JP2009162579A - Quality evaluator as to section of paper and quality evaluation method using same - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a quality evaluator as to a section of paper for detecting the existence of minute foreign matter on the section independently of the existence of disturbance factors such as the layer-built state of a paper pile or an imaging environment to acquire a highly reliable evaluation result, and a quality evaluation method using the same. <P>SOLUTION: This quality evaluator comprises an imaging device 10 installed opposite to the section 50a of the paper pile 50, a lighting system 20 arranged so that the optical axis A<SB>20</SB>of illuminating light makes 60° to 90° with the optical axis A<SB>10</SB>of the imaging device 10, a quality evaluation value preparation means 30, and an evaluation means 40. The preparation means 30 comprises a contrast evaluation value calculation means 36 for preparing a cooccurrence matrix concerning the density of each of pixels constituting an image on the section 50a acquired through the imaging of the imaging device 10 to find a contrast evaluation value within an evaluation object scope from the cooccurrence matrix. The evaluation means 40 is for evaluating whether or not the found contrast evaluation value is lower than a previously set reference value. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a paper cut surface quality evaluation apparatus and a quality evaluation method using the same, and more particularly to a paper cut surface quality evaluation apparatus that objectively determines whether a paper cut surface is good or not visually. And a quality evaluation method using the same.

2. Description of the Related Art Conventionally, in a paper mill or the like, an appearance inspection of a cut surface is included in a quality inspection of paper manufactured through a paper manufacturing process.
This cut surface appearance inspection (hereinafter referred to as cut surface inspection) is an inspection for cutting the bundled paper in a predetermined direction and evaluating the properties of the cut surface.
However, this cut surface inspection sometimes causes problems such as generation of fuzz due to tearing of some of the paper fibers without being completely cut, and adhesion of powdery paper called paper powder. As a cause of such a problem, there is a reduction in cutting ability of the cutting machine.

Further, when the paper surface is subjected to a coating treatment or the like in the paper making process, there may be a problem that the coating material peels off and adheres to the cut surface at the time of cutting.
The occurrence of fluffing and the adhesion of paper powder and peeled coating material (hereinafter referred to as coating material powder) are not merely aesthetic problems. In particular, when used as a recording medium in a printing apparatus, fine foreign matter such as paper powder and coating material powder adheres to the photosensitive drum of the printing apparatus, and the formed image portion falls out as white spots, resulting in an image defect. In some cases, the printer may fail.

Accordingly, the present situation is that prevention of troubles caused by these minute foreign matters is a very important issue in printing apparatuses such as electrophotographic copying machines and printers.
In order to prevent such problems, the current cut surface inspection visually evaluates the presence of fuzz and minute foreign objects, but there are individual differences in the evaluation and it is difficult to make an objective judgment. . For this reason, as a method for objectively evaluating the cut surface inspection, an inspection apparatus using image processing has been proposed (see Patent Document 1).

In Patent Document 1, a fuzzy portion (a portion having a brighter density than the surroundings) and a portion that does not receive illumination light due to the fuzzing (a portion having a darker density than the surroundings) are extracted by binarization processing, and the area and inspection of the fuzzy portion are examined. It is disclosed that the degree of fluffing is calculated from the ratio of the area of the entire range.
JP 2001-228142 A

However, in the evaluation method disclosed in Patent Document 1, since the fluff portion is extracted and evaluated by binarization processing, only a certain amount of fluff can be detected, and it is as small as paper powder or coating material powder. Foreign matter cannot be detected.
In addition, the threshold value used when performing the binarization process is a factor that has a significant influence on the evaluation result, but Patent Document 1 does not describe any objective determination method of the threshold value.

Therefore, even if the evaluation method described in Patent Document 1 is used, troubles such as image defects and failures of the printing apparatus caused by minute foreign matters such as paper powder and coating material powder have not been solved.
In addition, as a problem with the evaluation method using the binarization process in Patent Document 1, the reliability of the evaluation result is affected by the effect of the horizontal stripe pattern formed on the cut surface of the paper pile by stacking the paper. There is also a problem of lacking. Here, the horizontal direction refers to a direction orthogonal to the paper stacking direction in the cut surface of the stacked paper.

  Here, FIG. 19 is an image showing a conventional cut surface with irregularities. For example, as shown in FIG. 19, when there is unevenness in the paper stacking direction in the paper cut surface, the paper protruding portion (convex portion) is displayed brighter than the retracted portion (concave portion). . As described above, when the variation in brightness exists in the stacking direction, the striped pattern removal method disclosed in Patent Document 1 cannot cope with it. That is, the threshold value used for the optimal binarization process varies depending on the brightness of each layer. As a result, the area of the fluff portion extracted by the binarization process using a single threshold value has a value different from the original area, and there is a problem of evaluation with low reliability.

  Therefore, the present invention has been made in view of the above-described problems, and its purpose is to detect the presence of minute foreign objects on the cut surface regardless of the presence of disturbance factors such as the stacking state of paper piles and the imaging environment. Another object of the present invention is to provide a paper cut surface quality evaluation apparatus and a quality evaluation method using the same, which can obtain highly reliable evaluation results.

