JP2010078485A - Method for inspecting printed matter - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a method for inspecting a printed matter, capable of accurately inspecting the printed matter, with less false-defect detections. <P>SOLUTION: The method for inspecting the printed matter, which inspects the printed matter having at least one printed surface on its substrate, includes: an irradiation step of having the printed matter irradiated with white light; a photographic step of photographing the printed matter, in synchronization with the conveyance of the printed matter or at a prescribed time interval; a non-inspection area setting step of setting a non-inspection area which is not an object to be inspected in image data of the printed matter obtained in the photographing step; and an image-processing defect determining step of determining a defect present in the printed matter by using the image data. The method is such that the non-inspection area is set to a portion having a luminance difference in the image data, and the size of the non-inspection area is adjusted according to the magnitude of the brightness difference, in the non-inspection area setting step. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a printed matter defect inspection method.

  An example of a printing machine is a gravure printing machine. Hereinafter, the case of printing a color print with a gravure printing machine will be described as an example.

  In general, in a gravure printing machine, a belt-like printed material obtained by printing on an original fabric is conveyed, an image of the printed material is imaged by an imaging means such as a line sensor camera included in an inspection apparatus, and image data obtained at that time is obtained. Based on this, visual inspection is conducted. In such a printed matter inspection apparatus, a color camera is mainly used in order to inspect a color printed matter with high accuracy.

  A general color camera forms a color image by combining images captured by three types of light receiving elements such as red (hereinafter R), green (hereinafter G), and blue (hereinafter B). The color line sensor camera has a three-line system in which light receiving elements having sensitivity to R, G, and B wavelengths are arranged in parallel, and a light beam received by arranging a spectroscopic means such as a prism inside the camera. Are divided into R, G, and B wavelengths, and two types of methods are used, namely a 3CCD system (or a 3-plate system) in which light receiving elements are arranged according to the decomposed wavelengths.

  Moreover, in a gravure printing machine, not only paper but also a plurality of types of original fabrics including a special product such as a transparent plastic film and aluminum vapor deposition are printed by the same printer. For this reason, in the conventional illumination device for a printed matter inspection device, a necessary light source is selected from light sources such as irregular reflection, regular reflection, and transmission that the illumination device has every time the original of the printed matter to be inspected is switched. It was selected, and appropriate light intensity adjustment was performed to support inspection of various printed materials.

As a method for picking up an image of a printed matter with a CCD camera or the like and inspecting the printed matter using image processing, for example, a general technique for detecting defects proposed in Patent Document 1 is a pattern matching method. In this method, an image having no defect in a printed material in one cycle is set as a reference image, and an image to be inspected is set as an inspection image, and these two images are compared to classify defects. Specifically, the comparison process is performed by taking a difference between pixels at the same position in the image acquired by the camera, and determining a defect when the difference value exceeds the threshold.

  However, in this method, due to variations in the printing process, the positions of the two images reflected on the camera are relatively fluctuated, and the difference value of the brightness value exceeds the threshold in a portion where there is no defect in a simple comparison process, and it is determined as a defect. May be. This is a pseudo defect, which is overdetection as an inspection.

When a defect occurs during printing, it is necessary to visually determine whether it is a true defect or a pseudo defect. However, if there are many false, this work load increases. Therefore, the operator sets the inspection conditions sweetly and operates. If the inspection conditions are made sweeter, the detection power is reduced as a result.
Japanese Patent No. 3808937

The present invention has been devised in view of the above problems, and an object thereof is to provide an inspection method for suppressing the occurrence of pseudo defects.

  The invention according to claim 1 of the present invention is a method for inspecting a printed material on which at least one surface of a substrate is printed, an illumination step of irradiating the printed material with white light, conveyance of the printed material, An imaging stage for capturing images of the printed matter in synchronization or at a predetermined time interval, and a non-inspection area setting stage for setting a non-inspection area not to be inspected in the image data of the printed matter obtained in the imaging stage, An image processing defect determination step for determining defects present in the printed matter using the image data, and in the non-inspection area setting step, the non-inspection area is located at a portion having a luminance difference in the image data. The printed matter inspection method is characterized in that the size of the non-inspection area that is set is adjusted according to the brightness difference.

  According to the present invention, it is possible to adjust the size of a non-inspection area to be set in an area where a pseudo defect is likely to occur in a printed material by a luminance difference.

  Hereinafter, a printed matter inspection method according to the present invention will be described with reference to the drawings. FIG. 1 is a schematic configuration diagram showing a printed matter inspection apparatus.

