular to a detection system and a detection method for rotating disc type glass cover plate silk-screen.
Background
Aiming at the line frame printing inspection of the mobile phone glass cover plate, the prior art only simply observes the obvious poor line frame of the silk screen, and the detection method is mostly based on the gray scale comparison of images and then analyzes the integral integrity and the flatness of the images. The method cannot detect edge defects in the wire frame and fine sand holes caused by dust sticking of the silk screen, and a large amount of misjudgments exist because the gray scale change signals of the sand holes are close to the periphery.
Disclosure of Invention
In order to overcome the defects in the prior art, a detection system and a detection method for rotary table type glass cover plate silk screen printing are provided so as to solve the problem that a great amount of misjudgments exist in the detection of tiny defects such as sand holes and the like in the traditional mobile phone glass cover plate line frame printing detection.
In order to realize the purpose, the detection method for the silk screen of the rotating disc type glass cover plate comprises the following steps:
the method comprises the following steps:
dividing a silk-screen template into two end sections covering angular points at two opposite ends of an inner frame of the template silk-screen and a middle section connected to the two end sections along the length direction of the silk-screen template, wherein the inner frame is rectangular;
the three camera modules respectively acquire first sectional images of the two end head sections and the middle section;
obtaining coordinates of the angular points by using a fitted straight line four-edge rectangle positioning algorithm and calibrating the position of the template silk-screen in the first sectional image;
carrying out affine transformation on the three first sectional images by utilizing a three-point calculation affine transformation algorithm and a WarpAffini algorithm to obtain a template silk-screen sectional image;
the camera module respectively acquires second sectional images of the end section and the middle section of the cover plate to be detected;
performing affine transformation and normalization processing on the second segmented image according to the coordinates of the angular points to obtain a silk-screen segmented image to be detected;
carrying out image pixel subtraction on the silk-screen segmented image to be detected and the template silk-screen segmented image to obtain a difference segmented image;
extracting the interested contour of the difference segmented image to obtain a contour image of the screen printing to be detected of the cover plate to be detected;
comparing the outline of the silk screen to be detected in the outline image with the pixel points of the outline of the template silk screen in the first sectional image by utilizing an edge pixel cumulative change algorithm and recording difference pixel points to obtain an outline defect initial detection image;
and screening the defects in the profile defect initial inspection image based on preset defect parameters to obtain a defect profile image.
Further, the step of obtaining coordinates of the corner points and calibrating the position of the template silk screen in the first segmented image by using a fitted straight line four-edge rectangle-forming positioning algorithm comprises:
binarizing the first segmented image to obtain a vertical segment and a horizontal segment;
obtaining the outline of the template silk screen after merging the line segments of the acquired vertical line segment and the acquired horizontal line segment and eliminating the line segments which do not accord with preset elimination parameters;
establishing a two-dimensional rectangular coordinate system in the first segmented image, and obtaining two corner points P of an inner frame of the template silk screen based on intersection points of vertical line segments and horizontal line segments in the outline of the template silk screen1(x1，y1)、P2(x2，y2)；
Deriving and obtaining a corner point P by coordinates of two corner points3(x1，y1+. at), corner point P4(x2，y2+△t)；
Based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2And t) calibrating the position of the template silk screen in the first segmented image.
Further, the step of performing affine transformation on the three first segmented images by using the three-point affine transformation algorithm and the warp affine algorithm to obtain the template silk-screen segmented images comprises:
based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2+ DELTAt), obtaining a 2 multiplied by 3 output matrix by utilizing three-point to calculate affine transformation algorithm;
and performing affine transformation on the image by using the 2 multiplied by 3 output matrix by using a WarpAffini algorithm to obtain the obtained template silk-screen segmented image.
Further, the step of extracting the interested contour of the difference segmented image to obtain the contour image of the to-be-detected silk screen of the to-be-detected cover plate includes:
extracting a closed contour from the difference segmented image by using a contour extraction function of an OpenCV function library;
and extracting the interested inner and outer contours of the silk screen to be detected according to the size, shape and position of the contours so as to obtain a contour image of the silk screen to be detected.
Further, the defect parameters include the size of the pixel area of the defect and the position of the defect.
Further, after the step of screening the defects in the profile defect initial inspection image to obtain a defect profile image, the defect profile image is displayed after being spliced.
