JP2000211094A - Image processing method and system and gravure plate carving method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically measure the area and vol. of a cell even in such a case that the bank portion of the channel generating cell of a gravure printing cylinder is interrupted. SOLUTION: An image obtained by photographing a channel generating cell is taken in to be binarized (S1, S2) and the center segments of the bank portion of the channel generating cell are extracted on the basis of the binarized image (S4) and a straight line connecting the center segments mutually wherein the distance between the end points of the mutual center segments is nearest and the direction at the end points is opposed to a predetermined angle range is formed to obtain an interpollation straight line (S5) and the image of the interrupted bank portion is formed on the basis of the interpollation straight line and this image and the binarized image are synthesized by combining white and black colors (S6) and the contour forming the channel generating cell is extracted from the synthetic image (S7) and the cell is divided at the channel portion of the contour (89) and at least an area and a vol. are measured with respect to the divided cells (SIR).

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and a system for measuring the area and volume of a cell of a gravure printing cylinder, in particular, a cell called a channel generating cell, and to obtain a gravure plate using the same. Gravure engraving method and a gravure engraving system for the same.

[0002]

2. Description of the Related Art As a cell formed on a gravure plate surface of a gravure printing cylinder, there is a cell called a channel generation cell. FIG. 26 is a diagram showing an example of this. In FIG. 26, a portion painted black is an engraved portion, and ink is applied to this portion. The large diamond-shaped large area portion painted in black is generally called a cell, and the cells arranged in the vertical direction in the figure are called channels whose width in the horizontal direction is narrowed. Connected by parts. In the figure, the portion indicated by A is a channel. Further, in the figure, the portions represented in white are portions that are not engraved, and are generally called banks.

[0003] Such a channel generation cell is generally formed by engraving with an engraving needle.
It is known that the area and volume of cells formed vary depending on the wear of the engraved portion of the engraving needle, the conditions of the gravure plate surface of the cylinder, and the like.

[0004] Variations in the area and volume of the cells formed on the gravure plate surface, especially variations in the cell volume, greatly affect the quality of printing. That is, in gravure printing, printing is performed by transferring the ink contained in the concave cells formed on the gravure plate surface to the printing paper, and the printing density changes depending on the amount of ink contained in the cells, The formation of cells of the correct volume on the gravure printing plate is critical for obtaining proper print quality.

Therefore, when engraving a cell on a gravure printing plate, usually, prior to actual engraving for engraving a picture, a cell having a predetermined volume is formed for predetermined input data. The engraving conditions are set. For this reason, it is common practice to perform test engraving at an appropriate location on the gravure plate to determine whether the engraving conditions are appropriate.

Conventionally, whether or not the engraving conditions are appropriate is determined by observing the cells formed on the gravure plate surface by test engraving with a microscope with a scale and measuring the dimensions of the cells. Have been done. That is, it is determined whether the engraving condition is appropriate based on whether the vertical dimension or the left-right dimension of the cell is a predetermined dimension with respect to predetermined input data.

[0007]

However, in the conventional measurement, since the cursor is manually adjusted to the width and length of the enlarged and displayed cell, there is an individual difference, and reproduction when the same individual is repeatedly measured is performed. Poor nature. In addition, it is normal to repeatedly perform measurement → adjustment → engraving → measurement to increase the adjustment accuracy, but it is not easy to perform the adjustment sufficiently due to limitations in workability and work time.

When the cell size and the cell volume are in a one-to-one relationship, an appropriate determination can be made by the above-described method. Therefore, this method cannot determine with sufficient accuracy. In particular, this method cannot be used when the demand for print quality is high. For example, in order to shorten the production time of the gravure plate, engraving may be performed on the gravure plate surface using a plurality of mechanical engraving heads, for example, 4 to 8 mechanical engraving heads simultaneously. In that case, a slight difference in the engraving characteristics of the mechanical engraving head appears as a difference in color tone between adjacent patterns on the printed matter. The difference in color tone is extremely conspicuous because they are adjacent to each other, and the demand for print quality is high. However, the above-described method cannot cope with such a case.