  In FIG. 1, it is assumed that the printing machine is transporting the printed material 10 at a predetermined speed. The printed matter inspection apparatus is synchronized with the conveyance speed of the printing press and captures the imaging unit 30 for photographing the surface of the printed matter 10, the reflective illumination means 20 for irradiating the surface of the printed matter 10 with white light, and the back surface of the printed matter 10. Control / image for automatic determination by extracting defective portions present in the printed matter 10 using image data obtained by photographing the surface of the printed matter 10 by the transmission illumination means 21 for irradiating white light and the imaging means 30 And processing means 40.

  The printed material 10 moves in the imaged area at a predetermined speed. At this time, a signal for each unit distance is obtained from a unit that measures the amount of movement of the printed matter 10 attached to the printing machine with high accuracy, and the signal may be divided and distributed to the control / image processing means 40 in some cases. By sending a signal, scanning imaging is performed so as not to be affected by the speed fluctuation of the printing press. When the conveyance speed of the printing press can be regarded as constant within the range of the resolution of the image pickup means 30, a method of obtaining an image only by starting image pickup by a trigger signal and image pickup at a constant time interval is also conceivable. However, it is always more reliable to take images that are synchronized with the conveyance speed of the printing press.

  In addition to the speed fluctuation of the printing press, there is a possibility that the original fabric of the printed material 10 is an original fabric that tends to expand and contract, such as a plastic film. When the original fabric of the printed matter 10 is expanded or contracted, there is a possibility that the image will be affected even if imaging is performed in synchronization with the speed of the printing press. Therefore, in addition to the conveyance speed of the printing press, measurement in consideration of the expansion and contraction of the printed material 10 is necessary. Generally, the influence of expansion and contraction is often reduced by adopting a measurement method in which a part of the printed product 10 is brought into contact.

  FIG. 2 is a schematic diagram showing the reflection illumination means 20 and the transmission illumination means 21 of the printed matter inspection apparatus.

An imaging lens 32 is attached to the imaging means 30, and the printed material 10 to be imaged is arranged in the vertical direction. In the gravure printing machine, since the printed material 10 is manufactured in a roll shape, the conveyance speed is often a constant speed. In this case, since the imaging target always passes under the imaging means 30, the imaging means 30 mainly selects a line sensor in many cases, but the area depends on the type of printing press, particularly the form of conveyance. You may select a camera. Further, although the image pickup means 30 is arranged perpendicular to the printed material 10, in the case of a line sensor camera, an illumination system that can obtain an appropriate image can be realized, and each pixel is arranged in the horizontal direction of the line sensor. As long as it is possible to pick up images of the printed matter 10 at the same distance, the θ in FIG. 2 may be inclined (other than 90 °).

  The reflective illumination unit 20 includes a first illumination unit 22 that is disposed between the printed material 10 and the imaging unit 30 and that irradiates the surface of the printed material 10 with white light, and a second illumination unit 23 that irradiates light other than white light. ing. Any type of the reflective illumination means 20 may be used as long as the amount of light capable of obtaining an image that can be appropriately processed can be ensured. However, when a line sensor is used for the imaging means 30, a linear shape is used. An illumination system capable of irradiation is suitable. Specifically, a fluorescent lamp, a transmission light, an LED illumination or the like is selectively used.

  Further, the reflection illumination means 20 may be arranged to select one of them depending on the original fabric of the printed matter 10 while arranging irregular reflection, regular reflection, or both, or both, but the type of printing machine (printing method) Any arrangement may be selected depending on the material. Further, in order for the imaging unit 30 to receive an optimal amount of light, two or more first illumination units 22 that emit white light and two or more second illumination units 23 that emit light other than white light may be arranged. . In addition, when a fluorescent lamp is employed for the reflective illumination unit 20, a reflective member may be disposed around the reflective illumination unit 20 to increase the amount of light.

  The reflective illumination means 20 is provided with an imaging slit 31. Specifically, as long as the amount of light received by the imaging unit 30 is not affected, it may be a space or a transparent member such as transparent acrylic.

  The transmitted illumination means 21 is arranged between the imaging means 30 and the transmitted illumination means 21 in such a positional relationship that the printed product 10 is arranged, and a third illumination unit 24 that irradiates the back surface of the printed product 10 with white light. I have. Any kind of transmitted illumination means 21 may be used as long as the amount of light capable of obtaining an image that can be appropriately processed can be secured. However, when a line sensor is used for the image pickup means 30, a linear shape is used. An illumination system capable of irradiation is suitable. Specifically, a fluorescent lamp, a transmission light, an LED illumination or the like is selectively used. In addition, as a method for ensuring the maximum light amount, the transmitted illumination unit 20 can be arranged linearly with respect to the imaging unit 30. It is not necessary to have a typical arrangement.