The invention provides a detection system for rotating disc type glass cover plate silk screen printing, which comprises the following components:
the camera module comprises a three-CCD camera for acquiring a first sectional image of the silk-screen template and a second sectional image of the cover plate to be detected;
the modeling module comprises a calibration unit, a first calculation unit and a storage unit, wherein the calibration unit is used for acquiring coordinates of the angular points by using a fitted straight line four-edge rectangular positioning algorithm and calibrating the position of the template silk-screen in the first sectional image, the first calculation unit is used for performing affine transformation on the three first sectional images by using a three-point-to-point calculation affine transformation algorithm and a warp affine algorithm to acquire the template silk-screen sectional image, the storage unit is used for storing the coordinates of the angular points and the template silk-screen sectional image, the calibration unit is connected with the three CCD cameras, the first calculation unit and the storage unit, and the first calculation unit is connected with the storage unit; and
a comparison module, including a loading unit for retrieving coordinates of the corner points and the template silk-screen segmented image, a preprocessing unit for performing affine transformation and normalization processing on the second segmented image according to the coordinates of the corner points to obtain a silk-screen segmented image to be detected, a second computing unit for performing image pixel subtraction on the silk-screen segmented image to be detected and the template silk-screen segmented image to obtain a difference segmented image, an extraction unit for extracting an interesting profile in the difference segmented image to obtain the silk-screen to be detected of the cover plate to be detected, a third computing unit for comparing the profile of the silk-screen to be detected in the profile image with pixel points of the profile of the template silk-screen in the first segmented image by using an edge pixel accumulation variation algorithm and recording the difference pixel points to obtain a profile defect initial detection image, and a third computing unit for screening defects in the profile defect initial detection image based on preset defect parameters to obtain a defect wheel The device comprises a storage unit, a loading unit, a preprocessing unit, a CCD camera, a loading unit, a second calculating unit and an extracting unit, wherein the storage unit is used for storing the CCD camera, the loading unit, the second calculating unit and the extracting unit, and the third calculating unit is connected with the extracting unit and the screening unit.
The detection system and the detection method for the rotary disc type glass cover plate silk screen printing have the advantages that the three CCD cameras are adopted to cooperatively process images, the module structures of the mold making module and the comparison module are small in size and small in occupied space, and the detection system and the detection method can be flexibly applied to the rotary disc type cover plate silk screen printing production process; meanwhile, multiple groups of high-precision cameras are adopted to process images in a splicing mode, so that high resolution and high precision of the images can be ensured, and misjudgment of detection of tiny defects such as sand holes is greatly reduced. The detection system and the detection method for the rotary table type glass cover plate silk screen printing have remarkable effects on real-time detection of glass cover plate printing, solve the problem of automatic detection of rotary table type cover plate printing, and have universal application value on wide rotary table type printing defect detection.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the following detailed description of non-limiting embodiments thereof, made with reference to the accompanying drawings in which:
fig. 1 is a schematic block diagram of a detection system for screen printing of a rotating disk type glass cover plate according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of a template silk-screen printing according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of a first segment image of an end segment of a template screen printing according to an embodiment of the present invention.
Fig. 4 is a schematic diagram of a first segment image of a middle segment of a template screen printing according to an embodiment of the present invention.
Fig. 5 is a schematic diagram of a first segment image of another end segment of the template screen printing according to the embodiment of the present invention.
Fig. 6 is a schematic diagram of a template silk-screen segmented image of a tip segment of a template silk-screen according to an embodiment of the present invention.
Fig. 7 is a schematic diagram of an outline image of a tip section of a template screen printing according to an embodiment of the present invention.
Fig. 8 is a schematic diagram of a differential segmented image of a tip segment of a template screen printing according to an embodiment of the present invention.
Fig. 9 is a schematic diagram of an outline defect initial inspection image of a tip section of a template screen printing according to an embodiment of the present invention.
Fig. 10 is a schematic diagram of a template screen printing outline original position reference diagram of a tip section of the template screen printing according to the embodiment of the present invention.
Fig. 11 is a schematic diagram of a defect outline image of the other end head section of the template screen printing according to the embodiment of the invention.
Fig. 12 is a local defect diagram of the template screen printing according to the embodiment of the invention.
Fig. 13 is a defect effect diagram of the template silk-screen printing according to the embodiment of the invention.
Detailed Description
The present application will be described in further detail with reference to the following drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the relevant invention and not restrictive of the invention. It should be noted that, for convenience of description, only the portions related to the present invention are shown in the drawings.
It should be noted that the embodiments and features of the embodiments in the present application may be combined with each other without conflict. The present application will be described in detail below with reference to the embodiments with reference to the attached drawings.
Referring to fig. 1 to 13, the present invention provides a detection system for a rotary disk type glass cover plate screen printing, including: the device comprises a camera module, a modeling module 2 and a comparison module 3.
Specifically, the camera module includes a three-CCD (Charge-coupled Device) camera 1. The CCD camera 1 is used for obtaining a first sectional image of the silk-screen template and a second sectional image of the cover plate to be detected.
The modeling module 2 includes a calibration unit 21, a first calculation unit 22, and a storage unit 23. The calibration unit 21 is connected to the three CCD cameras 1. A first calculating unit 22 and a storage unit 23, the first calculating unit 22 being connected to the storage unit 23.
The calibration unit 21 is configured to obtain coordinates of the corner points by using a positioning algorithm for fitting a straight line and forming a rectangle on four sides of the straight line, and calibrate the position of the template silk-screen in the first segmented image.