In order to solve the above problem, the paper cut surface quality evaluation apparatus according to claim 1 of the present invention is installed at a position opposite to the cut surfaces obtained by cutting the stacked paper piles into a certain size. An imaging device, and an illuminating device arranged so that the optical axis of the irradiation light is 60 to 90 degrees with respect to the optical axis of the imaging device,
Concerning the image obtained by the imaging device imaging the cut surface irradiated with light by the illumination device, the density of each pixel constituting the image is set so that the average density of the evaluation target range is constant. The evaluation expressed by the appearance rate of (Gi, Gj) based on the density Gi of one pixel i converted and the density Gj of another pixel j laterally separated from the pixel i by a displacement δ. A quality evaluation value creating means for creating a co-occurrence matrix of the target range and obtaining a contrast evaluation value of the evaluation target range from the co-occurrence matrix, and the obtained contrast evaluation value is lower than a preset reference value It is characterized by having an evaluation means for evaluating whether or not.

  According to a second aspect of the present invention, there is provided a quality evaluation method for paper cut surfaces, wherein an image pickup device is installed at a position facing a cut surface obtained by cutting a pile of paper piles into a certain size, and the irradiation is performed. An illuminating device is arranged such that the optical axis of light is 60 degrees to 90 degrees with respect to the optical axis of the imaging device, and the imaging device images the cut surface irradiated with light by the illuminating device. Based on the obtained image, a density equalization step for converting the density so that the average density of the evaluation target range is constant, a density Gi of one pixel i, and a displacement δ from the pixel i in the horizontal direction. A co-occurrence matrix calculation step of creating a co-occurrence matrix of the evaluation target range represented by the appearance rate of (Gi, Gj) based on the density Gj of another pixel j, and the evaluation from the co-occurrence matrix Find the contrast evaluation value of the target range A quality evaluation value creation process including a contrast evaluation value computing step, the contrast evaluation value obtained is is characterized by comprising an evaluation step of evaluating whether or lower than a preset reference value.

According to the paper cut surface quality evaluation apparatus according to claim 1 of the present invention, the evaluation means uses the quality evaluation value calculated based on the co-occurrence matrix created for the lateral displacement of the target image. Since the evaluation is performed, the influence of disturbance as seen in the binarization process is extremely small, and even fine foreign objects such as paper powder and coating material powder are objective, reproducible, and highly reliable evaluation results are obtained. An apparatus for evaluating the quality of a paper cut surface can be provided.
According to the quality evaluation method for a paper cut surface according to claim 2 of the present invention, a co-occurrence matrix relating to the lateral displacement of the evaluation target image is created, and the quality evaluation value calculated based on this is used. Because of this, the influence of disturbance as seen in the binarization process is extremely low, and the paper cut surface that obtains an objective and reliable evaluation result even for minute foreign matter such as paper dust and coating material powder. The quality evaluation method can be provided.

Hereinafter, an embodiment of a paper cut surface quality evaluation apparatus and a quality evaluation method using the same according to the present invention will be described with reference to the drawings.
FIG. 1 is a schematic view showing the configuration of a paper cut surface quality evaluation apparatus according to the present invention.
As shown in FIG. 1, the quality evaluation device 1 includes an imaging device 10 that images a cut surface of a paper pile 50 that is an evaluation target, an illumination device 20, a quality evaluation value creation unit 30, an evaluation unit 40, a reference A value storage unit 41.

The imaging device 10 is installed so as to face the cut surface 50a of the paper pile 50 with respect to the paper pile 50 cut into a predetermined size and stacked with the cut surfaces 50a aligned. That is, the imaging device 10, the optical axis A _10, are disposed so as to coincide substantially normal to the direction of the cut surface 50a. The imaging device 10 has a function of outputting image data obtained by imaging the cut surface 50 a in response to a request from the quality evaluation value creating unit 30. An example of the imaging device 10 is a CCD camera.

The illuminating device 20 is installed so that an angle (incident angle) formed by incident light with respect to the cut surface 50a is 0 to 30 degrees. That is, the lighting device 20, the angle between the optical axis A ₂₀ and the normal direction of the cutting surface 50a alpha is arranged so as to be between 90 degrees 60 degrees. The angle alpha, in other words, the cutting surface 50a, an angle between the optical axis _{A 10} of the optical axis _{A 20} and the imaging device 10 of the lighting device 20.
In addition, the imaging device 10 should just be installed in the position which can image the cut surface 50a irradiated with light with the illuminating device 20 at least.

  The quality evaluation value creating unit 30 is a unit that is connected to the imaging device 10 and creates a quality evaluation value based on a cross-sectional image of the paper pile 50 captured by the imaging device 10. The quality evaluation value creating means 30 includes an image memory 31, an image size storage unit 32, an image size conversion unit 33, a density averaging processing unit 34, a co-occurrence matrix calculation unit 35, and a contrast evaluation value calculation unit 36. Have The quality evaluation value creation means 30 includes a co-occurrence matrix storage unit that stores the co-occurrence matrix created by the co-occurrence matrix calculation unit 35 and a contrast evaluation value storage unit that stores a contrast evaluation value as necessary. May be.

The image memory 31 is a storage unit that stores image data (digital gradation image data) supplied from the imaging device 10.
The image size storage unit 32 is a storage unit that stores a desired image size in order to specify the evaluation target range.
The image size conversion unit 33 is means for converting the size of the image stored in the image memory 31 into a desired image size stored in the image size storage unit 32 and storing the image after the size conversion in the image memory 31. .