  In addition, the transmission illumination means 21 may not be used when using a raw material with particularly low transparency depending on the type of printing press (printing method) and the raw material. However, it is suitable that the transmission illumination means 21 can be switched between ON and OFF, for example, when it is attached to a printing machine in which the original fabric is to be conveyed. Further, two or more third illumination means 24 for irradiating white light may be disposed in order to cause the imaging means 30 to receive an optimal amount of light. In addition, when a fluorescent lamp is used for the transmission illumination unit 20, a reflection member may be disposed around the transmission illumination unit 20 to increase the amount of light.

  When the color printed matter 10 is imaged with a monochrome camera, there is a case where there is no difference in output gradation values even when different colors are combined. On the other hand, in the case of a color camera, light is received by three types of elements having sensitivity to R, G, and B, respectively. Since the information of the printed matter 10 can be reflected, it is more suitable for imaging the colored printed matter 10 than a monochrome camera.

Image capture by a camera that is the imaging means 30 will be described. Images are captured by the camera in synchronization with the printing press. Therefore, if the printing speed changes, the image resolution also changes in the flow direction according to the capture time. Next, since there is a difference in brightness due to illumination within the capturing field, shading correction is performed. Thereafter, an image subjected to preprocessing such as a noise removal filter is acquired.

  In the printing process, a predetermined tension is applied to the printed matter 10 and control is performed so that the printed pattern always enters the same position. However, although it is a slight amount, meandering occurs in the width direction, and expansion and contraction occur due to the aforementioned tension in the flow direction. Therefore, in the image captured by the camera, the position of the printed pattern changes little by little in the width direction and the flow direction.

  In the inspection method using pattern matching due to the above-described positional deviation, overdetection occurs. Currently, as a method for minimizing the influence of this positional deviation, a method of setting a reference image as an image immediately before an inspection image and sequentially updating the image has been devised. When more precise position correction is required, a characteristic part in the image may be extracted, and the position of the printed pattern in the image may be finely adjusted in the width direction and the flow direction.

  Although these means can cope with variations in the position of the printed pattern, there are other causes of overdetection. That is, the print quality of the printed material varies for each printing cycle. This often occurs in places where there is a clear difference in shading, such as the edge of a printed pattern.

  FIGS. 3A to 3C schematically show examples of edge portions of a printed pattern, and in each case, a lightly colored portion 51 on the left side and a darkly colored portion 52 on the right side are those near the center. This shows a case where there is a boundary portion (edge portion) 50. If both the edges of the pattern of the reference image and the pattern of the inspection image are smooth as shown in FIG. 3A, no pseudo defect occurs when the pattern matching process is performed.

  However, it is difficult to completely smooth the edge portion of the printed pattern, and generally, the edge portion is uneven as shown in FIGS. If the unevenness of the edge portion is exactly the same in the pattern of the reference image and the pattern of the inspection image, no pseudo defect occurs when the pattern matching process is performed. However, the unevenness of the edge portion is generally different between printed materials having the same pattern as shown schematically in FIGS. 3B and 3C, for example. This is the print quality of the printed material. This is a variation.

  When pattern matching processing is performed using FIGS. 3B and 3C as a reference image and an inspection image, respectively, a large number of pseudo defects are generated at the edge portion.

  One of the methods for solving the above problem is to set a non-inspection area as a certain area in the vicinity of the edge portion of a pattern having a large difference in shading. The non-inspection area is usually set manually. In the area set as the non-inspection area, the inspection process is not performed at all, so that a defect actually generated in that portion is also overlooked. Therefore, it is preferable to keep the non-inspection area as small as possible.

  However, in manual area setting, it is best to make a uniform setting such that the area of the predetermined width on both sides of the edge is set as the non-inspection area. The time-consuming and time-consuming work of setting the size is difficult. In the present invention, the position and size of the non-inspection area can be set in consideration of the characteristics of the pattern to be inspected.

For comparison, a conventional processing flow is shown in FIG. In the conventional processing flow, the printed matter is imaged (S2), and the obtained inspection image is subjected to shading correction, preprocessing, and position correction (S3, S4), and then an edge portion is extracted and the vicinity of the edge portion is extracted. The predetermined number of pixels is set as a non-inspection area (S5). It is assumed that the number of pixels serving as the non-inspection area can be set as a parameter. Thereafter, pattern matching processing is performed to create a difference image between the reference image and the inspection image (S6). Further, the difference image is binarized with a predetermined threshold value to detect a defective portion, and the quality is determined (S7).

  In such a conventional non-inspection area setting method, as described above, the size of the non-inspection area is set regardless of variations in print quality. Therefore, a large non-inspection area is set even for a printed matter with small variations, and as a result, a defect that can be detected cannot be detected if the size of the non-inspection area is appropriate.

  The processing flow of the method of the present invention is shown in FIG. Also in the method of the present invention, the same processing as before is performed until the printed matter is imaged (S12), and the obtained inspection image is subjected to shading correction, preprocessing, and position correction (S13, S14).

