JPH1148456A - Plain region selecting method and apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To input a part low in density difference as an inspection image from the whole of the pattern on printed matter. SOLUTION: This plain region selecting apparatus selects a plain region from the whole of the pattern on printed matter and is adapted when a printing state is inspected on the basis of an inspection image obtained by inputting the plain region as an inspection point by the partial input camera of an inspection apparatus. In this case, an input part 10 inputting the whole of a printed pattern, a delivery part 12 delivering a partial image from a whole image, a calculation part 14 calculating an evaluation value showing the flatness of brightness with respect to each partial image, a judge part 22 selecting the plain region on an image from the partial image based on the evaluation value to set the same to the inspection point and the output part 24 outputting the position data relaxed to the inspection point to the inspection apparatus are provided.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for selecting a solid area, and more particularly to an inspection image in which a solid area is selected from the whole picture on a printed matter, and the solid area is image-input by a partial input camera of the inspection apparatus. The present invention relates to a method and an apparatus for selecting a plain area, which are preferably applied when inspecting a printing state based on a print condition.

[0002]

2. Description of the Related Art A running web is continuously printed by a rotary printing press, and a pattern printed on the web is optically input by a camera to inspect a printing state based on the input image. Is being done. Specifically, for example, in the case of an automatic inspection device, occurrence of a defect is determined by performing predetermined image processing on an input image.

As described above, when inspecting a defect generated on a printed matter by performing image processing on an input image, if the defect to be detected is minute, such as a doctor streak, the defect is a target. In some cases, a high-resolution inspection is performed by inputting an image while limiting a certain range in a picture. As described above, when an inspection is performed based on an image obtained by partially enlarging and inputting a printed material, there are the following methods for determining a location (inspection point) on the printed material to be input.

Here, for the sake of convenience, a case where six inspection points are set in the width direction with respect to the entire pattern corresponding to one cycle of the plate cylinder corresponding to the printing cycle will be described. (1) FIG. However, there is a method of determining a specific fixed position not only for the head but also for the entire picture, and (2) a method of determining a random position for the entire picture as shown in FIG.

[0005]

However, in the conventional determination methods such as (1) and (2) above, since individual pictures existing in the inspection target are not considered, the input is made from the determined inspection point. There is a high possibility that a part of the picture will be included in the inspection image. Therefore, when image processing is performed on such an inspection image in order to detect a minute defect, there is a problem that a pattern included in the inspection image causes erroneous detection.

In order to avoid this problem, if a process for preventing erroneous detection such as removing a picture portion from an input inspection image is performed, a signal corresponding to a defect may be removed. On the contrary, there is also a problem that this prevention processing causes a minute defect to be missed.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and has been made so that a picture portion included in an input inspection image does not cause erroneous detection of a fine defect. It is an object of the present invention to provide a method and an apparatus for selecting a plain area, which can input a portion having a small density difference as an inspection image from the following.

[0008]

According to the present invention, a plain area is selected from the whole picture on a printed matter, and the plain area is used as an inspection point based on an inspection image input by a partial input camera of the inspection apparatus. A plain area selection method applied when inspecting a printing state by inputting an entire image including an entire printed pattern, and determining a size corresponding to an inspection image from the input entire image. As a unit, a step of cutting out a partial image, a step of calculating an evaluation value representing a flatness of luminance for each of the cut out partial images, and a step of flattening the luminance from among the partial images based on the calculated evaluation value Selecting a high-potential area as a solid area on the image and determining a predetermined number as inspection points on the image; And outputting the information to the inspection apparatus, by having, it is obtained by solving the above problems.

That is, in the present invention, the whole image obtained by inputting the whole picture to be inspected is divided into partial images of a predetermined size, and the flatness of each partial image is evaluated. Is determined as an inspection point, and the position information of the inspection point is output to the inspection apparatus, so that a plain region having a minimum density difference on a printed matter can be selected and an image can be input. Here, the plain region means a region in which there is almost no difference in density on the printed matter or as close as possible.

According to the present invention, in the plain area selecting method, the evaluation value representing the flatness of the luminance is a maximum frequency of the luminance distribution for all pixels included in the partial image, a variance of the luminance distribution, and the luminance distribution. Is at least one of the luminance values that are divided into a predetermined ratio biased toward the high luminance side or the low luminance side.

According to the present invention, in the solid area selecting method, the partial image is sequentially cut out over the entire image while shifting the partial image by at least one pixel in the horizontal direction at a time. Things.