The storage unit 23 is configured to perform affine transformation on the three first segment images by using a three-point-to-point affine transformation algorithm and a warp affine algorithm (the warp affine is a rotation transformation matrix function, and is a main function in affine transformation) to obtain a first calculation unit 22 for storing coordinates of corner points and a template silk-screen segment image.
The comparing module 3 includes a loading unit 31, a preprocessing unit 32, a second calculating unit 33, an extracting unit 34, a third calculating unit 35, and a screening unit 36. Wherein, the loading unit 31 is connected to the storage unit 23, the preprocessing unit 32 is connected to the three CCD cameras 1, the loading unit 31, the second calculating unit 33 and the extracting unit 34, and the third calculating unit 35 is connected to the extracting unit 34 and the screening unit 36.
The loading unit 31 is configured to retrieve coordinates of corner points and a template silk-screen segmented image.
The preprocessing unit 32 is configured to perform affine transformation and normalization processing on the second segmented image according to the coordinates of the corner points to obtain a to-be-detected silk-screen segmented image.
The second calculating unit 33 is configured to perform image pixel subtraction on the to-be-detected silk-screen segmented image and the template silk-screen segmented image to obtain a difference segmented image.
The extracting unit 34 is configured to extract an interested contour in the difference segmented image to obtain a to-be-detected silk-screen contour image of the to-be-detected cover plate.
The third calculating unit 35 is configured to compare pixel points of the profile of the to-be-detected silk screen in the profile image with pixel points of the profile of the template silk screen in the first segmented image by using an edge pixel cumulative change algorithm, and record difference pixel points to obtain a profile defect initial inspection image.
The screening unit 36 is configured to screen the defects in the profile defect preliminary inspection image based on preset defect parameters to obtain a defect profile image.
As a preferred embodiment, the detection system for screen printing of the rotating disk type glass cover plate of the present invention further comprises a display module 4 connected to the screening unit 36. Specifically, the display module displays the defect outline image after splicing through the splicing unit, so that the defect outline image is convenient for a detector to observe.
In some embodiments, the detection system for silk-screen printing of the rotating disc type glass cover plate further comprises a judgment module 5. The judgment module 5 is connected to the screening unit 36 and the display module 4.
The judgment template comprises a classification unit and a judgment unit. And classifying and sorting the edge defects and the defects between the inner frame and the outer frame of the silk screen based on the position range of the defects. The judgment unit is preset with a detection standard set by a user, whether the memory unit for detecting the adjacent images is in error reporting, shut down or added is judged based on the detection standard, and the batch defects are further detected according to the poor repeatability of the adjacent images and the user is reported in time.
The invention provides a detection method for silk screen printing of a rotating disc type glass cover plate, which comprises the following steps:
s1: dividing the silk screen template into two end head sections covering angular points at two opposite ends of an inner frame of the template silk screen and a middle section connected to the two end head sections along the length direction of the silk screen template, wherein the inner frame is rectangular.
S2: the three camera modules respectively acquire first sectional images of head sections and middle sections at two ends.
When the template is manufactured, the three CCD cameras are respectively aligned to the end section and the middle section of the silk-screen template, and first sectional images of the end section and the middle section of the silk-screen template are obtained through shooting.
S3: and acquiring coordinates of the corner points by using a fitted straight line four-edge rectangle positioning algorithm and calibrating the position of the template silk-screen in the first sectional image.
Wherein, step S3 includes the following steps:
s31: a calibration unit of the modeling module carries out binarization on the first segmented image to obtain a vertical line segment and a horizontal line segment;
s32: obtaining the outline of the template silk screen after merging the line segments of the acquired vertical line segment and the acquired horizontal line segment and eliminating the line segments which do not accord with preset elimination parameters;
s33: establishing a two-dimensional rectangular coordinate system in the first segmented image, and obtaining two corner points P of an inner frame of the template silk screen based on intersection points of vertical line segments and horizontal line segments in the outline of the template silk screen1(x1，y1)、P2(x2，y2)。
S34: deriving and obtaining a corner point P by coordinates of two corner points3(x1，y1+. at), corner point P4(x2，y2+△t)。
S35: based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2And +/-deltat) calibrating the position of the template silk screen in the first sectional image.
In the first segmented image of an end segment, the coordinates of the four corner points (i.e., P) in FIG. 3 are saved1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P3(x2，y2+. at)) to load call seats at the time of official testAnd (4) marking.
Saving the coordinates of the four corner points in fig. 5 (i.e., P)5(x1，y3)、P6(x2，y4)、P7(x1，y3+△t)，P8(x2，y4+. DELTA.t)) to load the call coordinates at the time of the official test. In the first segment image of the other end head segment, the positioning mode of the template silk screen printing is the same as that in fig. 3, but the x coordinates of two corner points of the other end head segment are consistent with the x coordinates in fig. 3, so as to ensure that the first segment images of the two end head segments and the middle segment are consistent in the vertical direction, store the coordinates of four corner points of the other end head segment and store the template image.
