KR101732820B1 - Vision system for inspecting defects on surface of gravure printing roller - Google Patents

Vision system for inspecting defects on surface of gravure printing roller Download PDF

Info

Publication number
KR101732820B1
KR101732820B1 KR1020150173739A KR20150173739A KR101732820B1 KR 101732820 B1 KR101732820 B1 KR 101732820B1 KR 1020150173739 A KR1020150173739 A KR 1020150173739A KR 20150173739 A KR20150173739 A KR 20150173739A KR 101732820 B1 KR101732820 B1 KR 101732820B1
Authority
KR
South Korea
Prior art keywords
image
defect
pattern
roller
vision system
Prior art date
Application number
KR1020150173739A
Other languages
Korean (ko)
Inventor
영하 전
종문 이
상준 송
종협 김
Original Assignee
(주)제이 앤 엘 테크
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by (주)제이 앤 엘 테크 filed Critical (주)제이 앤 엘 테크
Priority to KR1020150173739A priority Critical patent/KR101732820B1/en
Application granted granted Critical
Publication of KR101732820B1 publication Critical patent/KR101732820B1/en

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
    • G01N21/84—Systems specially adapted for particular applications
    • G01N21/88—Investigating the presence of flaws or contamination
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B41—PRINTING; LINING MACHINES; TYPEWRITERS; STAMPS
    • B41F—PRINTING MACHINES OR PRESSES
    • B41F13/00—Common details of rotary presses or machines
    • B41F13/08—Cylinders
    • B41F13/10—Forme cylinders
    • B41F13/11—Gravure cylinders
    • 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
    • G01N21/952—Inspecting the exterior surface of cylindrical bodies or wires
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/0002—Inspection of images, e.g. flaw detection
    • G06T7/0004—Industrial image inspection
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
    • G01N2201/00—Features of devices classified in G01N21/00
    • G01N2201/12—Circuits of general importance; Signal processing
    • G01N2201/126—Microprocessor processing
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/10—Image acquisition modality
    • G06T2207/10004—Still image; Photographic image

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Pathology (AREA)
  • Health & Medical Sciences (AREA)
  • Quality & Reliability (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Theoretical Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)

Abstract

The objective of the present invention is to provide a vision system to automatically identify defect from a surface pattern of a gravure printing roller to display a defect. According to the present invention, to find a defect part of a margin part between patterns, a predetermined part of a gravure printing roller is captured and digitized to identify the pattern formed with a plurality of cells as one object, and the pattern processed as the one object is compared with a size of an image object by the defect of the margin part to find the defect of the margin part. Moreover, discovery of the defect included in the pattern of the gravure printing roller is determined by digitizing and storing an image of a reference of a perfect pattern image, digitizing and subtracting the image of an actual captured pattern from the reference pattern image, and determining the defect when pixels of which a difference value is equal or greater than a given threshold exist within a predetermined region over a predetermined number.

Description

[0001] VISION SYSTEM FOR INSPECTING DEFECTS ON SURFACE OF GRAVURE PRINTING ROLLER [0002]

The present invention relates to a vision system for inspecting surface defects of a gravure printing roller, and more particularly, to a vision system including an optical system and an algorithm capable of inspecting the entire surface area of the roller at a high resolution in a shorter time .

In recent years, research has been actively conducted to replace conventional electronic device manufacturing processes by applying conventional printing technology in the electronics industry. This printing electronic technology refers to a technique for manufacturing various electronic devices through a low-cost printing process using a functional electronic ink material capable of printing, and it is a low cost, large-sized, low-temperature / high- It is expected that it will be a new paradigm in the field of electronic devices and components by enabling simple and eco-friendly processes. In particular, it is advantageous in mass production and manufacturing cost reduction in MLCC, RFID tag, E-paper, solar cell, and smart sensor, and gravure printing ) Are attracting attention.