According to the present invention, a plain area is selected from the entire picture on a printed matter, and the printing state is inspected based on an inspection image input by a partial input camera of the inspection apparatus, using the plain area as an inspection point. Means for inputting an entire image including the entire printed pattern, and a partial image in units of a size corresponding to an inspection image from the input entire image. Means for cutting out, for each cut-out partial image, means for calculating an evaluation value representing the flatness of brightness, and for each of the partial images, an image having high flatness of brightness is imaged based on the calculated evaluation value. Select it as the solid area above,
The above-mentioned problem is similarly solved by providing a means for determining a predetermined number as an inspection point on an image and a means for outputting position information on the determined inspection point to the inspection device.

According to the present invention, in the plain region selecting apparatus, the evaluation value representing the flatness of the luminance is a maximum frequency of the luminance distribution for all pixels included in the partial image, a variance of the luminance distribution, and the luminance distribution. Is at least one of the luminance values that are divided into a predetermined ratio biased toward the high luminance side or the low luminance side.

According to the present invention, in the plain area selecting apparatus, the partial image is sequentially cut out over the entire image while shifting at least one pixel at a time in the horizontal direction on the entire image. Things.

[0015]

Embodiments of the present invention will be described below in detail with reference to the drawings.

FIG. 1 is a block diagram showing a schematic configuration of a solid area selection device according to a first embodiment of the present invention.

The apparatus for selecting a plain area according to the present embodiment inspects a pattern (printed matter) printed on a running raw material,
A part input camera that can be moved in the width direction of the raw material of an inspection device (not shown), which is applied when an image of a plain region is input from the entire pattern and the plain region is selected to inspect the printing state. It is.

The solid area selection device includes a whole image input unit 10 for inputting a whole image including the entire printed picture, a whole image storage unit 12 for storing data of the inputted whole image, and a memory for storing the whole image. A partial image cutout unit 1 that cuts out a partial image from the entire image read from the unit 12 in units of a size corresponding to an inspection image in accordance with a method described later in detail.
4, a parameter storage unit 16 for storing parameters such as the size of a partial image used for processing in the cutout unit 14 and the number of partial images to be selected as inspection points in a determination unit 22, which will be described later. An evaluation value calculator 18 that calculates an evaluation value representing the flatness of luminance for each partial image cut out by the cutout unit 14, and an evaluation value that stores the evaluation value of each partial image calculated by the calculation unit 18 Storage unit 20
Based on the evaluation value read out from the storage unit 20, one of the partial images having a high luminance flatness is selected as a plain region on the image, and is input from the parameter storage unit 16. A determination unit 22 that determines the number of inspection points on an image as many as the number of images, and an output unit 24 that outputs position (address) information of the partial image determined as the inspection point to an inspection device (not shown).

When position information relating to the inspection point determined by the determination unit 22 is input from the output unit 24 to the inspection apparatus, the inspection apparatus assigns the partial input camera to the inspection point on the printed material corresponding to the inspection point. (Not shown), and an inspection image is input from the inspection point by the camera.

The solid area selecting apparatus of the present embodiment will be described in detail. The whole image input unit 10 can input an image of a normal printed matter in advance even if it is an image pickup apparatus such as a CCD camera installed on a printing line. It may be a storage device such as a stored database.

As shown conceptually in FIG. 2, the partial image cutout unit 14 uses a size corresponding to a range on a printed material to be actually inspected from the whole image input by the whole image input unit 10 as a unit. A process of extracting (cutting out) a partial image to be performed is performed. As described above, the size of the partial image is input from the parameter storage unit 16 as a separately set parameter.

As a specific extraction method, for example, as conceptually shown in FIG. 3, a partial image PG2 in which the entire image is shifted rightward by one pixel from the partial image PG1 at the upper left end of the entire image. Then, while sequentially shifting in the horizontal direction by one pixel, the image is moved and cut out in the direction indicated by the arrow, and finally cut out to the final partial image PGn at the lower right end. At that time, the first-stage extraction for the entire width in the horizontal direction is completed, and the amount of vertical movement when moving to the next-stage extraction position is arbitrary as long as there is no pixel in the vertical direction.

The evaluation value calculating section 18 calculates an evaluation value representing the flatness of the luminance for all the partial images included in the whole image, which are extracted by the partial image extracting section 14 by, for example, the method shown in FIG. I do.