Saving the coordinates of the four corner points in figure 4 (i.e. P)9(x1，y5)、P10(x2，y6)、P11(x1，y5+h)，P12(x2，y5+ h)) to load the call coordinates at the time of official testing. For the first segment image of the middle segment, the accurate positioning of the silk screen of the template only needs to be kept consistent with the x coordinate in fig. 3, the y coordinate is the fixed view field height h of the CCD camera, and the template image is saved.
S4: and performing affine transformation on the three first sectional images by utilizing a three-point calculation affine transformation algorithm and a WarpAffini algorithm to obtain a template silk-screen sectional image.
Wherein the first computing unit of the modeling module utilizes the coordinates of the four corner points (i.e., P) maintained in the first segmented image1～P12) An affine transformation algorithm can be calculated by using GetAffiniTransform (image affine transformation) three-point pair to obtain a 2 x 3 output matrix, and then the 2 x 3 output matrix is combined with a WarpAffinine algorithm to perform affine transformation on the image to obtain the template silk-screen segmented image.
Specifically, for the first segmented image of the other end segment, the step of performing affine transformation on the three first segmented images by using the three-point affine transformation algorithm and the warp affine algorithm to obtain the template silk-screen segmented image comprises the following steps of:
based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2+ DELTAt), obtaining a 2 multiplied by 3 output matrix by utilizing three-point to calculate affine transformation algorithm;
and performing affine transformation on the image by using the 2 multiplied by 3 output matrix by using a WarpAffini algorithm to obtain the obtained template silk-screen segmented image. Template silk-screen segmented image and angular point (P)1～P12) Is stored in the storage unit.
Referring to fig. 3 to 5, the screen printing stencil is divided into two end sections covering corner points at two opposite ends of the inner frame of the stencil screen and a middle section connected to the two end sections along a length direction of the screen printing stencil. The inner frame of the template screen printing is rectangular, the outer frame of the template screen printing is similar to a rectangle, and the corner parts are rounded corners. In this embodiment, the image capturing module uses 3 sets of 20M CCD cameras to capture a part of images, and then combines with a 35mm lens to achieve high-precision positioning. Specifically, when a local shooting product line is adopted, a group of 20M camera pixels is 5496 × 3672; by adopting a 35mm lens, the fixed 275mm height can reach the actual product visual field of 30mm, the short side positioning precision is 30/5496-0.005 mm/pixel, the long side positioning precision is 30/3672-0.008 mm/pixel, and the integral precision is less than 0.01 mm. If a lens with the thickness of 25mm is adopted, the fixed height of 275mm can reach the visual field of an actual product of 55mm, the positioning accuracy of the short side is 55/5496-0.01 mm/pixel, and the positioning accuracy of the long side is 55/3672-0.014 mm/pixel; in contrast, the precision of the whole long edge and the short edge of the 35mm lens is obviously higher than that of the 25mm lens and is 2 times higher than that of the 35mm lens. Because a plurality of CCD cameras can only collect one third of template silk-screen images, at most two angular points (such as figures 3 to 4) can only be obtained, affine transformation positioning can not be directly carried out, and the problem can be solved by adopting an advanced positioning algorithm to fit four sides of a straight line into rectangular positioning.
S5: the three camera modules respectively acquire second sectional images of the end section and the middle section of the cover plate to be detected.
When the cover plate to be tested is tested, the three CCD cameras respectively aim at the end section and the middle section of the cover plate to be tested, and second sectional images of the end section and the middle section of the cover plate to be tested are obtained through shooting.
S6: and performing affine transformation and normalization processing on the second segmented image according to the coordinates of the angular points to obtain the silk-screen segmented image to be detected.
After the cover plate to be tested is shot in a segmented mode, the loading units of the comparison module respectively load the coordinates of the angular points in the storage units in the three CCD cameras, the preprocessing unit of the comparison module conducts affine transformation on the angular points in the second segmented image obtained by shooting of the CCD cameras in the positioning reference template silk-screen segmented image, the silk-screen segmented image to be tested at the position consistent with the template silk-screen segmented image is obtained, and normalization is guaranteed.
S7: and carrying out image pixel subtraction on the silk-screen segmented image to be detected and the template silk-screen segmented image to obtain a difference segmented image.
Specifically, the second calculating unit of the comparison module performs a subtraction algorithm on the silk-screen segmented image to be detected after the positioning compensation and the template silk-screen segmented image, all pixel points in the two images are subtracted, the pixel points without difference are subtracted to be 0, the difference is retained, and finally the difference is subtracted to form a new image, namely the difference segmented image, which is retained for next analysis.
S8: and extracting the interested contour of the difference segmented image to obtain a contour image of the screen printing to be detected of the cover plate to be detected.
Specifically, step S8 includes:
an extraction unit of the comparison module extracts a closed contour in the difference segmented image by using a contour extraction function of an OpenCV function library;
and extracting the interested inner and outer contours of the silk screen to be detected according to the size, shape and position of the contours so as to obtain a contour image of the silk screen to be detected.