Gravure printing is a method in which a pattern is formed on the surface of a cylindrical cylinder of a cylinder and an ink is injected into the pattern, and then the pattern is transferred onto the surface of a continuous film-like object wound in a roll form. It has been used extensively in the field of printing for photographs, books, packaging materials and textiles, because it has advantages of high-speed large-area application to printing and excellent printing quality. Such gravure printing technology has recently been applied to a wide range of fields including IT and electronic industry beyond the existing printing application field due to its low production cost and excellent mass productivity.

MLCC is the most representative example of electronic devices to which gravure printing technology is applied. MLCC (Multi-Layer Ceramic Capacitors) are chips-type capacitors that are mounted on printed circuit boards of various electronic products such as mobile communication terminals, notebook computers, personal computers and personal portable terminals and play an important role in charging or discharging electricity. The dielectric layer and the metal electrode are alternately stacked, and the thickness of the dielectric layer and the total number of layers within the permissible volume serve as key design elements for the implementation of the high-capacity device. Like other electronic devices, MLCCs also follow the tendency to miniaturize with high capacity and high density, so that the printing pattern by gravure roller which is the base technology of MLCC production also requires high precision and reliability.

When the pattern itself formed on such a gravure printing roller has surface defects such as scratches, shots, and pinholes, a large number of defects are generated with respect to the parts produced by the roller. A plurality of fine cells are gathered to form one pattern and the patterns are arranged with a certain margin to form a predetermined pattern group as a whole. When a capacitor is formed by using a gravure printing roller, a plurality of substrates are printed using rollers having such a pattern, and then the substrates are overlapped, and the cutter is approached to the margin portion and cut and modularized to produce capacitors of desired capacities.

If there is a defect in the margin of the roller, the cutter recognizes this as a kind of pattern and does not perform the cutting operation, so that the whole process is interrupted and a large number of the entire boards are treated as defective. If the pattern itself is defective, the device itself is not normally fabricated and the defects increase again.

      To prevent this problem, before the printing, the inspector visually inspects the roller with a cylindrical vision showing an enlarged pattern of the pattern formed on the roller surface. This is a very hard work, and it takes a lot of time to inspect one roller. Therefore, there is a need for a new inspection system because the operation is slow and the worker's work fatigue is high and the defect may be missing.

Korean Patent Laid-Open No. 10-2004-0030312 describes a technique for vision inspection of a printed circuit board, but relates to an optical system.

SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an automatic inspection vision system capable of automatically recognizing and displaying defects on fine patterns formed on the surface of a gravure printing roller.

In addition, the present invention provides a vision application for displaying an image of a defect part so as to increase the resolution of the vision system to improve the accuracy and reliability of inspection, and to easily recognize a corresponding point when a defect part is found.

According to the above object, according to the present invention, a certain portion of a roller is photographed with an image sensor so that a pattern composed of a plurality of cells is recognized as a single entity in order to find a defect in a margin portion between the pattern and the pattern. The signal processing is performed and the size of the image object by the defect of the margin portion is compared with the pattern of the single object to find the defect of the margin portion.

Further, the defect found in the roller pattern itself can be detected by inputting an image of a standard pattern having no defect at all, inputting the image of the actually photographed pattern, and then processing the standard pattern image and the subtraction process, When a pixel having a value equal to or greater than a given threshold is present in a certain region, a defect is judged to exist. The standard pattern image may be directly registered by the user or may be generated by automatically recognizing an image of a standard pattern by averaging a plurality of patterns in the case of a pattern that is periodically repeated in one image.

Meanwhile, in the present invention, coaxial light is applied to the image sensor so that the light irradiating the roller can be clearly displayed in black so that the fine pattern and the margin defect can be photographed at a high resolution.

Further, the present invention linearly configures the image sensor to avoid image distortion due to the curvature of the roller and to take a large area at high resolution.

Also, in order to quickly process a large amount of image data, the present invention records and reads image data through a file processing technique that can use a file of a disk as a virtual memory.