  Thereafter, in the x direction of the inspection image (perpendicular to the flow direction of the printed material), a difference in luminance value between adjacent pixels is taken (S15), and the difference value exceeds a predetermined threshold value. In this case, the pixels before and after that are set as non-inspection areas (S16). The same process is performed for the y direction (flow direction of the printed material) of the inspection image. Note that there is no particular order in which order of processing in the x direction or y direction is performed first. Further, depending on the pattern to be inspected, this process may not be performed in both the x direction and the y direction (for example, when the printed pattern is a vertical stripe, if this process is performed only in the x direction, the process is not performed in the non-inspection area. Can be set).

  The subsequent processing flow is the same as the conventional one, and pattern matching processing is performed to create a difference image between the reference image and the inspection image (S17), and the difference image is binarized with a predetermined threshold value to detect defects. A certain point is detected and the quality is judged (S18).

  In the non-inspection area setting method of the present invention, the position and size of the non-inspection area can be set in accordance with the uneven shape of the edge portion in the inspection image. Therefore, even if there are irregularities in the edge portion, only the very vicinity of the edge portion is set as a non-inspection area, and other portions are inspection target areas, and the possibility of missing a defect can be reduced.

  Furthermore, in the above-described non-inspection area setting method of the present invention, a portion where the difference in luminance value is small on the inspection image, that is, a portion where the difference in shading is originally small on the printed material does not become a non-inspection area but becomes an inspection target area. The size of the non-inspection area can be reduced.

  6A and 6B are schematic views for explaining this. 6A and 6B both show eight pixels arranged in a horizontal row in the region including the edge portion of the pattern in the inspection image (G1 to G8 in FIG. 6A and H1 in FIG. 6B). -H8) are schematically shown. In any case, it is assumed that a portion with high brightness (bright) is on the left side and a portion with low brightness (dark) is on the right side. The brightness of the bright part is the same in (A) and (B), but the brightness of the dark part is lower in (A), and the difference in brightness between the bright part and the dark part is larger in (A). Shall. More specifically, in (A), when the luminance difference is (A), the difference value of the luminance value at each stage is a threshold value in four stages (G1, G2, G3), G4, G5, and (G6, G7, G8). This is the case. In (B), there are two stages (H1, H2, H3, H4) and (H5, H6, H7, H8), and the difference value of the luminance value at that stage exceeds the threshold value.

  In the example of FIG. 6, if a portion having a brightness difference is simply set as a non-inspection area, 4 pixels in (A) and 2 pixels in (B) are set as non-inspection areas. Will be. In this way, the size of the inspection target area can be adjusted according to the density of the printed pattern.

  In a portion where the density difference is large, such as an edge portion or a barcode printing portion in the printed pattern, the density gradient also varies due to variations in print quality, and the difference value in the difference image becomes larger. On the other hand, in the case of an edge portion having a small density gradient, the print quality variation has little influence on the density gradient variation, and the difference value in the difference image is also small.

  Therefore, by setting the non-inspection area large in the former case and setting the non-inspection area small in the latter case, pseudo defects are less likely to occur and the detection power can be increased. The printed matter can be inspected with high accuracy.

The schematic block diagram which shows the principal part structure of the inspection apparatus which concerns on this invention. Cross-sectional explanatory drawing which shows the imaging and illumination means of this invention. The figure which showed the example of the edge part of a printing pattern typically. The flowchart which shows the whole schematic operation | movement of the conventional inspection apparatus. The flowchart which shows the whole schematic operation | movement of the test | inspection apparatus of this invention. Schematic diagram explaining changing the size of the non-inspection area depending on the difference in brightness

Explanation of symbols

10 .. Printed material 20 .. Reflective illumination means 21 .. Transmitted illumination means 22... First illumination means 23 .. 2nd illumination means 24 .. 3rd illumination means 30. The slit 32 of the imaging lens 40, the control, the image processing means 50, the edge portion 51, the light-colored portion 52, the dark portion

Claims (1)

  A method for inspecting a printed material on which at least one surface of a substrate is printed, the illumination step of irradiating the printed material with white light, and synchronizing with conveyance of the printed material or at predetermined time intervals, An imaging stage for imaging a printed matter, a non-inspection area setting stage for setting a non-inspection area not to be inspected in the image data of the printed matter obtained in the imaging stage, and the printed matter using the image data An image processing defect determination step for determining an existing defect, and in the non-inspection area setting step, the non-inspection area is set at a location having a luminance difference in the image data, and the size of the non-inspection area is set. A method for inspecting printed matter, wherein the printed matter is adjusted according to the magnitude of the luminance difference.

Images