The extraction unit of the one-step comparison module adopts a conventional contour extraction function in an OpenCV function library to extract closed contours one by one, and extracts interested inner and outer contours according to the size, shape and position of the contours so as to be used for further contour analysis, as shown in fig. 7.
S9: and comparing the outline of the screen printing to be detected in the outline image with the pixel points of the outline of the template screen printing of the first sectional image by utilizing an edge pixel cumulative change algorithm and recording the difference pixel points to obtain an outline defect initial detection image.
After the outline image of the to-be-detected silk screen of the to-be-detected cover plate is obtained, the third calculation unit of the comparison module uses an edge pixel cumulative change algorithm, pixel points of the to-be-detected silk screen outline of the to-be-detected cover plate are compared one by adopting a linear search method, the coordinates of the pixel points are greatly different in pairwise comparison, the pixel points with large differences are recorded, and very fine pixels are selected according to pixel precision, so that the concave-convex defect of the edge can be detected, and the effect is very obvious, as shown in fig..
S10: and screening the defects 6 in the profile defect initial inspection image based on preset defect parameters to obtain a defect profile image.
The defect parameters comprise the size of the area of the pixel of the defect and the position of the defect.
Due to the fact that the defects comprise sand holes, scratches, light leakage and the like, the size of the area of the pixels of the defects and the positions of the defects are set inside a silk-screen wire frame, the defects are selected by combining a reference image of the original position of the silk-screen outline of the ZnMap template, a screening unit of a comparison module screens the defects 6 in an outline defect primary detection image to obtain a defect outline image, interference defects such as dust/dirt and the like on the periphery of the wire frame can be avoided, and a final defect outline effect image is shown in fig. 11.
The reference picture of the original position of the silk-screen outline of the template is obtained by the following steps:
a. and carrying out binarization processing on the template silk-screen segmented image to generate a contour map.
b. And traversing the outline, and separating the inner frame and the outer frame of the template silk screen.
c. And copying a new image from the template silk-screen segmented image to be used as a ZnMap initial size image.
d. And (c) on the initial size diagram, drawing and widening the inner frame and the outer frame in the step (b), drawing different gray scales, drawing the gray scale of a printing area at the same time, and finally forming ZnMap diagrams (namely template silk-screen outline original position reference diagrams) with different gray scales.
The reference image of the original position of the silk-screen outline of the template is mainly used for distinguishing the positions of defects in an outer frame, an inner frame, a printing area and holes, and the areas can be classified and distinguished without gray levels.
After the step of screening the defects in the outline defect initial inspection image to obtain the defect outline image, based on the position range of the defects, the classification unit of the judgment template classifies and arranges the edge defects and the defects in the wire frame, the judgment unit of the judgment template for detecting the silk screen wire frame result in the previous step determines whether to report errors, stop the machine or add a memory unit for detecting adjacent images according to the detection standard set by a user, further detects the batch defects according to the poor repeatability of the adjacent images and reports the batch defects to the user in time, as shown in fig. 12.
After the step of screening the defects in the initial inspection image of the profile defect to obtain the defect profile image, the defect profile image is spliced and displayed, and the processed result image is displayed as shown in fig. 13, which is more convenient for people to observe.
According to the detection system and the detection method for the rotary disc type glass cover plate silk screen printing, three CCD cameras are adopted to cooperatively process images, the module structures of the mold making module and the comparison module are small in size and small in occupied space, and the detection system and the detection method can be flexibly applied to the rotary disc type cover plate silk screen printing production process; meanwhile, multiple groups of high-precision cameras are adopted to process images in a splicing mode, so that high resolution and high precision of the images can be ensured, and misjudgment of detection of tiny defects such as sand holes is greatly reduced. The detection system and the detection method for the rotary table type glass cover plate silk screen printing have remarkable effects on real-time detection of glass cover plate printing, solve the problem of automatic detection of rotary table type cover plate printing, and have universal application value on wide rotary table type printing defect detection.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by a person skilled in the art that the scope of the invention as referred to in the present application is not limited to the embodiments with a specific combination of the above-mentioned features, but also covers other embodiments with any combination of the above-mentioned features or their equivalents without departing from the inventive concept. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (7)
1. A detection method of a detection system for screen printing of a rotating disc type glass cover plate is characterized by comprising the following steps:
dividing a silk-screen template into two end sections covering angular points at two opposite ends of an inner frame of the template silk-screen and a middle section connected to the two end sections along the length direction of the silk-screen template, wherein the inner frame is rectangular;
the three camera modules respectively acquire first sectional images of the two end head sections and the middle section;
obtaining coordinates of the angular points by using a fitted straight line four-edge rectangle positioning algorithm and calibrating the position of the template silk-screen in the first sectional image;
carrying out affine transformation on the three first sectional images by utilizing a three-point calculation affine transformation algorithm and a WarpAffini algorithm to obtain a template silk-screen sectional image;
the camera module respectively acquires second sectional images of the end section and the middle section of the cover plate to be detected;
performing affine transformation and normalization processing on the second segmented image according to the coordinates of the angular points to obtain a silk-screen segmented image to be detected;
carrying out image pixel subtraction on the silk-screen segmented image to be detected and the template silk-screen segmented image to obtain a difference segmented image;
extracting the interested contour of the difference segmented image to obtain a contour image of the screen printing to be detected of the cover plate to be detected;
comparing the outline of the silk screen to be detected in the outline image with the pixel points of the outline of the template silk screen in the first sectional image by utilizing an edge pixel cumulative change algorithm and recording difference pixel points to obtain an outline defect initial detection image;
and screening the defects in the profile defect initial inspection image based on preset defect parameters to obtain a defect profile image.