According to another aspect of the present invention, there is provided a display module for displaying an image of a roller part on a monitor screen for each predetermined area, and providing an enlarged view of the selected area when the image is selected, When a specific portion of the enlarged view is clicked, an enlarged view with a higher magnification is provided,

Observation is convenient by separately providing an image of a part where a defect is found by the defect judgment module to a monitor.

According to the present invention, a pattern defect of the gravure printing roller can be found very easily and quickly and accurately.

That is, according to the present invention, a pattern defect of a gravure printing roller is automatically inspected by a signal processing without requiring a human to directly inspect the pattern defect, and the defect portion is displayed on the monitor, so that defect inspection can be carried out very conveniently and quickly.

Further, the inspection system of the present invention employs a linear image sensor to obtain a distortion-free image with respect to the curved surface of the roller.

In addition, the present invention increases the reliability by allowing coaxial light to be applied to the image sensor to cause the light irradiating the roller to be detected without missing a defect, which is formed concavely with respect to the plane.

Further, according to the present invention, the screen configuration shown on the monitor can facilitate the inspection work of the observer.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a device overview diagram showing a vision system for surface defect inspection of a printing roller of the present invention. FIG.
2 is a photograph showing an example of a configuration of a roller defect detection monitor screen by the vision system of the present invention.
3 is a schematic cross-sectional view illustrating a defect detection method by the coaxial illumination device of the vision system of the present invention.
4 is a flowchart illustrating a signal processing method for finding a margin between a roller pattern and a pattern in a vision system of the present invention and a defect portion inherent in the pattern itself.
5 to 8 are schematic diagrams for explaining a signal processing method for finding a margin between a roller pattern and a pattern in a vision system of the present invention and a defect portion inherent in the pattern itself.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a device schematic diagram showing a vision system for surface defect inspection of a gravure print roller 110 of the present invention.

The vision system of the present invention includes a rotation shaft 120 capable of rotating the apparatus main body 100 and the gravure printing roller 110, an optical system 140 capable of photographing the surface of the roller 110 from the rear surface of the roller 110, 140) in a direction orthogonal to the longitudinal direction and the longitudinal direction of the roller.

In addition, a defect determination module 201 for analyzing image data photographed by the optical system 140 to determine whether a pattern defect is present, a display design module 202 for easily browsing the monitoring result, And a file processing module 203 for preventing the speed from slowing down in computer processing (such as reading, writing, and recalling a file).

In addition, a monitor 300 is provided for easily observing an image photographed by the optical system 140. The monitor 300 may be installed in the vision system itself, but may be installed in a remote place at a distance from the vision system, or may be configured as a mobile terminal such as a tablet PC or a smart phone.

First, the apparatus main body 100 is schematically shown in the form of a plate, but a control box and an outer casing may be additionally mounted on the apparatus main body 100. It is also possible to provide a vibration damping device or the like capable of suppressing vibration from the outside.

The rotary shaft 120 is connected to the gravure print roller 110 on the apparatus main body 100. The rotary shaft 120 is provided on the apparatus main body 100 It rotates.

The linear stage 130 includes an X-axis stage 131 and a Y-axis stage 132. The linear stage 130 is mounted on the apparatus main body 100 and mounts the optical system 140 on the apparatus main body 100, ) Along the longitudinal direction and the longitudinal direction. The X-axis stage 131 is used when the linear image scanner 141 can not photograph the entire area of the roller at one time, so that the X-axis stage 131 can be photographed while moving at regular intervals in the longitudinal direction of the gravure printing roller 110. The Y-axis stage 132 is then used to focus the linear image scanner 141 and the frame image scanner 142.

The optical system 140 is disposed in the linear stage 130 and includes a linear image scanner 141 for photographing the entire area of the gravure roller and a frame image scanner 142 for magnifying and observing an area of interest such as a defect part . The linear image scanner 141 is a one-dimensional image sensor, and the frame image scanner 142 is a two-dimensional image sensor, and the frame image scanner 142 is relatively large in comparison with the linear image scanner 141 Is applied.