[0009] For this reason, in the prior art, poor color tone was found only after printing was actually performed in the printing process, and the gravure plate had to be corrected or reprinted (reproduced).
The waste of material, time, labor and downtime of the printing press is extremely high.

Further, in the case of expressing high density in the channel generating cell, the amount of digging of the cylinder is increased in order to increase the volume of the cell where the ink is loaded, but the engraving needle wears. In such a case, when engraving a cell expressing a high density, the bank portion may be cut off as shown in FIG. 27, and the cell may be connected to an adjacent cell. Measuring the volume of each cell is important for quality control. However, in the related art, it has been difficult to accurately measure the volume of each of the channel generating cells whose banks have been shaved as shown in FIG.

Therefore, the present invention can automatically and accurately measure the area and volume of a cell even in a channel generating cell whose bank has been cut off, and thus can manage the gravure plate well. Moreover, it is an object of the present invention to provide an image processing method and an image processing system capable of greatly reducing the burden on an operator when measuring the area and volume of a cell.

It is another object of the present invention to provide a gravure engraving method and a gravure engraving system capable of forming cells having a good volume regardless of the degree of wear of the engraving needle. Is what you do.

[0013]

In order to achieve the above object, an image processing method according to claim 1 takes an image of a channel generation cell, binarizes the image, and performs binarization based on the binarized image. To extract the center line segment of the bank of the channel generation cell, the distance between the end points of the center line segment is closest,
In addition, an interpolation straight line is obtained by creating a straight line connecting the center line segments whose directions at the end points face each other within a predetermined angle range, and an image of a broken bank portion is created based on the interpolation straight line. The image of the bank and the binarized image are combined by combining black and white colors, a contour forming a channel generation cell is extracted from the combined image, and the cell is divided at the channel portion of the contour. At least the area and the volume of the divided cells are measured. The image processing system according to claim 2 captures an image from the imaging unit for obtaining an image of the channel generating cell engraved on the gravure plate surface of the cylinder, binarizes the image, and binarizes the image. The center line segment of the bank portion of the channel generation cell is extracted based on the image, and the distance between the center line segments whose end points are closest to each other and the direction of the end points oppose each other within a predetermined angle range is determined. A straight line is created to obtain an interpolation straight line, an image of the broken bank portion is created based on the interpolation straight line, and the black and white color of the broken bank image and the binarized image are matched. From the composite image, extract the contour forming the channel generation cell, divide the cell at the channel portion of the contour,
An image processing means for measuring at least the area and the volume of the divided cells is provided. The gravure plate engraving system according to claim 3, wherein an image pickup means for picking up an image of a channel generating cell formed on the gravure plate surface to obtain image data, and engraving conditions for forming the channel generating cell, and inputting engraving condition data. The image from the engraving condition input means to be obtained and the image from the image pickup means are taken in, binarized, and the center line segment of the bank portion of the channel generating cell is extracted based on the binarized image. Is the closest, and the direction at the end point creates a straight line connecting the center line segments facing each other within a predetermined angle range to obtain an interpolation straight line, and based on the interpolation straight line, an image of the bank part cut off based on the interpolation straight line is obtained. Then, the image of the broken bank portion and the binarized image are combined by combining black and white colors, a contour forming a channel generation cell is extracted from the combined image, and the contour is checked. Dividing the cell in panel portion, characterized in that it comprises a cell volume calculating means for measuring the volume for the divided cell. The gravure plate engraving system according to claim 4, wherein in the gravure plate engraving system according to claim 3, the engraving condition data includes an engraving needle angle which is an angle of an engraving needle attached to an engraving head, and engraving of the engraving needle. It is characterized by a needle abrasion degree. The gravure printing method according to claim 5, wherein a test engraving is performed based on the reference data to form a test engraving channel generating cell on the gravure printing plate, and the test engraving channel generating cell is imaged. An image is fetched and binarized, and the center line segment of the bank portion of the channel generation cell is extracted based on the binarized image, and the distance between the end points of the center line segments is the shortest and the direction at the end point is predetermined. A straight line connecting the center line segments facing each other in the angle range is obtained to obtain an interpolation straight line, and an image of the broken bank portion is created based on the interpolation straight line, and an image of the broken bank portion is created. The binarized image is combined with black and white colors and combined, the contour forming the channel generation cell is extracted from the combined image, the cell is divided at the channel portion of the contour, and the divided cell is divided. And the cell volume calculation step of calculating the volume for,
When the measurement volume data obtained in the cell volume measurement process is compared with the reference volume data to determine the necessity of correcting the engraving condition, and when the necessity of correction is determined by the comparison determination process, In the engraving condition correcting step of correcting the engraving conditions, and when the necessity of correction is determined in the comparison determining step, main engraving is performed based on the engraving conditions to form a channel generating cell on the gravure plate surface, and the gravure is performed. It is characterized by the main engraving process of completing the plate. The gravure engraving method according to claim 6, wherein, in the gravure engraving method according to claim 5, when the engraving condition correcting step is performed, the test engraving step, the cell volume measuring step, and the comparison determining step are performed. It is characterized by repeating.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments will now be described with reference to the drawings. First, an image processing system according to the present invention, that is, the area of a channel generating cell formed on a gravure plate surface and having a bank cut off. And an image processing system for automatically measuring the volume will be described, and then an embodiment of the gravure engraving system according to the present invention will be described.