The density averaging processing unit 34 is a means for calculating equalization of the average value of image density based on the size-converted image data stored in the image memory 31.
The co-occurrence matrix calculation unit 35 is a means for calculating a co-occurrence matrix for each pixel constituting the image data whose density is averaged. A specific calculation method will be described later.
The contrast evaluation value calculation unit 36 is a calculation unit that calculates a contrast evaluation value as a quality evaluation value, and is a unit that calculates a contrast evaluation value based on the co-occurrence matrix calculated by the co-occurrence matrix calculation unit 35.

  The reference value storage unit 41 is a means in which a value set as a maximum value allowed as a contrast evaluation value (hereinafter referred to as a reference value) is stored in advance. The input of the reference value is performed by using, for example, an input device connected to the outside. As this reference value, it is preferable to store contrast evaluation values serving as a plurality of evaluation criteria according to the type of paper, the state of the paper pile, the cutting environment, the image capturing environment, and the like.

  The evaluation unit 40 compares the contrast evaluation value calculated by the contrast evaluation value calculation unit 36 with a predetermined reference value stored in the reference value storage unit 41. When the contrast evaluation value is lower than the reference value, It is a means for outputting evaluation result data such as “foreign matter almost does not exist on the cut surface”. Further, the evaluation unit 40 determines that “foreign matter exists in the cut surface” when the contrast evaluation value calculated by the contrast evaluation value calculation unit 36 is higher than a predetermined reference value stored in the reference value storage unit 41. Is a means for outputting the evaluation result data. Furthermore, the evaluation means 40 has a function of outputting evaluation result data to the outside. That is, the evaluation result data may be output as character data to a display device such as a display connected to the evaluation unit 40, or the evaluation result data may be transmitted to another device connected so as to be able to transmit data. Furthermore, the evaluation means 40 may have a trend function for storing the evaluation results in a predetermined storage means and arranging them in time series and displaying them on a display device or the like. The evaluation means 40 may include a selection means (not shown) for selecting which reference value is used as the evaluation reference when a plurality of reference values are stored in the reference value storage unit 41. .

Here, in the image of the cut surface 50a obtained by the imaging device 10, what kind of difference appears in the image depending on whether or not the cut surface 50a has deposits such as fuzz, paper powder, and coating powder. This will be described with reference to FIGS. 2 (A) and 2 (B).
FIG. 2 is a diagram for explaining the influence of the irradiation light on the optical path depending on whether or not there is a deposit on the cut surface of the paper, and FIG. FIG. 2B is a diagram showing the optical path of the irradiation light with respect to the deposits present on the cut surface of the paper pile.

As shown in FIG. 2 (A), if the deposit on the cut surface 50a is not observed, since the illumination light is often light path P ₁ to continue to reflected at the same angle as the incident angle, the light entering the imaging device 10 Only the light component irregularly reflected at the cut surface 50a is obtained. Therefore, in such a case, image data with uniform density to some extent and almost no contrast can be obtained by the imaging apparatus 10.
On the other hand, as shown in FIG. 2B, when there is an attachment on the cut surface 50 a, the optical path P ₂ where the illumination light is reflected by the attachment portion 51 and the regular reflection component are reflected in the direction of the imaging device 10. often it proceeds with the optical path P _3. Incidentally, the optical path P _3, by light reflected in deposits is directed to the imaging apparatus 10, deposits appear bright, the lateral appear dark shadow. Even when the reflected light does not go in the direction of the imaging device 10 and does not become bright, the side appears dark due to the shadow generated in the deposit portion 51. Therefore, it can be seen that the image with the attached object on the cut surface 50a can be obtained by the imaging device 10 with a higher contrast with a difference in brightness than the image without the attached object on the cut surface 50a.

Here, the illumination device 20 can obtain an image with high contrast by using illumination (for example, LED) in which the direction of the irradiated light is aligned rather than illumination such as a fluorescent lamp that irradiates diffused light. it can.
As described above, in order to evaluate the presence of foreign matter such as fuzz on the cut surface 50a as a result of verifying the presence or absence of an adhering material on the cut surface 50a, the image in the evaluation target range is evaluated with high or low contrast. Is effective. In order to obtain the image data of the cut surface 50a as a high-contrast image, it is necessary to relatively increase the angle formed by the imaging device 10 and the illumination device 20 so that a shadow portion can be seen large.

Next, the installation positions of the imaging device 10 and the illumination device 20 in the configuration of the quality evaluation device 1 will be described.
3 to 5 are the same when the angle α formed by the optical axis A ₁₀ of the imaging device 10 and the optical axis A ₂₀ of the light emitted from the illumination device 20 is changed in the configuration shown in FIG. It is the image of the deposit | attachment which exists on the cut surface 50a.
6 and 7 show the case where the angle α formed by the optical axis A ₁₀ of the imaging device ₁₀ and the optical axis A ₂₀ of the light emitted from the illumination device 20 is changed in the configuration shown in FIG. It is an image of paper dust existing on the same cut surface 50a.