In addition, the present invention linearly configures the image scanner to avoid image distortion due to the curvature of the roller 110 and to take a large area at a high resolution. The linear image scanner 141 scans the periphery of the roller 110 and moves in the longitudinal direction of the roller by the field of view of the scanner and then scans the periphery of the roller 110 again, ) Is photographed. Accordingly, even if the entire surface of the roller 110 is large, an image free from distortion can be obtained with respect to the entire surface area. In addition, since the linear image scanner 141 generates a large image by synthesizing images acquired in units of lines, there is no limit to the image size, and it is possible to photograph the surface of the roller 110 at a high resolution. When the surface of the roller 110 is photographed at one time by the frame image scanner, the image distortion due to the curvature of the roller 110 occurs. However, the linear image scanner 141 of the present invention has the curvature of the roller 110 Scanning is performed for a width of about 1 mm, so that a highly reliable image can be obtained.

FIG. 2 shows an exemplary screen configuration of the roller 110 defect detection monitor 300 by the vision system of the present invention. A list menu including a defect list, an inspection record, a hardware and an end button for automatic inspection is displayed on the left side, an image of the roller 110 side is shown on the right side of the menu, and a defective image is enlarged in the center. Below it are detailed images of the defect, and on the far right is an image of the defect at the maximum magnification.

7 is a schematic diagram for explaining a signal processing method for finding a defect portion in a margin portion between a pattern of a roller 110 and a pattern in a vision system of the present invention.

The present invention relates to a defect determination module that processes a video signal acquired by a linear image scanner 141 and detects a defect in a margin portion between patterns of the roller 110 and a defect inherent in the pattern itself Respectively. In order to find a defect in a margin portion between a pattern and a pattern, a pattern composed of a plurality of cells is processed to be recognized as one object, and a pattern of one object and a size of an image object To identify defects in the margins. For this purpose, the image including the margin portion and the pattern was subjected to dilation treatment and erosion treatment.

When the expansion process is performed, the outline of the fine cells constituting the pattern is thickened to form one face, and the pattern is integrated with the neighboring cells so that the pattern itself becomes one face. When an erosion process is performed in the state of a single face, only the outline is left as an object consisting of the outline of the original pattern.

The marginal portion should be normal without any image, and if there is a defect, the defect appears as an independent entity by expansion and erosion treatment.

Since the patterns are originally formed in a uniform size, the size of the defect caused by the defects in the margin portion is much smaller than that of the outline of the pattern. Therefore, if the size of the pattern is much smaller than the average, or if a pattern having a small object size is found based on the size of the original pattern, they are judged to be defects of the margin portion.

In addition, the discovery of defects inherent in the gravure print roller 110 pattern itself can be accomplished by signaling an image of a standard pattern with no defect at all, signaling the image of the actually photographed pattern, And judged the defect as a defect when a pixel having a difference value equal to or greater than a given threshold value exists in a certain region in a certain number or more.

8 illustrates a signal processing method for finding defects inherent in the gravure print roller 110 itself in the vision system of the present invention.

The image acquiring process and the defect inspection method will be described in more detail as follows.

4 is a flowchart of a defect inspection method.

In the image acquisition, the linear image scanner 141 applies a method of stacking images photographed on a line-by-line basis and synthesizing the photographed images into an entire image. Therefore, the entire image of the periphery of the roller 110 can be photographed only by rotating the rollers. Accordingly, it is necessary to provide a rotating shaft 120 capable of rotating the roller 110 at a constant speed. Further, in order to prevent distortion of the photographed image, the photographing speed and the rotational speed of the roller 110 must be accurately synchronized. Accordingly, the rotation shaft 120 includes a high-resolution encoder and controls the camera and the illumination by using the pulse signal output from the encoder as feedback.