In the calculation section 18, (1) the maximum frequency of the luminance distribution (histogram) for all pixels included in the partial image, (2) the variance of the luminance distribution, and (3) the evaluation values Any one of the luminance values (hereinafter, also referred to as partial division luminance values) that divide the luminance distribution into a predetermined ratio biased toward the high luminance side or the low luminance side is adopted.

The above (1) maximum frequency of the luminance distribution is shown in FIG.
As shown in FIG. 2B, the luminance distribution of all pixels included in one partial image conceptually shown in FIG. 2A is a pixel having the same luminance value (gradation value) in the partial image. Is the maximum value in the histogram with respect to the number (frequency).

In the present embodiment, the maximum frequency is calculated for all the partial images in the whole image and stored in the evaluation value storage unit 20, and the judgment unit 22 determines that the larger the maximum frequency is, the larger the printed matter is. In the above, it is determined that the image is close to a solid portion having a small shading (a small difference in density), that is, a plain region. In addition, when adopting the maximum frequency in this way,
In consideration of the influence of noise mixed into the entire image, the evaluation value may be a value obtained by summing up the respective frequencies of the pixel values included in an arbitrary width before and after the pixel value having the maximum frequency as a center.

When such a maximum frequency is used as the evaluation value, compared with the case of using the variance described below,
Even when the calculation process is fast and there are almost no solid portions, or when there are a plurality of peaks in the luminance distribution, there is an advantage that a region closer to solid can be selected.

The variance of the luminance distribution in the above (2) is calculated by the following equation (1).

[0029]

(Equation 1)

When the above variance σ is adopted as the evaluation value, this value is obtained for all the partial images cut out from the entire image and stored in the storage unit 20 in the same manner. Then, the determining unit 22 determines that the smaller the variance is, the closer to the solid portion on the printed matter corresponding to the partial image.

Further, as shown in FIG. 5, the partial split luminance value of the partial distribution (3) conceptually shows an example in which the luminance distribution of the i-th partial image is biased toward the high luminance side. 80% on the high luminance side and 20% on the low luminance side
It is obtained as a luminance value (pixel value) Pi that can be divided such that

This partial pixel value is calculated for all partial images, stored in the storage unit 20 in the same manner, and used for determination by the determination unit 22. The larger the partial pixel value Pi, the brighter the area. Also, this partial division pixel value P
i has such a feature that the larger the value is, the higher the possibility of becoming a solid portion is.

By changing the ratio of dividing the luminance distribution, it is possible to change the evaluation of the flatness and to change the direction of the partial division, for example, to the above-mentioned 20%:
By changing 80% to 80%: 20%, a solid portion can be selected in a similar manner for a dark portion.

Further, the determination unit 22 includes the storage unit 20
Is read out, and based on the evaluation values, the inspection points are selected by the preset number inputted from the parameter storage unit 16. At this time, if a plurality of inspection points are selected in an order that the evaluation points are considered to be closer to the solid portion only from the evaluation values, the inspection points may be concentrated in a certain narrow range. Therefore, in order to prevent this, using the position information of each partial image, if the evaluation values have the same level and the areas on the images overlap, one of them is excluded and the inspection point is selected. You may do so.

In the output section 24, as described above,
The position information of the inspection point on the image selected by the determination unit 22 is output to an inspection device (not shown).

In this embodiment, a plain area is selected in accordance with the flowchart shown in FIG.

First, the whole image input section 10 obtains (inputs) a whole image including the whole picture on a printed material in a normal printing state, and stores it in the whole image storage section 12 (step 1). .

The whole image can be obtained by directly inputting the image with the whole input camera (not shown) at the time when the printing is stabilized after the start of printing, or when the whole image data is already stored for the same picture. For example, it may be input from a database or the like.

Next, in the partial image cutout section 14,
From the entire image read out from the storage unit 12, FIG.
The partial image is cut out as shown in (2) (step 2).

Next, an evaluation value is calculated for the first partial image cut out (step 3), and the same calculation is sequentially performed for all partial images (step 4).

Next, the judgment unit 22 judges (selects) a predetermined number of inspection points on the image from those having high flatness using the evaluation values for all the partial images.
Then, the position information of each inspection point is output to the inspection device.

When the position information of each inspection point on the image is input to the inspection device, the inspection device moves the partial input camera to a widthwise position corresponding to the inspection point on the corresponding printed matter based on the position information. Then, an image is input at a timing corresponding to the position of the inspection point in the flow direction to obtain an inspection image.