  Specifically, FIG. 3 is an image of the deposit when the angle α is 80 degrees, FIG. 4 is an image of the deposit when the angle α is 60 degrees, and FIG. FIG. 6 is an image of paper dust when the angle α is 85 degrees, and FIG. 7 is an image of paper dust when the angle α is 45 degrees. The angle α formed with the optical axis of the light emitted from the illumination device 20 is such that the optical axis of the imaging device 10 faces the cut surface 50a of the paper pile 50, that is, the normal direction of the cut surface 50a and the imaging. This is an angle in a state where the optical axes of the apparatus 10 are substantially coincident. Therefore, in this premise, the angle α is also an angle formed between the normal direction of the cut surface 50a and the optical axis of the light emitted from the illumination device 20. In the state where the normal direction of the cut surface 50a and the optical axis of the imaging device 10 are substantially coincident, the incident angle of the light emitted from the illumination device 20 with respect to the cut surface 50a is (90−α) degrees. Can be considered.

  As shown in FIGS. 3 to 5, the image shown in FIG. 4 has a higher contrast than the image shown in FIG. In addition, the image shown in FIG. 3 has higher contrast than the image shown in FIG. Further, in the image shown in FIG. 5, a large number of recognizable deposits could not be confirmed in the images shown in FIGS. Therefore, it can be seen that the image of the deposit on the cut surface 50a is preferably set to have an angle α of 60 degrees or more which is adopted when the image shown in FIG. 4 is acquired.

  Further, as shown in FIGS. 6 and 7, the image shown in FIG. 6 has a higher contrast than the image shown in FIG. Further, in the image shown in FIG. 7, many recognizable paper dusts in the image shown in FIG. 6 could not be confirmed. Therefore, it can be seen that the image of the paper dust present on the cut surface 50a is preferably set to have an angle α of 60 degrees or more which is adopted when the image shown in FIG. 4 is acquired.

Here, a preferable angle α is described below with reference to FIG.
When the illuminating device 20 is moved so that the angle α becomes small, the illuminating device 20 illuminates from above the deposit, and the shadow portion becomes small and the contrast is lowered. Therefore, in order to leave a shadow on the image obtained by the imaging device 10, it is necessary to arrange the lighting device 20 so that the angle α is increased to some extent.

  Moreover, the reason why it is preferable to make the imaging device 10 face the cut surface 50a can be explained as follows. If the imaging device 10 is tilted to the lighting device 20 side, it becomes difficult to capture a shadow portion generated by irradiation of the lighting device 20, and conversely, if the imaging device 10 is tilted to the opposite side to the lighting device 20 side, illumination is performed. It becomes difficult to capture the bright part irradiated by the device 20. In other words, the contrast is lowered regardless of which direction the imaging device 10 is tilted in the illumination device 20. Therefore, as shown in the present invention, the image pickup apparatus 10 faces the cut surface 50a, and the angle α formed between the illumination device 20 and the image pickup apparatus 10 is suitably 60 degrees to 90 degrees.

As described above, by setting the angle α in the configuration shown in FIG. 1 to 60 degrees to 90 degrees, a high-contrast image can be obtained, so that the presence of deposits on the cut surface can be evaluated with high accuracy. The angle α is preferably 60 to 89 degrees.
Next, a paper cutting surface quality evaluation method for evaluating the quality of an image obtained by the imaging device imaging a paper cutting surface to be evaluated will be described below with reference to the drawings.

  As shown in FIG. 8, the paper cut surface quality evaluation method according to the present invention includes a cut surface imaging step S1, an evaluation target range specifying step S2, a density averaging step S4, and a quality evaluation value creating step S5. And evaluation step S6. The quality evaluation value creation step S5 includes a co-occurrence matrix calculation step S5a and a contrast evaluation value calculation step S5b. In addition, as the quality evaluation method of the paper cut surface according to the present invention, an image size conversion step S3 may be included between the evaluation target range specifying step S2 and the density averaging step S4 as necessary.

The cut surface imaging step S <b> 1 is a step of imaging the paper cut surface 50 a irradiated with light by the illumination device 20 with the imaging device 10 as in the configuration illustrated in FIG. 1.
The evaluation target range specifying step S2 is a step of determining the evaluation target range in the image data acquired from the imaging device 10 after the cut surface is imaged, as shown in FIGS. For example, the information for determining the evaluation target range is displayed by displaying an image of the acquired image data on a display device (not shown), and using an input device (not shown) connected to the outside based on the image. Then, the number of pixels in the X and Y directions is input and the evaluation target range displayed by a frame or the like is designated (see FIG. 14).

  As shown in FIGS. 1 and 8, the image size conversion step S <b> 3 determines the evaluation target range in the evaluation target range specifying step S <b> 2, and then based on the image size data stored in the image size storage unit 32. The image size conversion unit 33 converts the size of the image data acquired from the above. The input of the image size data is performed by using, for example, an input device connected to the outside.

In the quality evaluation method using the quality evaluation apparatus 1 shown in FIG. 1, various disturbances such as the stacking state of the paper piles 50 and the unevenness of illumination light due to deterioration of the illumination apparatus 20 may be received. For example, when the lighting device 20 deteriorates and the overall brightness becomes dark, the contrast may decrease regardless of the presence or absence of foreign matter on the cut surface 50a.
Therefore, in order to reduce such a decrease in contrast due to a change in brightness in the evaluation target, a density averaging step S4 is performed to equalize the average value of the density of the image to be evaluated.

Here, a specific method of the concentration averaging step S4 will be described below.
In FIG. 1, image data whose evaluation target range has been specified in the evaluation target range specifying step S <b> 2 or image data whose image size has been converted in the image size conversion step S <b> 3 is stored in the image memory 31. This image data is divided for each charge element (pixel) arranged in a number of two-dimensional coordinates (XY coordinates). The image data is a set of coordinate positions in XY coordinates and brightness numerical values (density) at the positions.