An image of the surface of the roller is obtained using the linear image scanner 141 (operation 400). The first image obtained is a gray level multivalued signal from 0 to 255. Since the light irradiating the roller 110 is photographed by applying coaxial light to the image sensor, the printed pattern formed at the engraved surface on the surface of the roller 110 and the defective portion having the depth are expressed darker than the plane.

When the image of the multilevel signal is binarized to a threshold value lower than the gray level of the plane portion, the plane portion larger than the threshold value is expressed by white, and the print pattern formed by the engraved surface on the surface of the roller 110, pinhole, Defective areas with the same depth are represented in black (step 410).

In the strict sense, binarization is to express the value as 0 or 1. However, if the pixel value of the image is expressed as 0 or 1, it is difficult to distinguish it from the view of the eye. Therefore, in the binarization of the image, the pixel value is converted into 0 or 255. Binarization of images is used in various image processing fields, and is especially used as a preprocessing process for locating a desired object in an image.

At this stage, however, patterns as well as defects are detected. Therefore, in order to detect only defects, it is necessary to exclude the region where the patterns are formed from the inspection region.

The present invention provides a method of generating a mask image of a pattern through signal processing for recognizing the pattern as one opening and inspecting only a region excluding the region where the pattern is formed using the generated mask image.

All the patterns formed on the surface of the roller 110 have a common feature that a plurality of fine cells are grouped adjacent to each other (see FIG. 5). If the signal is processed to recognize such a pattern as one object as described below, the defect maintains its original size, but the pattern becomes a single object in which a plurality of cells are combined to be relatively larger than the defect size. Here, a mask image is generated using only objects (patterns) having a size exceeding a predetermined threshold value.

 The steps 1 to 3 are performed to generate the mask image of the pattern unit (step 420).

1. Opening process removes small objects such as defects leaving a large size pattern. The open process goes through the process of erosion and expansion.

2. Closing process makes a pattern composed of adjacent cells into one object. The closure process goes through a process of expansion and erosion.

3. When the above process is performed, a mask image (see FIG. 6) in which only a pattern-formed region is displayed in black is generated.

A mask image of the pattern portion generated in the above is overlaid on the original image, and objects (black spots) exceeding a given threshold size are detected in an area indicated by white. Through the above process, a margin defect detection screen can be obtained as shown in FIG. At present, it is possible to detect an object having a minimum size of 14 mu m in the inspection device specification (step 430).

The discovery of defects inherent in the pattern itself of the roller 110 can be accomplished by detecting the exact position of the patterns with subpixel resolution through a position detection algorithm and by signaling and inputting an image of a standard pattern with no defect at all, After the image is signalized, the standard pattern image and the subtraction process are performed to determine that the defect is present when a pixel having a difference value equal to or larger than a given threshold exists in a certain region in a certain region (Step 440). 8 shows detection of a defect in a pattern portion by a subtraction operation.

The standard pattern image may be directly registered by the user, or may be generated by automatically recognizing an image of a standard pattern by averaging a plurality of patterns in the case of a pattern that is periodically repeated in one image.

Finally, the defects detected through the image processing are subjected to a step of mapping the position information obtained from the rotation axis 120 and the linear stage 130 to the defective image (operation 450).

Meanwhile, in the present invention, coaxial light is applied to the image scanner so that the light illuminating the roller 110 can be clearly displayed in black so that the pattern and margin defects can be photographed at a high resolution. That is, the illumination and image scanner are installed coaxially.

3 is a schematic diagram for explaining a defect detection method by the coaxial optical system 140 of the vision system of the present invention. Since the conventional optical system is incident on the image scanner with a light angle, the defective portion is incident on the image scanner from a defective portion such as a pinhole so that the defective portion appears bright. However, since the pinhole itself is not a square pit made up of vertical and horizontal, the image is very small, and it is often blurred or can not be captured. On the other hand, when coaxial light is used as in the present invention, light is scattered toward a side where there is no image sensor and appears as a black spot in a defective portion such as a pinhole. The size of such a black spot is so large that it is easy to find a defect and reliability is high because it is not omitted.