2. The method for detecting the screen printing of the rotating disc type glass cover plate according to claim 1, wherein the step of obtaining the coordinates of the corner points and calibrating the position of the screen printing of the template in the first segmented image by using a fitted straight line four-side-rectangle positioning algorithm comprises:
binarizing the first segmented image to obtain a vertical segment and a horizontal segment;
obtaining the outline of the template silk screen after merging the line segments of the acquired vertical line segment and the acquired horizontal line segment and eliminating the line segments which do not accord with preset elimination parameters;
establishing a two-dimensional rectangular coordinate system in the first segmented image, and obtaining two corner points P of an inner frame of the template silk screen based on intersection points of vertical line segments and horizontal line segments in the outline of the template silk screen1(x1，y1)、P2(x2，y2)；
Deriving and obtaining a corner point P by coordinates of two corner points3(x1，y1+. at), corner point P4(x2，y2+△t)；
Based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2And t) calibrating the position of the template silk screen in the first segmented image.
3. The detection method for the silk screen of the rotating disc type glass cover plate according to claim 2, wherein the step of performing affine transformation on the three first sectional images by using three-point calculation affine transformation algorithm and a warp affine algorithm to obtain the template silk screen sectional images comprises the following steps:
based on P1(x1，y1)、P2(x2，y2)、P3(x1，y1+△t)，P4(x2，y2+ DELTAt), obtaining a 2 multiplied by 3 output matrix by utilizing three-point to calculate affine transformation algorithm;
and performing affine transformation on the image by using the 2 multiplied by 3 output matrix by using a WarpAffini algorithm to obtain the obtained template silk-screen segmented image.
4. The method for detecting the silkscreen of the rotating disc type glass cover plate according to claim 1, wherein the step of performing the interested contour extraction on the difference segmented image to obtain the contour image of the silkscreen to be detected of the cover plate to be detected comprises the following steps of:
extracting a closed contour from the difference segmented image by using a contour extraction function of an OpenCV function library;
and extracting the interested inner and outer contours of the silk screen to be detected according to the size, shape and position of the contours so as to obtain a contour image of the silk screen to be detected.
5. The method for detecting the silkscreen of the rotating disc type glass cover plate according to claim 1, wherein the defect parameters comprise the size of the area of the pixel of the defect and the position of the defect.
6. The method for inspecting screen printing of rotary table type glass cover plates according to claim 1, wherein after the step of screening the defects in the outline defect initial inspection image to obtain a defect outline image, the defect outline image is displayed after being spliced.
7. The detection system for the detection system of the screen printing of the rotating disc type glass cover plate as claimed in any one of claims 1 to 6, wherein the detection system comprises:
the camera module comprises a three-CCD camera for acquiring a first sectional image of the silk-screen template and a second sectional image of the cover plate to be detected;
the modeling module comprises a calibration unit, a first calculation unit and a storage unit, wherein the calibration unit is used for acquiring coordinates of the angular points by using a fitted straight line four-edge rectangular positioning algorithm and calibrating the position of the template silk-screen in the first sectional image, the first calculation unit is used for performing affine transformation on the three first sectional images by using a three-point-to-point calculation affine transformation algorithm and a warp affine algorithm to acquire the template silk-screen sectional image, the storage unit is used for storing the coordinates of the angular points and the template silk-screen sectional image, the calibration unit is connected with the three CCD cameras, the first calculation unit and the storage unit, and the first calculation unit is connected with the storage unit; and