FIG. 1 is a diagram showing an embodiment of an image processing system according to the present invention. In the figure, 1 is a microscope, 2 is an image processing device, 3 is an input unit, 4 is a monitor, 5 is a printer, 6
Is video capture board, 7 is CPU, 8 is RO
M and 9 are RAMs and 10 is a storage unit.

The microscope 1 is for picking up an enlarged image on the gravure plate surface on which the channel generating cells are formed, and is a light source for illuminating the surface of the gravure plate surface, and for obtaining an enlarged image of the surface of the gravure plate surface. The microscope is provided with a microscope and a CCD camera for capturing an enlarged image obtained by the microscope. The CCD camera may be a black and white camera. That is, the microscope 1 functions as an image acquisition unit.

As the image processing device 2, a personal computer or a workstation can be used. C
The PU 7 executes processing to be described later, and the ROM 8
Stores a program for executing processing to be described later, various data used when executing processing to be described later, and the like. The RAM 9 is used when the CPU 7 executes processing to be described later.

The storage unit 10 is a large-capacity storage device such as a hard disk. Although not shown, the storage unit 10 stores engraving condition data used when calculating the cell volume, and various tables described later.

The image processing apparatus 2 includes a microscope 1
The video capture board 6 for taking in the image picked up by the device is connected. The input unit 3 includes an input device such as a keyboard and a mouse.

Here, the engraving condition data will be described as follows. Obviously, when determining the volume of a cell, the depth of the cell must be known. By the way, when engraving a cell with an engraving needle, the three-dimensional shape of the cell to be engraved differs depending on the shape of the needle of the engraving needle. In this case, the angle of the engraving needle is particularly important. If the angle is different, the depth of the cell will be different even if the opening shape of the cell is the same. This indicates that it is necessary to know the angle of the engraving needle when obtaining the cell volume.

Also, the degree of wear of the engraving needle becomes a problem. This is because the wear of the engraving needle progresses according to the use time, and the needle shape of the engraving needle gradually changes. As a result, even when the engraving conditions are the same except for the wear of the engraving needle, the cell obtained by engraving is obtained. This is because the shape of the opening differs from the depth of the cell. The progress of wear of the engraving needle is determined by the use time. However, strictly speaking,
The degree of wear depends not only on the usage time, but also on the quality of the engraving needle, the difference in the image data to be engraved, the difference in the physical properties of the gravure cylinder plate to be engraved, etc.
It can be approximated that the degree of progress of wear is determined by the use time. Therefore, at least an engraving needle angle and a use time corresponding to the degree of abrasion of the engraving needle are input from the input unit 3 as engraving conditions, and registered in the storage unit 10 as engraving condition data.