As a result, as a test image, a solid portion or a solid portion close to a plain color can be selected as an inspection point from the whole picture and input as an image, so that minute print defects such as doctor streaks can be reliably detected. Can be.

FIG. 7 is a block diagram showing a schematic configuration of a solid area selection device according to a second embodiment of the present invention.

As shown by the reference numerals 18-1 to 18-n and 20-1 to 20-n in the figure, the solid area selection device of this embodiment has an evaluation value calculation unit 18 and an evaluation value storage unit. 20
There are n (a plurality of) evaluation values, and n types of evaluation values can be calculated and stored independently of each other. In addition, the determination unit 22 determines a predetermined number of test values set in advance for each evaluation value. The points are determined, and the output unit 24 executes a parallel process of sequentially outputting position information on the inspection points for each evaluation value.

In the present embodiment, when the above-described parallel processing is performed, there are two types of evaluation values for convenience, and each evaluation value is divided into a plurality of inspection areas obtained by dividing the entire image into a plurality of parts with a predetermined width in the vertical direction. In the case where two inspection points are set one by one for each value, that is, a total of two inspection points are set for each inspection area. Are alternately output in units of the predetermined number.

Therefore, when position information relating to inspection points for each evaluation value is input from the output unit 24 to the inspection device, the inspection device may determine that five inspection points are determined by the first evaluation value. As shown conceptually in FIG. 8A, the relationship with the printed matter including the whole picture, the inspection points on the printed matter indicated by five squares in the figure can be input as the inspection image.

After that, when the position information about the inspection point similarly determined based on the second evaluation value is input, as shown in FIG. It is possible to input an image of an inspection point on a printed material having a different pattern tendency from the case.

In the present embodiment, the inspection based on the first evaluation value (for example, the maximum frequency of the above (1)),
The inspection based on the third evaluation value (for example, the partial pixel value of (3) above) can be performed alternately. Therefore, as shown in FIG. 9, when the inspection point is determined in the area of the width W in the whole picture, assuming that the position at the first evaluation value is selected as the inspection point, As long as inspection is performed at a point, even if a doctor streak or the like occurs at this position in the same color as the color of the picture at that point (or a color that is invisible on the inspection point depending on the printing order of each color), the defect is detected. Can not do it.

However, in the present embodiment, as shown in FIG. 10, an evaluation value different from the first evaluation value used for selecting the inspection point, in particular, the brightness in the area is also considered. Another inspection point can be selected based on the second evaluation value (the partial pixel value), and the inspection can be performed at this point. As a result, for example, when the inspection point is blue and the doctor muscles continuously occurring in the vertical direction are also blue or a color similar thereto, the doctor muscles are detected even when the inspection point is input and inspected. I can't,
When the inspection point is selected on a substantially white background, the blue doctor muscle can also be reliably detected. Therefore,
By using two evaluation values of the flatness having different characteristics as image data in this way, the reliability of the inspection can be further improved.

In the above description, the number of inspection points is two for each area. However, three or more types of evaluation values may be used, and three or more inspection points obtained from each of the evaluation points may be selected. .

In the present embodiment, the case has been described in which the determination unit 22 selects inspection points separately based on each of a plurality of types of evaluation values. However, the calculation results of the plurality of evaluation values are comprehensively determined. Thus, one optimal inspection point may be selected for each inspection area in the vertical direction.

That is, in the example of FIG. 8 where the evaluation value is two, two inspection points are alternately set for the same inspection area as shown in FIGS. Instead of performing the inspection, the optimal one may be selected from FIGS. 9A and 9B, and one inspection point may be selected for the same inspection area.

FIG. 11 is a block diagram showing a schematic configuration of a solid region selecting apparatus according to a third embodiment of the present invention.

The solid area selecting apparatus according to the present embodiment, as indicated by reference numerals 22-1 to 22-n in the drawing,
Evaluation value calculation unit 18 and evaluation value storage unit 20 in the case of the embodiment
In addition, n (plural) determination units 22 are provided, and
Each of the first and second evaluation values is connected in series from the evaluation value calculation unit 18 to the determination unit 22, and the final determination result is output from the output unit 24.
It is substantially the same as the embodiment.

In the present embodiment, the processing from the first evaluation value to the n-th evaluation value can be executed up to the determination. Therefore, by making a determination from the first evaluation value, the candidate for the inspection point is gradually determined. Can be narrowed down.

Therefore, for example, by making it possible to calculate the evaluation value earlier as the calculation processing becomes faster, an appropriate inspection point can be quickly selected.