a comparison module, including a loading unit for retrieving coordinates of the corner points and the template silk-screen segmented image, a preprocessing unit for performing affine transformation and normalization processing on the second segmented image according to the coordinates of the corner points to obtain a silk-screen segmented image to be detected, a second computing unit for performing image pixel subtraction on the silk-screen segmented image to be detected and the template silk-screen segmented image to obtain a difference segmented image, an extraction unit for extracting an interesting profile in the difference segmented image to obtain the silk-screen to be detected of the cover plate to be detected, a third computing unit for comparing the profile of the silk-screen to be detected in the profile image with pixel points of the profile of the template silk-screen in the first segmented image by using an edge pixel accumulation variation algorithm and recording the difference pixel points to obtain a profile defect initial detection image, and a third computing unit for screening defects in the profile defect initial detection image based on preset defect parameters to obtain a defect wheel The device comprises a storage unit, a loading unit, a preprocessing unit, a CCD camera, a loading unit, a second calculating unit and an extracting unit, wherein the storage unit is used for storing the CCD camera, the loading unit, the second calculating unit and the extracting unit, and the third calculating unit is connected with the extracting unit and the screening unit.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110312387.4A CN113063731B (en) | 2021-03-24 | 2021-03-24 | Detection system and detection method for rotary disc type glass cover plate silk screen printing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110312387.4A CN113063731B (en) | 2021-03-24 | 2021-03-24 | Detection system and detection method for rotary disc type glass cover plate silk screen printing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113063731A true CN113063731A (en) | 2021-07-02 |
| CN113063731B CN113063731B (en) | 2023-01-20 |
Family
ID=76561662
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110312387.4A Active CN113063731B (en) | 2021-03-24 | 2021-03-24 | Detection system and detection method for rotary disc type glass cover plate silk screen printing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113063731B (en) |
Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006284308A (en) * | 2005-03-31 | 2006-10-19 | Toshiba Corp | Visual examination method of semiconductor device |
| JP2010181254A (en) * | 2009-02-05 | 2010-08-19 | Ushio Inc | Device for inspecting screen print making |
| KR20100093213A (en) * | 2009-02-16 | 2010-08-25 | 유근홍 | System for inspecting defects on glass substrate using contrast value, and method of the same |
| US8103085B1 (en) * | 2007-09-25 | 2012-01-24 | Cognex Corporation | System and method for detecting flaws in objects using machine vision |
| CN105354844A (en) * | 2015-10-30 | 2016-02-24 | 清华大学 | Edge inflection point detection method and detection apparatus based on shape fitting |
| CN106127779A (en) * | 2016-06-29 | 2016-11-16 | 上海晨兴希姆通电子科技有限公司 | The defect inspection method of view-based access control model identification and system |
| CN108355987A (en) * | 2018-01-08 | 2018-08-03 | 西安交通大学 | A kind of screen printing of battery quality determining method based on piecemeal template matches |
| CN109064420A (en) * | 2018-07-13 | 2018-12-21 | 华南理工大学 | A kind of multi-layer PCB cross-sectional image technological parameter extracting method |
| CN109961464A (en) * | 2019-03-22 | 2019-07-02 | 武汉艾维视智能技术有限公司 | A kind of high-precision screen printing net plate scan image method for registering |
| CN110567976A (en) * | 2019-08-30 | 2019-12-13 | 中国地质大学（武汉） | mobile phone cover plate silk-screen defect detection device and detection method based on machine vision |
| CN110567965A (en) * | 2019-07-31 | 2019-12-13 | 华南理工大学 | Smartphone glass cover plate edge visual defect detection method |
| CN110672617A (en) * | 2019-09-14 | 2020-01-10 | 华南理工大学 | Method for detecting defects of silk-screen area of glass cover plate of smart phone based on machine vision |
| CN110909750A (en) * | 2019-11-14 | 2020-03-24 | 展讯通信（上海）有限公司 | Image difference detection method and device, storage medium and terminal |
| CN111204116A (en) * | 2020-02-13 | 2020-05-29 | 深圳市和力泰智能制造有限公司 | Silk screen printing equipment |
| CN111275697A (en) * | 2020-02-10 | 2020-06-12 | 西安交通大学 | Battery silk-screen quality detection method based on ORB feature matching and LK optical flow method |
| CN111487259A (en) * | 2020-04-24 | 2020-08-04 | 上海帆声图像科技有限公司 | Glass cover plate silk-screen appearance detection device and detection algorithm |
| CN111693549A (en) * | 2020-05-14 | 2020-09-22 | 西安电子科技大学 | Method for detecting and classifying defects of mobile phone cover plate glass |
| CN112330678A (en) * | 2021-01-07 | 2021-02-05 | 中科慧远视觉技术(北京)有限公司 | Product edge defect detection method |
-
2021
- 2021-03-24 CN CN202110312387.4A patent/CN113063731B/en active Active
Patent Citations (18)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2006284308A (en) * | 2005-03-31 | 2006-10-19 | Toshiba Corp | Visual examination method of semiconductor device |
| US8103085B1 (en) * | 2007-09-25 | 2012-01-24 | Cognex Corporation | System and method for detecting flaws in objects using machine vision |
| JP2010181254A (en) * | 2009-02-05 | 2010-08-19 | Ushio Inc | Device for inspecting screen print making |
| KR20100093213A (en) * | 2009-02-16 | 2010-08-25 | 유근홍 | System for inspecting defects on glass substrate using contrast value, and method of the same |