Next, the processing performed by the image processing system shown in FIG. 1 will be described with reference to FIG. 2 together with the image processing method according to the present invention.

First, the operator sets the microscope 1 so that an enlarged image of a channel generation cell at a desired portion of the gravure plate surface to be measured can be obtained. The input unit 3 sets the density to be achieved when printing and the magnification of the microscope 1 at this time. Then, execution of image capture is instructed from the input unit 3.

Thus, the image data from the microscope 1 is transmitted to the image processing device 2 via the video capture board 6.
It is taken in. The digitization of the image captured by the microscope 1 may be performed inside the microscope 1 and the digital image data may be transferred to the video capture board 6, or the microscope 1 may output an analog image signal. Alternatively, the video may be digitized by the video capture board 6.

Then, the CPU 7 develops the captured image data in the RAM 9 and displays the image data on the monitor 4.
At this time, it goes without saying that the image data may be stored in the storage unit 10. It is assumed here that the image shown in FIG. 4 has been captured.

The above is the processing for capturing an image from the microscope 1 in step S1 in FIG. 2. After the processing in step S1, the operator instructs the input unit 3 to execute image processing.

When execution of image processing is instructed, the CPU 7
Performs the process of binarizing the captured image data,
Prior to this, first, a process of determining a threshold value T for binarization is performed. In the process of determining this threshold T, C
The PU 7 obtains the gradation value of each pixel of the image data expanded in the RAM 9, creates a histogram for each gradation value, and determines the threshold T based on the histogram.

Specifically, the operation is as follows. When the histogram has two distinct peaks on the black side and the white side as shown in FIG. 3A, the grayscale value that is the minimum value between the two peaks may be determined as the threshold value T. However, actually, the peak on the white side in FIG. 3A does not clearly appear due to an image of a scratch or dirt on the gravure plate surface or dust attached on the gravure plate surface, and is schematically shown in FIG. 3B. It may be like this. However, even in the case of the histogram as shown in FIG. 3B by various experiments, it is preferable that the tone value at the position where the frequency suddenly decreases after passing the black side peak and becomes gentle again is set as the threshold value T. It was confirmed that binarization was possible. The method of determining such a threshold is shown in FIG.
Naturally, it is also effective in the case shown in FIG. In FIG. 3, the gradation value is 2 from all black to all white.
It has 56 gradations.

Then, based on the histogram, the CPU 7 determines that the frequency suddenly decreases after passing the peak on the black side,
A process of setting a tone value at a position where the tone becomes gentle again as a threshold value T is performed. As a result, in the case of FIG. 3B, the gradation value indicated by T in the figure is determined as the threshold value. In addition,
It has been confirmed that the threshold value T is a value between 80 and 120 in most cases when the gradation value is 256 gradations from all black to all white.

When the threshold T is determined as described above, C
The PU 7 binarizes the image data developed in the RAM 9 using the threshold T. Then, the CPU 7 displays the binarized image on the monitor 4. At this time, it is natural that the binarized image data may be stored in the storage unit 10. The above is the binarization processing in step S2 in FIG.

FIG. 5 shows an example of an image obtained by the binarization processing in step S2. FIG. 5 shows an example of an image obtained by binarizing the image shown in FIG. 4 captured in step S1.

Incidentally, as described above, the gravure plate surface of the cylinder may have scratches or dirt, or dust may adhere thereto, and these scratches, dirt, or dust portions are the binary values of step S2. Although a black image is obtained by the conversion process, an image based on such scratches, dirt, dust, and the like is unnecessary when measuring the cell area and needs to be removed. For this purpose, the noise removal processing in step S3 in FIG. 2 is performed. When the binarization processing is completed, the CPU 7 performs noise removal on the obtained binary image by, for example, a morphological method. The resulting image is displayed on the monitor 4. At this time, it is natural that the binarized image data may be stored in the storage unit 10. This is the noise removal processing in step S3. Since the morphological processing is well known, a detailed description thereof will be omitted.