In addition, the present invention uses a telecentric lens to focus the light incident on the image sensor to prevent peripheral image distortion.

In addition, the present invention includes a file processing module 203 for preventing an image file from slowing down in computer processing (such as reading, writing, and reading a file) in establishing an interface with a vision system. That is, in order to quickly process a large amount of image data, image data is recorded and read using virtual memory. For example, assuming that the diameter of the roller 110 is 150 mm and the total width is 550 mm, the memory size of the image file obtained by scanning the roller 110 by one rotation with respect to the 14 mm section is approximately 500 MB, and X Axis, and if the entire section is scanned, the total size of the image file is about 22 GB. It takes over an hour to read and write files of this size on a regular PC. Therefore, the processing time is shortened by applying a part of the disk area to the virtual memory so that it can be processed like a physical memory (RAM). Virtual memory is implemented using the memory map file technique.

Windows uses the paging file (Paging) of the hard disk instead of memory when there is insufficient physical memory (RAM). The feature is the same as physical memory (RAM), which is a little slower with paging file. A memory map file is a technique for connecting the contents of a file existing on a hard disk to the address space of a process based on this theory. In short, it uses a file as memory.

On the other hand, the display design module 202 is provided so that the observer can view images of the gravure print roller 110 surface provided by the vision system of the present invention.

Providing the screen continuously with respect to the portion of the roller 110 that is photographed by the image sensor, and when a specific screen is selected, provides an enlarged view thereof, and when a specific portion of the enlarged view is clicked, Lt; / RTI > Since the roller 110 has a cylindrical shape, a cylindrical coordinate system can be used to specify the position of each image in terms of coordinates. In the case of an image for a defect, it is easy to determine at which point on the roller 110 a defect exists, which is specified in this coordinate system together with the serial number.

In addition, the system can additionally include a three-dimensional microscope capable of measuring the cell depth in the pattern, and a pen marking device capable of marking a defective area for inspection of non-conforming products after inspection.

In this manner, a vision system for automatically detecting defects in the gravure printing roller can be realized.

It is to be understood that the present invention is not limited to the above-described embodiment, but is defined by the scope of the claims, and those skilled in the art can make various changes and modifications within the scope of the claims It is self-evident.

100:
110: Gravure roller
120:
130: linear stage
131: X-axis stage
132: Y-axis stage
140: Optical system
141: Linear Image Scanner
141a: Lighting
141b: lens
141c: camera
142: Frame Image Scanner
201: Defect determination module
202: Display Design Module
203: File processing module
300: Monitor

Claims (9)