Here, when the numerical value of the density of the pixel at a certain coordinate position (x, y) is f, pixel information relating to the density can be expressed as f (x, y).
Therefore, in the density averaging step S4, the density average processing unit 34 reads out the image data stored in the image memory 31 in FIG. 1, and the calculation according to the following formula (1) is performed on the image data for each pixel. Then, pixel information g (x, y) is obtained. In the following formula (1), h is an arbitrary constant determined so that the processed image does not become too dark, and AVE (A) represents the average density value of the image in the evaluation target range.
g (x, y) = f (x, y) × h / AVE (A) (1)

The co-occurrence matrix calculation step S5a is a step of digitizing the level of contrast of the image data obtained by averaging the density of each pixel in the density averaging step S4. Specifically, as shown in FIG. 1, the co-occurrence matrix calculation unit 35 creates a co-occurrence matrix for each pixel constituting the image data in which the average value of the density is uniformed by the density averaging processing unit 34. .
Here, the co-occurrence matrix is represented by S (Gi, Gj), where Gi and Gj are the densities of pixels j that are relatively distant from a certain pixel i by δ on the image. Is a matrix display of the appearance probabilities of density combinations S.

Note that when the density of each pixel in the image data from which the co-occurrence matrix is created is expressed by n (bits), the co-occurrence matrix is expressed by 2 ⁿ rows × 2 ⁿ columns.
For the co-occurrence matrix, see, for example, Mikio Takagi, Yoshihisa Shimoda: “Texture Feature Extraction”, “Image Analysis Handbook”, University of Tokyo Press, pp. 518-521.
The co-occurrence matrix calculation step S5a includes a sampling step and a normalization step.

  9 to 11 are diagrams showing specific examples of the process of the co-occurrence matrix calculation step S5a used in the present invention, and FIG. 9 is a diagram showing the coordinates and density of image data to be evaluated. 10 is a diagram showing a matrix of density combinations S sampled from image data to be evaluated, and FIG. 11 is a diagram showing a co-occurrence matrix created in the normalization process. FIG. 9 shows the coordinates (upper stage) and density (lower stage) of each pixel in image data composed of 5 rows and 5 columns as a specific example.

In the sampling process, for example, in the image data of the image to be evaluated, a combination (Gi) of a density Gi of one pixel i and a density Gj of another pixel j that is laterally displaced δ from the one pixel i. , Gj) is sampled in a matrix format. The “lateral direction” refers to a direction orthogonal to the paper stacking direction in the cut surface of the stacked paper. Further, the matrix sampled in this sampling process is expressed by 2 ⁿ rows × 2 ⁿ columns when the density of the image data to be sampled is expressed by n (bit). The dimension of the displacement δ is preferably 0.3 mm to 1.0 mm. In addition to the dimensions, it is also preferable to set the displacement δ to a pixel equivalent to 0.3 mm to 1.0 mm.

Here, first, the reason why the direction of the displacement δ is specified in the lateral direction will be described below.
In general, the displacement δ used when calculating the co-occurrence matrix is a pixel to be combined (positioned in a predetermined angular direction (stacking direction and lateral direction) in the cut surface with respect to the reference pixel (pixel i) ( Pixel j) is selected as appropriate.
In the present invention, since the cut surfaces of the laminated paper have the following characteristics (1) to (3), from the reference pixel (pixel i) to the pixel to be combined (pixel j) Was defined as the amount of displacement moved laterally.

(1) In the cut surface image, the density in the stacking direction changes depending on the stacking state of the paper (see FIG. 19). The density also changes when illumination light is not uniformly irradiated in the stacking direction of the cut surface.
(2) Although the lateral density is affected by the uniformity of the illumination light, it can be considered that the illumination light is illuminated uniformly if the lateral spacing is very small.
(3) A deposit can be understood as a change in concentration in a minute range.

In this way, if the density difference between the pixels at a small displacement δ in the horizontal direction is obtained, the density values above and below the reference pixel are not referred to for the calculation of the quality evaluation value. Without considering the pattern density difference, it is not necessary to correct the pixel density each time for the non-uniformity of the illumination light in the cut surface, and by making the displacement δ a minute value, the illumination is non-uniform. A small foreign matter such as paper dust can be recognized without being affected by the difference in brightness of the image in the horizontal direction.
In other words, by specifying the direction of the minute displacement δ in the horizontal direction, it is possible to accurately detect changes in density due to deposits without being affected by disturbances such as paper pile stacking and illumination light non-uniformity. Can be expressed.

  Here, the reason why the value of the displacement δ is set to 0.3 mm to 1.0 mm will be described with reference to FIGS. FIG. 12 is a graph showing changes in the quality evaluation value (described later) when the displacement δ is changed depending on the degree of adhesion of paper dust. FIG. 12 shows the evaluation results for three samples: “paper dust adhered (very much)”, “paper dust adhered (large)”, and “paper dust adhered little”. “Paper dust adheres (very much)” and “paper dust adheres (large)” are levels at which the product cannot be shipped. Further, in FIG. 12, the unit of displacement δ is changed from 1 pixel to 30 pixels in one pixel, and one pixel is equivalent to 0.08 mm. FIGS. 13A and 13B are diagrams showing the concentration values at two locations in the deposit portion 51 irradiated by the illuminating device 20 and their distance (displacement δ).