| CN105354844A (en) * | 2015-10-30 | 2016-02-24 | 清华大学 | Edge inflection point detection method and detection apparatus based on shape fitting |
| CN106127779A (en) * | 2016-06-29 | 2016-11-16 | 上海晨兴希姆通电子科技有限公司 | The defect inspection method of view-based access control model identification and system |
| CN108355987A (en) * | 2018-01-08 | 2018-08-03 | 西安交通大学 | A kind of screen printing of battery quality determining method based on piecemeal template matches |
| CN109064420A (en) * | 2018-07-13 | 2018-12-21 | 华南理工大学 | A kind of multi-layer PCB cross-sectional image technological parameter extracting method |
| CN109961464A (en) * | 2019-03-22 | 2019-07-02 | 武汉艾维视智能技术有限公司 | A kind of high-precision screen printing net plate scan image method for registering |
| CN110567965A (en) * | 2019-07-31 | 2019-12-13 | 华南理工大学 | Smartphone glass cover plate edge visual defect detection method |
| CN110567976A (en) * | 2019-08-30 | 2019-12-13 | 中国地质大学（武汉） | mobile phone cover plate silk-screen defect detection device and detection method based on machine vision |
| CN110672617A (en) * | 2019-09-14 | 2020-01-10 | 华南理工大学 | Method for detecting defects of silk-screen area of glass cover plate of smart phone based on machine vision |
| CN110909750A (en) * | 2019-11-14 | 2020-03-24 | 展讯通信（上海）有限公司 | Image difference detection method and device, storage medium and terminal |
| CN111275697A (en) * | 2020-02-10 | 2020-06-12 | 西安交通大学 | Battery silk-screen quality detection method based on ORB feature matching and LK optical flow method |
| CN111204116A (en) * | 2020-02-13 | 2020-05-29 | 深圳市和力泰智能制造有限公司 | Silk screen printing equipment |
| CN111487259A (en) * | 2020-04-24 | 2020-08-04 | 上海帆声图像科技有限公司 | Glass cover plate silk-screen appearance detection device and detection algorithm |
| CN111693549A (en) * | 2020-05-14 | 2020-09-22 | 西安电子科技大学 | Method for detecting and classifying defects of mobile phone cover plate glass |
| CN112330678A (en) * | 2021-01-07 | 2021-02-05 | 中科慧远视觉技术(北京)有限公司 | Product edge defect detection method |
Non-Patent Citations (4)
| Title |
|---|
| CHUANXIA JIAN ET AL.: "Automatic surface defect detection for mobile phone screen glass based on machine vision", 《APPLIED SOFT COMPUTING》 * |
| YONGFA LV ET AL.: "A Mobile Phone Screen Cover Glass Defect Detection MODEL Based on Small Samples Learning", 《2019 IEEE 4TH INTERNATIONAL CONFERENCE ON SIGNAL AND IMAGE PROCESSING》 * |
| 吴闯等: "基于轮廓修补和图像差分的手机玻璃盖板表面缺陷检测", 《软件工程》 * |
| 高如新等: "基于图像处理的手机屏幕缺陷检测系统研究", 《测控技术》 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113063731B (en) | 2023-01-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109100363B (en) | Method and system for distinguishing defects of attached foreign matters from dust | |
| TWI484161B (en) | Defect inspection system and filming device for defect inspection, image processing device for defect inspection, image processing program for defect inspection, recording media, and image processing method for defect inspection used therein | |
| JP4269005B2 (en) | Glass bottle inspection equipment | |
| US8073236B2 (en) | System and method for detecting blemishes on image sensor package | |
| CN109900711A (en) | Workpiece, defect detection method based on machine vision | |
| CN109856156A (en) | A kind of display panel tiny flaw determination method and device based on AOI | |
| CN110044405B (en) | Automatic automobile instrument detection device and method based on machine vision | |
| CN112033965B (en) | 3D arc surface defect detection method based on differential image analysis | |
| WO2017092427A1 (en) | Electronic element positioning method and apparatus | |
| CN105160652A (en) | Handset casing testing apparatus and method based on computer vision | |
| CN104683793B (en) | Detecting method for camera | |
| JPH04166751A (en) | Method and apparatus for inspecting defect in bottle and the like | |
| CN105812790B (en) | Method for evaluating verticality between photosensitive surface and optical axis of image sensor and optical test card | |
| JP5303405B2 (en) | Vehicle inspection device | |
| CN110567965A (en) | Smartphone glass cover plate edge visual defect detection method | |
| CN104103069B (en) | Image processing apparatus, image processing method and recording medium | |
| CN111524134B (en) | Method and device for detecting regularity of honeycomb products on production line | |
| CN113063731B (en) | Detection system and detection method for rotary disc type glass cover plate silk screen printing | |
| CN116091506B (en) | Machine vision defect quality inspection method based on YOLOV5 | |
| CN111583241B (en) | Mobile detection method and device for regularity of ultra-large area honeycomb products | |
| JP7362324B2 (en) | Inspection method, manufacturing method and inspection device for image display device | |
| TW201522949A (en) | Inspection method for image data | |
| TWI493177B (en) | Method of detecting defect on optical film with periodic structure and device thereof | |
| CN113128499A (en) | Vibration testing method of visual imaging device, computer device and storage medium | |
| CN113624457B (en) | Thin film uniformity detection system based on optical diffraction |
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 |