FIG. 6 shows an example of an image obtained by the noise removal processing in step S3. In the binarized image shown in FIG. 5, white dots appear in the cell portion, and black portions also appear in the bank portion in a dot or straight line. This is a noise component based on scratches, dirt, or attached dust on the gravure plate surface. In FIG. 6, the horizontal direction of the image is defined as the x-axis, and the vertical direction is defined as the y-axis. This is the same in the following.

When this image is subjected to noise removal processing by a morphological method in step S3,
The image shown in FIG. 6 is obtained, and fine black lines and small dots based on scratches, dirt, or attached dust can be removed.

In the above description, morphology is used as a noise removing method. However, the method is not limited to the morphological method, and any method capable of removing noise components may be used.

When the CPU 7 completes the noise removal processing in step S3, the CPU 7 next performs thinning processing on the noise-removed image, and further removes noise line segments (step S4). This thinning process is a process for extracting the center line segment of the bank shown in white in FIG. 6, and a method widely known as the thinning process may be used. The following process may be performed.
First, a white pixel is searched for in the x direction, and if white pixels are continuous, the process of obtaining the center position is performed for all y pixels.
At the position of. FIG. 7 shows an example. FIG. 7 illustrates a pixel portion at a certain y position, where a hatched rectangle indicates a black pixel, and a non-hatched rectangle indicates a white pixel. Therefore, in the case shown in FIG. 7, the center of the white pixel with a circle is obtained as the center position. Then, when the center position of the white pixel is obtained over the entire image, next, the center positions that are close to each other within a predetermined distance are connected and line-divided, and each center line segment is used as vector data. register. For example, if the position indicated by the black point in FIG. 8 is obtained as the center position of the white pixel, L,
Two center lines as shown in L 'are obtained, and these center lines L and L' are converted into vector data. Therefore, the vector data of each center line is composed of the number of center positions and the coordinate values of those center positions.

In this manner, the center line segment of the bank can be extracted. However, it is not desirable that the center line segment having a small number of center positions is the bank. , Is desirably removed as a noise line segment. Further, even if the end point of at least one of the center line segments is interrupted at the left and right ends of the image, it is unnecessary in the subsequent processing, and is deleted at this point as a noise line segment. This is because, for the former, the number of center positions of the vector data of each center line is searched, and if the number of center positions is equal to or less than a predetermined number, those vector data may be deleted. Finds the coordinates of the center position of the vector data of each center line,
If the coordinate value of the center position of the end point is within a predetermined distance from the left and right ends of the image, those vector data may be deleted.

FIG. 9 shows an example of the image of the center line of the bank obtained in this way. FIG. 9 is an image obtained by performing a thinning process on the image shown in FIG. 6 and removing a noise line segment.

After obtaining the center line of the bank in this way, the CPU 7 performs a process of creating a straight line (hereinafter, referred to as an interpolation straight line) for interpolating the broken portion of the bank (step S5). This processing is performed by creating a straight line that connects the center line segments whose end points are closest to each other and whose directions at the end points face each other within a predetermined angle range. Here, the direction at the end point of the center line segment is
The direction at the end of the center line segment, for example, an end point,
The direction may be the direction of a line connecting the center position immediately next to the end point. Therefore, for example, two center line segments L, shown in FIG.
The directions at both ends of L 'are the directions indicated by the arrows in the figure.

[0040] Now, Paying attention to the center line shown by L ₁ in FIG. 9, the end points of the other center line segment that is closest to the end point P ₁ of the lower side of the center line is at L ₂ in FIG. The upper end point P ₂ of the indicated center line segment, and the center line segment L ₁
The direction of the end point P ₁ of the case where the direction of the center line end point P ₂ of the L ₂ is opposed at a predetermined angular range, CPU 7 and the end point P ₁ of the center line L _1, the center line L
To create a straight line connecting ₂ of the end point P _2. This is the interpolation straight line. The same applies to other center line segments. When there is no opposing end point as indicated by P ₅ in FIG. 9, generating an interpolation straight line in the direction of the end point P _5.
By the above processing, the center line shown in FIG.
An interpolation straight line as shown in FIG. 10 is created.