  As shown in FIG. 12, in the region where the displacement δ is 3 pixels or less, the deposit cannot be correctly evaluated, and the three samples show similar quality evaluation values. As shown in FIG. 13A, since the displacement δ is too small with respect to the size of the deposit to be detected, the density value Gi and the places where the density value Gj changes greatly are reduced. This is thought to be because the bright spots and shadow spots generated by the camera cannot be captured.

On the other hand, as shown in FIG. 12, in the region where the displacement δ is 4 pixels to 10 pixels, when there are few deposits, a stable quality evaluation value (variation range of the quality evaluation value is ± 3) is shown. This is presumably because, as shown in FIG. 13B, when the value of the displacement δ is an appropriate value, there are many places where the density value Gi and the density value Gj change greatly.
As shown in FIG. 12, when the displacement δ exceeds 15 pixels, the quality evaluation value starts to increase. This cause is considered to be affected by uneven lighting and undulation of paper piles. That is, when the displacement δ exceeds 15 pixels, the displacement δ is no longer a minute displacement, and the quality evaluation value increases due to various disturbances.

Thus, the displacement δ has an appropriate range for evaluating the deposit. In the present invention, this range is about 4 to 12 pixels, and when converted to a distance value, this range is 0.3 to 1.0 mm.
However, in general, since the magnification of the lens of the imaging device is changed according to the size of the deposit or the like to be detected, the appropriate range of the displacement δ is not limited to the size of the deposit or the like to be detected. Varies depending on the imaging conditions.
Therefore, the displacement δ is appropriately selected based on the imaging conditions and the image obtained by the imaging device.

Next, a specific method of the sampling process will be described with reference to FIGS. In this description, the image data shown in FIG. 9 is assumed to be image data composed of 5 × 5 25 pixels, and the displacement δ is assumed to be a distance of one pixel.
In the sampling process, for example, in the image data shown in FIG. 9 that is an evaluation target, the density G ₁ of the pixel (0, 0) at the coordinate (0, 0) is 0, and the horizontal direction from the pixel (0, 0). Since the density G ₂ of the pixel (1, 0) that is displaced δ (= 1 pixel) in the direction is 0, it can be expressed as (G ₁ , G ₂ ) = (0, 0). The coordinate concentration _{G 3} pixels (1,1) (1,1) is 1, the pixel displacement from (1,1) in the transverse direction [delta] (= 1 pixel) apart pixel (2,1) Since the density G ₄ of the second is 2, it can be expressed as (G ₃ , G ₄ ) = (1, 2).

In this way, a sample of (Gi, Gj) is created for the pixels constituting the evaluation target range, but in the image data shown in FIG. 9, the density range is 0 to 3, so FIG. Can be represented by a 4 × 4 matrix as shown in FIG.
Here, the matrix shown in FIG. 10 indicates that the number of samples in which the combination (Gi, Gj) of density when the displacement δ is (0, 0) in the image data shown in FIG. 9 is three. means. That is, a sample whose density combination (Gi, Gj) is (0, 0), (1, 1), (2, 2), and (3, 3) It means that the density difference is zero.

In addition, in the image data composed of 25 pixels of 5 × 5 as shown in FIG. 9, when the displacement δ is sampled as a distance of one pixel, the sampling showing the density difference between adjacent pixels in the horizontal direction Is performed for each row, so the total number of samples is 20.
The normalization step is a step of converting each element (number of concentration combinations S) of the matrix created in the above-described sampling step into its appearance probability. Specifically, the co-occurrence matrix calculator 35 divides each element (number of concentration combinations S) of the matrix created in the sampling step by the total number of samples.

Next, a specific method of the normalization process will be described with reference to FIGS. Here, in this description, the target to be normalized is a matrix shown in FIG. 10 obtained by sampling image data as shown in FIG.
The normalization process is a process of dividing each element (number of samples) of the matrix shown in FIG. 10 by the total number of samples 20 and representing it as a matrix shown in FIG. That is, in the matrix shown in FIG. 11, for example, the sample probability that the combination S (Gi, Gj) of the density when the displacement δ is (0, 0) is 0.15.

The contrast evaluation value calculation step S5b obtains a frequency distribution for each absolute value of the density difference between two points from the co-occurrence matrix created in the co-occurrence matrix calculation step S5a, and evaluates contrast as a quality evaluation value therefrom. This is a step of calculating a value (hereinafter referred to as a contrast evaluation value). Various calculation methods have been proposed. In the present invention, a method for calculating the contrast evaluation value C expressed by the following formula (2) is adopted. In the following formula (2), C is a contrast evaluation value, d is a density difference between two points, and K [d] is a probability that the density difference between two points is d.
C = Σ {d ² × K [d]} (2)

Hereinafter, a specific example of calculating the contrast evaluation value using Equation (2) will be described with reference to the drawings.
In the contrast evaluation value calculation step S5b, the contrast evaluation value calculation unit 36 performs the following calculation based on Expression (2) for the co-occurrence matrix (see FIG. 11) created by the co-occurrence matrix calculation unit 35.
Contrast evaluation value C = (sum of density difference between two points × probability) = (0−1) ² × 0.1 + (1-2) ² × 0.1 + (2-3) ² × 0. 05+ (3-0) ² × 0.05
= 1 × 0.1 + 1 × 0.1 + 1 × 0.05 + 9 × 0.05
= 0.7

Here, in the above calculation, “(0−1) ² × 0.1” means that the element of (Gi, Gj) = (1, 0) of the co-occurrence matrix (see FIG. 11) is “between two points. This is because “density difference” is “Gj−Gi”, that is, “0-1”, and “probability” is “0.1”. Further, in the contrast evaluation value calculation step S5b, each probability shown in the first row, first column, second row, second column, third row, third column, and fourth row, fourth column in the co-occurrence matrix (see FIG. 11) is between two points. Since the density difference is 0, it is not calculated.