A vision system for automatically detecting and displaying surface defects of a gravure printing roller,
An image scanner for photographing a certain portion of the gravure print roller surface;
A defect determination module for analyzing image data of a photographed image scanner to detect whether a pattern defect is present; And
And a display module for displaying a photographed image and an image of a part where a defect is found, on a monitor,
Wherein a plurality of pattern groups are arranged on the surface of the gravure printing roller with a margin portion therebetween, the pattern group is a set of a plurality of patterns,
The defect determination module includes:
In the process of finding a defect part in a margin part between a pattern and a pattern, a pattern composed of one or more cells is signal-processed so as to be recognized as one object, and a pattern composed of one object and an image object The size of the margin is compared to find the defect of the margin,
In the process of finding defects inherent in the pattern itself, an image of a standard pattern with no defects is signaled and input, the image of the actually photographed pattern is signalized, and then the standard pattern image is subtracted and processed,
A light irradiating roller is provided coaxially with the image scanner so as to photograph a defect on the surface of the gravure printing roller, and coaxial light is irradiated so that the light incident on the defective portion is scattered toward the image sensor without black Wherein the reliability of defect detection is improved. The defect determination module of claim 1, wherein, in the process of searching for a defect inherent in the pattern itself, the defect determination module performs a subtraction process with respect to a standard pattern image, and when a pixel whose difference value is equal to or greater than a given threshold exists in a certain region, And judges that the defect is defective. The image scanner of claim 1, wherein the image scanner is a linear image scanner in which the image sensor is linearly arranged,
Wherein the linear image scanner rotates the gravure print roller and detects a defect on the roller surface. delete 2. The vision system of claim 1, wherein the lens that focuses light to the image scanner is a telecentric lens. The system of claim 1, wherein the vision system comprises an image data processing module,
Wherein the image data processing module sets a part of the disk area as a virtual memory, and records and reads the image data. The display module according to claim 1, wherein the display module displays an image of the roller portion displayed on the monitor screen for each predetermined region, and provides an enlarged view of the selected region when the image is selected for the specific region, If you click on a specific part of it,
And an image of a portion where a defect is found by the defect determination module is separately provided to the monitor. 2. The vision system of claim 1, wherein the gravure printing roller is rotated by an axis of rotation and comprises an encoder on the axis of rotation for synchronizing the rotation speed of the rotation axis with the imaging speed of the linear image scanner. 2. The image processing apparatus according to claim 1, wherein the standard pattern image is registered through a module that is directly input by the user or automatically recognizes and generates an image of a standard pattern by averaging a plurality of patterns that are repeated periodically Vision system.
KR1020150173739A 2015-12-08 2015-12-08 Vision system for inspecting defects on surface of gravure printing roller KR101732820B1 (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020150173739A KR101732820B1 (en) 2015-12-08 2015-12-08 Vision system for inspecting defects on surface of gravure printing roller

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020150173739A KR101732820B1 (en) 2015-12-08 2015-12-08 Vision system for inspecting defects on surface of gravure printing roller

Publications (1)

Publication Number Publication Date
KR101732820B1 true KR101732820B1 (en) 2017-05-08

Family

ID=60163823

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020150173739A KR101732820B1 (en) 2015-12-08 2015-12-08 Vision system for inspecting defects on surface of gravure printing roller

Country Status (1)

Country Link
KR (1) KR101732820B1 (en)

Cited By (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108481891A (en) * 2018-04-26 2018-09-04 深圳劲嘉集团股份有限公司 A kind of gravure printing roller detection device, detecting system and its detection method
KR101938095B1 (en) * 2017-10-18 2019-01-11 김희춘 Gravure Printing Engraving Roll having Having Vision System
KR101938995B1 (en) * 2017-10-18 2019-04-10 김희종 Vision System For Gravure Printing Engraving Roll
CN112630224A (en) * 2020-12-17 2021-04-09 黑龙江职业学院(黑龙江省经济管理干部学院) Visual workstation based on cell-phone screen flaw detects
CN114324400A (en) * 2021-12-21 2022-04-12 龙游运城压纹制版有限公司 Embossing roller surface defect detection device
CN115870237A (en) * 2022-12-19 2023-03-31 北京航空航天大学杭州创新研究院 System and method for detecting product quality of assembly line and eliminating defective products
CN117190965A (en) * 2023-11-03 2023-12-08 南通际维机电设备有限公司 Radial surface defect detection device of auto-parts in shaft hole formula
WO2023244500A1 (en) * 2022-06-16 2023-12-21 Schaeffler Technologies AG & Co. KG Method for defect detection for rolling elements

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002254590A (en) 2001-03-05 2002-09-11 Nyuurii Kk Inspection device for gravure cylinder
JP2006194702A (en) * 2005-01-12 2006-07-27 Think Laboratory Co Ltd Inspection device for gravure cylinder and inspection method using it
JP2008122097A (en) * 2006-11-08 2008-05-29 Hitachi Kokusai Electric Inc Inspection measuring device

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002254590A (en) 2001-03-05 2002-09-11 Nyuurii Kk Inspection device for gravure cylinder
JP2006194702A (en) * 2005-01-12 2006-07-27 Think Laboratory Co Ltd Inspection device for gravure cylinder and inspection method using it
JP2008122097A (en) * 2006-11-08 2008-05-29 Hitachi Kokusai Electric Inc Inspection measuring device