Next, the CPU 7 draws an image of the interpolated straight line created in step S5, assigns a predetermined width to the interpolated straight line, and outputs an image of the interrupted bank portion (hereinafter, referred to as an image).
Simply called a bank image), and then the bank image,
The image obtained as a result of the processing in step S3 is combined with black and white colors. This is the process of step S6. Here, the reason why the width of the interpolation straight line is set when the bank image is created is that the image of the interpolation straight line created in step S5 has a narrow width, so that the width suitable for the bank is provided. Therefore, if an appropriate line width is defined for the interpolation straight line in step S5, it is not necessary to perform the width in step S6.

When the bank image is combined with the image obtained as a result of the processing in step S3, the bank image may be combined as it is when the background is drawn in black and the bank is drawn in white. However, if the background is drawn white and the bank is drawn in black, it is necessary to invert the black and white of the bank image and combine them. Accordingly, if the image obtained as a result of the processing in step S3 is as shown in FIG. 6 and the bank image is as shown in FIG.
A composite image as shown in FIG.

Next, the CPU 7 performs a contour extraction process on the composite image obtained in step S6 (step S6).
7). For this, a known method of sequentially following the boundary between a white pixel and a black pixel may be used. Then, the route that has been followed is generated as vector data.

By the contour extraction processing in step S7, a contour as shown in FIG. 13 is extracted from the binary image shown in FIG. When the CPU 7 completes the contour extraction processing in step S7, the CPU 7 displays a contour image as shown in FIG. At this time, it is natural that the image data of the contour may be stored in the storage unit 10. Note that, in this contour extraction processing, C
When sequentially tracing the boundary between the white pixel and the black pixel, the PU 7 recognizes whether the pixel on the left side of the boundary is black or white, and adds the information as an attribute of the outline vector data. This attribute information is used in the process of removing incomplete contours later. Therefore, the vector data of the outline indicated by R ₁ in FIG. 13, black information is added as an attribute, information of white is added as attribute to the vector data of the contour shown by R _2.

Next, the incomplete contour is removed (step S).
8). In order to obtain the cell area and volume, step S
It is obvious that the contour extracted by the contour extraction of No. 7 must be continuous from at least one end of the image to the other end, and must not be interrupted in the middle. For example, in the case of FIG. 13, one outline needs to be continuous at least from the upper end to the lower end of the figure.

Therefore, as a first stage of the incomplete contour removal, a contour that is not continuous from one end to the other end of the image is removed as an incomplete contour. In the process of the first stage, in the case of the image shown in FIG. 13, the outline that is in contact with the left end or the right end of the image may be removed. Specifically, as shown in FIG. 13, assuming that the size of the image in the x direction is x _max , the CPU 7 recognizes the size x _max of the image, and checks the vector data of each contour. Then, the vector data including the value of x = 0 or x = y _max may be regarded as an incomplete contour, and the process of deleting the vector data of the contour may be performed.

By the above-described first-stage processing, the outline shown by B in FIG. 13 is removed in FIG. 13, and an image composed of four outlines shown by R _{1 to} R ₄ is obtained. Naturally, these four contours R _{1 to} R ₄ are continuous from the upper end to the lower end in the y direction of the image.

By the way, in the image shown in FIG. 12, since the channel generation cells are connected in the vertical direction, the contours of the channel generation cells in a certain row are arranged two in the horizontal direction. In this way, two contours form a pair.