In this way, by not adopting a matrix element having a density difference of 0 between two points as an evaluation element, even if a paper pile rises due to inappropriate adjustment of paper moisture, the paper It can be evaluated in the same way that there is no mountain lift.
This is because Formula (2) is only subject to calculation only in the horizontal direction, so the brightness difference (density difference) in the stacking direction at the raised portion having the same density in the horizontal direction is not affected. is there.

Next, the evaluation step S5 in the quality evaluation method according to the present invention will be described with reference to FIGS.
FIGS. 14 to 18 are images showing the properties of the cut surface for each contrast evaluation value. FIG. 14 is an image of the cut surface (contrast evaluation value is determined by visual inspection). 185). FIG. 15 is an image of the cut surface (contrast evaluation value is 98) which is judged to be possible even if it is normally used in a printing apparatus without the presence of any deposit on the cut surface. FIG. 16 is an image of a cut surface (contrast evaluation value is 58) in which it was determined that the presence of minute foreign matters such as paper dust was not recognized visually. FIG. 17 is an image of a cut surface (contrast evaluation value is 400) where it was determined that the presence of fluff was visually recognized. FIG. 18 is an image of a cut surface (contrast evaluation value is 128) in which it was determined that the presence of the coating material was visually confirmed. Note that these contrast evaluation values are contrast evaluation values calculated based on image data constituted by pixels whose density is expressed by 8 bits, and a co-occurrence matrix created based on the image data has 256 rows. X Expressed in 256 columns.

As shown in FIGS. 14 to 18, it can be confirmed that there is a correlation between the visual evaluation and the contrast evaluation value regarding the presence of the deposit on the cut surface.
Specifically, according to the image in FIG. 14, it can be seen that when the contrast evaluation value is 185, the presence of minute foreign matter on the cut surface can hardly be visually confirmed.
Further, according to the image of FIG. 15, it can be seen that if the contrast evaluation value is 98, the presence of minute foreign matter on the cut surface can hardly be visually confirmed.

Further, according to the image of FIG. 16, when the contrast evaluation value is 58, it can be seen that the presence of minute foreign matters on the cut surface is not visually observed.
Moreover, according to the image of FIG. 17, when the contrast evaluation value is 400, it can be seen that the presence of fluff is visually recognized in the cut surface.
Moreover, according to the image of FIG. 18, when the contrast evaluation value is 128, it can be seen that the presence of the coating material is visually recognized in the cut surface.

  Therefore, in the present invention, the evaluation unit 40 compares the contrast evaluation value calculated by the contrast evaluation value calculation unit 36 with the reference value stored in the reference value storage unit 41, and evaluation result data based on the comparison result. Is output. Specifically, the contrast evaluation value and the reference value stored in the reference value storage unit 41 are compared, and when the contrast evaluation value is lower than the reference value, “the minute foreign matter is not substantially present in the cut surface”. When the contrast evaluation value is higher than the reference value, the evaluation result data that “a minute foreign matter exists in the cut surface” is output.

  In the evaluation of the cut surfaces shown in FIGS. 14 to 18 described above, it is preferable to treat the contrast evaluation value shown in FIG. 18 as a reference value. The image shown in FIG. 14 was judged to have no problem with the properties of the cut surface by visual inspection with a contrast evaluation value 185, but the image shown in FIG. This is because the presence was confirmed visually.

  As described above, the reference value (128) is stored in the reference value storage unit 41, and the cut surface shown in FIGS. 15 and 16 showing the contrast evaluation value lower than the reference value (128) is cut by minute foreign matter. The evaluation means 40 determines that there is almost no surface. 14, 17, and 18, which show a contrast evaluation value higher than the reference value (128), is judged by the evaluation means 40 that minute foreign matter exists in the cut surface.

  Then, for the image of FIG. 14 with a contrast evaluation value of 185, evaluation result information that “if this cut paper pile is used in the printing apparatus, there is a possibility of causing an image defect or a failure of the printing apparatus”. Is output by the evaluation means 40. Further, for the image of FIG. 15 having a contrast evaluation value of about 98, evaluation result information that “the property of the cut surface is at a level that does not require consideration of the influence on the printing apparatus even if it is normally used” is evaluated. The means 40 outputs. For the image in FIG. 16 having a contrast evaluation value of about 58, the evaluation means 40 outputs evaluation result information “a state in which there are almost no minute foreign matters”. For the image of FIG. 17 having a contrast evaluation value of 400, evaluation result information that “when this cut paper pile is used in a printing apparatus, there is a very high possibility of causing an image defect or a malfunction of the printing apparatus”. Is output by the evaluation means 40. For the image of FIG. 18 having a contrast evaluation value of 128, evaluation result information that “if this cut paper pile is used in a printing apparatus, there is a possibility of causing an image defect or a malfunction of the printing apparatus” is evaluated. The means 40 outputs.