Cited By (12)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
KR101938095B1 (en) * 2017-10-18 2019-01-11 김희춘 Gravure Printing Engraving Roll having Having Vision System
KR101938995B1 (en) * 2017-10-18 2019-04-10 김희종 Vision System For Gravure Printing Engraving Roll
CN108481891A (en) * 2018-04-26 2018-09-04 深圳劲嘉集团股份有限公司 A kind of gravure printing roller detection device, detecting system and its detection method
CN108481891B (en) * 2018-04-26 2024-05-28 深圳劲嘉集团股份有限公司 Gravure plate roller detection equipment, gravure plate roller detection system and gravure plate roller detection method
CN112630224A (en) * 2020-12-17 2021-04-09 黑龙江职业学院(黑龙江省经济管理干部学院) Visual workstation based on cell-phone screen flaw detects
CN112630224B (en) * 2020-12-17 2021-09-24 黑龙江职业学院(黑龙江省经济管理干部学院) Visual workstation based on cell-phone screen flaw detects
CN114324400A (en) * 2021-12-21 2022-04-12 龙游运城压纹制版有限公司 Embossing roller surface defect detection device
CN114324400B (en) * 2021-12-21 2024-05-07 龙游运城压纹制版有限公司 Embossing plate roller surface defect detection device
WO2023244500A1 (en) * 2022-06-16 2023-12-21 Schaeffler Technologies AG & Co. KG Method for defect detection for rolling elements
CN115870237A (en) * 2022-12-19 2023-03-31 北京航空航天大学杭州创新研究院 System and method for detecting product quality of assembly line and eliminating defective products
CN117190965A (en) * 2023-11-03 2023-12-08 南通际维机电设备有限公司 Radial surface defect detection device of auto-parts in shaft hole formula
CN117190965B (en) * 2023-11-03 2024-01-23 南通际维机电设备有限公司 Radial surface defect detection device of auto-parts in shaft hole formula

Similar Documents

Publication Publication Date Title
KR101732820B1 (en) Vision system for inspecting defects on surface of gravure printing roller
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
TWI225927B (en) Defect inspection system
KR101687163B1 (en) Apparatus and method for inspection of touch panel
JP3132565B2 (en) Defect inspection method and apparatus
CN110136120B (en) Silk-screen printing sample plate size measuring method based on machine vision
CN112508826A (en) Printed matter defect detection method based on feature registration and gradient shape matching fusion
US20090213214A1 (en) Microscope System, Image Generating Method, and Program for Practising the Same
JPH0452992B2 (en)
JPS60219504A (en) Measuring device for height of circuit element on substrate
EP1277042B1 (en) Directional lighting and method to distinguish three dimensional information
JP2008175686A (en) Visual examination device and visual examination method
CN111861980A (en) Imaging detection method, electronic equipment and computer readable storage medium
CN104749801B (en) High Precision Automatic optical detecting method and system
JP2897890B2 (en) Printing accuracy measuring method and its device
CN110533660B (en) Method for detecting silk-screen defect of electronic product shell
KR20230096057A (en) Defect Layering Detection Method and System Based on Light Field Camera and Detection Production Line
Liu et al. Industrial Printing Image Defect Detection Using Multi‐Edge Feature Fusion Algorithm
Nardin et al. Automation of a series of cutaneous topography measurements from silicon rubber replicas
CN110838135A (en) PCB bare board image registration method
CN116818785A (en) Defect detection method, system and medium based on machine vision
CN115839957A (en) Method, device and equipment for detecting interlayer defect of display module and storage medium
Qian et al. Sub-Pixel defect detection for super large LCD images using quadrant gradient operator
CN115035071A (en) Visual detection method for black spot defect of PAD light guide plate
CN115222728A (en) Screen detection method and device based on visual technology and computer equipment

Legal Events

Date Code Title Description
E701 Decision to grant or registration of patent right
GRNT Written decision to grant