Therefore, as a second stage of the process of removing incomplete contours, a search is made as to whether there is any unpaired contour among the contours left in the first stage. , The contour is removed as an incomplete contour. In performing this process, the CPU 7 refers to the attribute of the vector data of each contour, and pairs the vector data of the contour with the attribute white and the vector data of the contour with the attribute black right to the right of the contour. recognize. For example, at the position of y = 0, the vector data of the outline is searched from the smaller value of x to the larger value of x, and the vector data of the outline having the white attribute and the vector of the outline having the black attribute immediately to the right are obtained. What is necessary is just to perform the process which pairs with data.

By the process of the second stage, in the case shown in FIG. 13, only the vector data of the contours of R ₂ and R ₃ are left as complete contours as shown in FIG. And
CPU7 recognizes the contour of the R ₂ and R ₃ are paired. As described above, the incomplete contour is removed by the two-stage processing.

When the processing in step S8 ends, the CPU
7 performs the cell division processing in step S9. This process is a process of dividing cells connected by channels into individual cells. For gravure version management,
Since it is important to manage the area and volume of each cell, this cell division processing is performed.

Now, in order to divide the cells one by one,
What is necessary is just to divide at the position of a channel. For this purpose, in this processing, first, the number of pixels in the x direction between the two contours recognized as a pair in step S8 is obtained at all pixel positions in the y direction, and a data string of the number of pixels between the two contours is obtained. A smoothing process is performed to recognize the position of the y coordinate having the minimum value as the position of the channel, to provide a gap having a width of 1 pixel at the position of the channel, and to define a straight line between the contours forming a pair on both sides of the gap. To perform cell division.

Here, the reason why the smoothing process is performed on the data string of the number of pixels between the two contours is as follows.
That is, for example, when a data sequence in the vicinity of the channel of the number of pixels between the contours forming a pair is represented by a graph, a line graph shown in FIG. Therefore, a smoothing process is performed on these data strings, and the y position at which the number of pixels is minimized is obtained as in a smooth curve shown in FIG.

When the above-described cell division in step S9 is performed on the image shown in FIG. 14, an image as shown in FIG. 16 is obtained. In FIG. 16, cells arranged in the vertical direction in the figure are divided at channel positions.

Next, the CPU 7 removes incomplete cells from the divided cells (step S10). This may be performed by a process of extracting a closed region surrounded by black pixels in white pixels, and can be performed by a known method. For example, in FIG. 17 (a), when a shaded rectangle is a black pixel and a rectangle not shaded is a white pixel, if there is a black pixel P adjacent to the white pixel, the black pixel P is indicated by an arrow. As shown in the figure, a method may be used in which a black pixel adjacent to a white pixel is traced, and only when the pixel returns to the first black pixel is recognized as a closed region, and the traced route is generated as vector data. .

Therefore, in the area of the black pixel as shown in FIG. 17B, even if the black pixel adjacent to the white pixel is traced from the black pixel shown by P in the figure as shown by the solid arrow, Even if the tracing is performed as indicated by the arrow, the end reaches the end of the binary image data, and since there is no image data thereafter, it is not possible to return to the first black pixel P, so that it is not recognized as a closed area. . According to this, for example, as shown by D in FIG. 16, an image in which the entire closed region is not completely included in the image is recognized as an incomplete cell and is removed. This is useful. This is because it is meaningless to measure the area and volume of the region shown by D in FIG. Accordingly, the image shown in FIG.
When performing processing of 0, only one closed area indicated by Q ₁ in FIG. 16 is extracted as a complete cell.

Next, the CPU 7 measures the area and volume of the complete cell extracted in step S10,
The result is displayed (step S11). The area of each cell may be obtained by counting the number of pixels in a closed region of a complete cell, and using the number of pixels as a pseudo cell area. It is also possible to obtain an average area value, a maximum area value, a minimum area value, a standard deviation, and the like of the extracted cells.