Further, the blade of the cutting machine can be managed by storing the quality evaluation value in a predetermined storage means and displaying it on a display device by graphing it in time series.
As described above, the evaluation unit 40 determines whether or not the contrast evaluation value of the evaluation target image is lower than the reference value stored in advance in the reference value storage unit 41, so that a minute foreign object on the cut surface is obtained. The presence of can be evaluated. The reference value is determined according to the type of paper, the state of the paper pile, the cutting environment, the image capturing environment, and the like.
As mentioned above, although embodiment of this invention has been described, this invention is not limited to this, A various change and improvement can be performed.
For example, when evaluating only a relatively large object such as fluff, it is possible to reduce the density level by neglecting minute density fluctuations by converting the density of the target image before obtaining the co-occurrence matrix.

It is the schematic which shows the structure of the quality evaluation apparatus of the cut surface of the paper which concerns on this invention. It is a figure explaining the influence which it has on the optical path of irradiation light by the quality evaluation method of the cut surface of the paper which concerns on this invention by the presence or absence of the deposit | attachment in the cut surface of paper. It is an image which shows fuzz when the incident angle of illumination is 10 degree | times in the quality evaluation method of the cut surface of the paper concerning this invention. It is an image which shows fuzz when the incident angle (alpha) of an illuminating device is 20 degree | times in the quality evaluation method of the cut surface of the paper concerning this invention. It is an image which shows fuzz when the incident angle (alpha) of an illuminating device is 45 degree | times in the quality evaluation method of the cut surface of the paper which concerns on this invention. It is an image which shows paper dust when the incident angle (alpha) of an illuminating device is 5 degree | times in the quality evaluation method of the cut surface of the paper concerning this invention. It is an image which shows paper dust when the incident angle (alpha) of an illuminating device is 45 degree | times in the quality evaluation method of the cut surface of the paper concerning this invention. It is a flowchart of the quality evaluation method of the cut surface of the paper which concerns on this invention. It is a figure which shows the outline | summary of the image data used as evaluation object in the quality evaluation method of the cut surface of the paper concerning this invention. It is a figure which shows the matrix which shows the sample number of the combination S (Gi, Gj) of the density | concentration of two pixels from the image data used as evaluation object in the quality evaluation method of the cut surface of the paper concerning this invention. It is a figure which shows the co-occurrence matrix produced in the quality evaluation method of the cut surface of the paper concerning this invention. 6 is a graph showing the relationship between the displacement δ and the quality evaluation value when the displacement δ is changed in the paper cut surface quality evaluation method according to the present invention. In the paper cut surface quality evaluation method according to the present invention, it is a diagram for explaining the detection state of the density value of the deposit when the displacement δ is changed. 5 is an image of a cut surface having a contrast evaluation value of 185 in the paper cut surface quality evaluation method according to the present invention. It is an image of a cut surface with a contrast evaluation value of 98 in the paper cut surface quality evaluation method according to the present invention. 6 is an image of a cut surface having a contrast evaluation value of 58 in the paper cut surface quality evaluation method according to the present invention. 4 is an image of a cut surface having a contrast evaluation value of 400 in the paper cut surface quality evaluation method according to the present invention. It is an image of a cut surface with a contrast evaluation value of 128 in the paper cut surface quality evaluation method according to the present invention. It is the image of the conventional cut surface in case a cut surface has an unevenness | corrugation.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Quality evaluation apparatus 10 Imaging device 20 Illumination apparatus 30 Quality evaluation value preparation means 40 Evaluation means 50 Paper pile 50a Cut surface

Claims (2)

An imaging device installed at a position facing a cut surface obtained by cutting a pile of paper piles to a certain size, and an optical axis of irradiation light is 60 to 90 degrees with respect to the optical axis of the imaging device. A lighting device arranged to make
Concerning the image obtained by the imaging device imaging the cut surface irradiated with light by the illumination device, the density of each pixel constituting the image is set so that the average density of the evaluation target range is constant. The evaluation expressed by the appearance rate of (Gi, Gj) based on the density Gi of one pixel i converted and the density Gj of another pixel j laterally separated from the pixel i by a displacement δ. A quality evaluation value creating means for creating a co-occurrence matrix of the target range and obtaining a contrast evaluation value of the evaluation target range from the co-occurrence matrix, and the obtained contrast evaluation value is lower than a preset reference value An evaluation means for evaluating whether or not the paper has a cut surface.   An imaging device is installed at a position opposite to the cut surface obtained by cutting the piles of paper piles to a certain size, and the optical axis of the irradiated light is 60 to 90 degrees with respect to the optical axis of the imaging device. Based on an image obtained by the imaging device imaging the cut surface irradiated with light by the illumination device, the density of the evaluation target range is constant. The rate of appearance of (Gi, Gj) based on the density equalization step for converting, the density Gi of one pixel i, and the density Gj of another pixel j laterally separated from the pixel i by a displacement δ A co-occurrence matrix calculation step of creating a co-occurrence matrix of the evaluation target range represented by the above, and a quality evaluation value generation step including a contrast evaluation value calculation step of obtaining a contrast evaluation value of the evaluation target range from the co-occurrence matrix; Seeking The contrast evaluation value is, the quality evaluation method of cutting surfaces of the paper, characterized in that it comprises an evaluation step of evaluating whether or lower than a preset reference value.