Although the present invention has been specifically described above, the present invention is not limited to the above-described embodiments, and can be variously modified without departing from the gist thereof.

For example, although three types of specific evaluation values have been described, the present invention is not limited to these, and any type can be used without particular limitation as long as it can evaluate the flatness of the luminance value.

[0060]

As described above, according to the present invention,
A part having a small density difference can be input as the inspection image from the entire pattern on the printed matter so that the pattern part included in the input inspection image does not cause erroneous detection of a minute defect.

[Brief description of the drawings]

FIG. 1 is a block diagram schematically showing a solid region selection device according to a first embodiment of the present invention;

FIG. 2 is a diagram showing a relationship between an entire image and a partial image.

FIG. 3 is a diagram showing a method for extracting a partial image.

FIG. 4 is a diagram showing an example of a partial image and its luminance distribution.

FIG. 5 is a diagram for explaining a partially divided pixel value;

FIG. 6 is a flowchart illustrating a processing procedure according to the embodiment;

FIG. 7 is a block diagram schematically showing a solid region selection device according to a second embodiment of the present invention;

FIG. 8 is an explanatory diagram showing parallel processing according to the second embodiment;

FIG. 9 is an explanatory diagram showing inspection points when only one evaluation value is used.

FIG. 10 is an explanatory diagram showing inspection points when two evaluation values are used together.

FIG. 11 is a block diagram schematically showing a solid region selection device according to a third embodiment of the present invention;

FIG. 12 is an explanatory diagram showing an example of a method of setting inspection points.

FIG. 13 is an explanatory diagram showing another example of a method of setting inspection points.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 ... Whole image input part 12 ... Whole image storage part 14 ... Partial image cutout part 16 ... Parameter storage part 18 ... Evaluation value calculation part 20 ... Evaluation value storage part 22 ... Judgment part 24 ... Output part

Continuation of front page (72) Inventor Kenta Hayashi 1-1-1 Ichigaya-Kaga-cho, Shinjuku-ku, Tokyo Dai Nippon Printing Co., Ltd.

Claims (6)

[Claims] When a print state is inspected based on an inspection image input by a partial input camera of an inspection apparatus, a plain area is selected from the entire pattern on a printed matter, and the plain area is used as an inspection point. A solid area selection method to be applied, wherein a step of inputting an entire image including an entire printed pattern and a step of cutting out a partial image from the input entire image in units of a size corresponding to an inspection image And calculating an evaluation value representing the flatness of the luminance for each of the cut-out partial images. Based on the calculated evaluation value, one of the partial images having a high flatness of the luminance is displayed on the image. Selecting a plain area and determining a predetermined number as inspection points on the image; and providing the inspection apparatus with positional information on the determined inspection points. Solid area selection method comprising the steps of force, the. 2. The method according to claim 1, wherein the evaluation value representing the flatness of the luminance is a maximum frequency of a luminance distribution for all pixels included in the partial image, a variance of the luminance distribution,
And at least one of luminance values for dividing the luminance distribution into a predetermined ratio biased toward a high luminance side or a low luminance side. 3. The method according to claim 1, wherein the partial image is sequentially cut out over the entire image while shifting the partial image by at least one pixel in the horizontal direction. . 4. A method according to claim 1, wherein a plain area is selected from the whole picture on the printed matter, and said plain area is used as an inspection point to inspect a printing state based on an inspection image input by a partial input camera of said inspection apparatus. A solid area selection device to be applied, a means for inputting a whole image including the entire printed picture, and a means for cutting out a partial image from the input whole image in units of a size corresponding to an inspection image Means for calculating an evaluation value representing luminance flatness for each of the cut out partial images; and, based on the calculated evaluation value, a partial image having high luminance flatness on the image. Means for selecting as a plain area and determining a predetermined number as inspection points on the image; means for outputting position information on the determined inspection points to the inspection apparatus; Provided by which solid area selection device. 5. The evaluation value representing the flatness of luminance according to claim 4, wherein the evaluation value represents a maximum frequency of a luminance distribution for all pixels included in the partial image, a variance of the luminance distribution,
And at least one of luminance values for dividing the luminance distribution into a predetermined ratio biased toward a high luminance side or a low luminance side. 6. The solid region selection apparatus according to claim 4, wherein the partial image is sequentially cut out over the entire image while shifting the partial image by at least one pixel in the horizontal direction at a time. .