The volume of each cell is obtained as follows. For example, in the example described so far, the channel generating cell is engraved in the y direction, but the engraving needle angle at the time of engraving is
It is known in advance from the input engraving condition data. The relationship between the width of the cell in the y direction and the cross-sectional shape in the depth direction of the cell at that time is also known. Thus, for example the shape of one cell is like shown in FIG. 18 (a), y _n
Assuming that the width in the x direction at the position of is δ,
The cross-sectional shape in the depth direction of the cell at the position _n is shown in FIG.
It is known in advance that it is as shown in (b), and the area of the cross section is also known. Accordingly, at all the y-coordinate positions of the cell, the area of the cross section of the cell at the y-coordinate position is obtained from the width in the x direction, and the sum of the obtained cross-sectional areas is calculated to obtain the volume of the cell. Can be. Here, the number of pixels in the cross-sectional shape may be used for the area of the cross-section of the cell.

In order to obtain the cell volume by such an operation, a table in which the area of the cross section with respect to the width of the extracted cell in the x direction is prepared in advance for each engraving needle angle, and engraving condition data is prepared. Select the table to be used based on the angle of the engraving needle, and use the table to determine the area of the cross section of the cell at the y coordinate position from the width in the x direction at all y coordinate positions of each cell. What is necessary is just to calculate the sum of the areas of the obtained cell cross sections.

The CPU 7 measures the cell volume by performing the above processing. Here, as described above, the CPU 7
It is natural that a table in which the area of the cross section of the cell with respect to the width of the cell in the x direction is written for each angle of the engraving needle is registered. As for the volume, similarly to the area, an average volume value, a maximum area value, a minimum area value, a standard deviation, and the like of the volume value may be obtained.

As described above, the volume of each cell can be obtained, but the cell volume obtained above does not take into account the degree of wear of the engraving needle. Therefore, a method that takes into account the degree of wear of the engraving needle will be described below. When calculating the cell volume, there are a simple method and a more accurate method in which the degree of wear of the engraving needle is taken into consideration.

First, a simple method will be described. In this method, a table of correction coefficients relating to the degree of wear of the engraving needle is obtained in advance by experiments or the like according to the engraving time or cell shape. Then, based on the engraving needle abrasion degree of the engraving condition data, an engraving needle abrasion degree correction coefficient is obtained by referring to a correction coefficient table relating to the engraving needle abrasion degree. Further, the cell volume obtained above is multiplied by the engraving needle wear degree correction coefficient obtained from the table. This makes it possible to calculate the cell volume in consideration of the engraving needle angle and the engraving needle wear degree of the engraving condition data.

Next, a more accurate method is as follows. In this method, a three-dimensional table in which three numerical values of the degree of wear of the engraving needle, the width of the cell in the x direction, and the area of the cross section of the cell are prepared in advance for each of various engraving needle angles. deep. In this three-dimensional table, the numerical value of the area of the cross section of the cell is present at two-dimensional coordinates (lattice points) determined by two numerical values of the numerical value of the degree of wear of the engraving needle and the numerical value of the width in the x direction. Such a three-dimensional table can be created, for example, by actually measuring channel engraved cells engraved at various engraving needle angles and at various engraving needle abrasion degrees.

Then, based on the engraving needle angle obtained from the engraving condition data, a corresponding three-dimensional table is selected from a plurality of tables, and the numerical value of the engraving needle wear degree obtained from the engraving condition data and the x direction From the numerical value of the cell width, the numerical value of the cross-sectional area of the cell is obtained with reference to the selected three-dimensional table. The numerical value of the area of the cross section of the cell naturally takes into account the engraving condition data. Then, using the numerical value of the area of the cross section of the cell, the cell volume is calculated as described above.

As described above, when the number of dimensions increases, the number of data to be recorded in the table increases, which may cause a problem in data creation and storage of the table. There is a method of reducing the data amount by setting each numerical value of the cell width in the x direction and the area of the cell cross section as a discrete numerical value. The engraving needle angle is a geometric shape and is usually a discrete number. When obtaining the area of the cross section of the cell corresponding to the numerical value not recorded in the table, the area can be obtained by applying a known interpolation method.

When the area and volume of each extracted cell are measured as described above, the CPU 7 displays the measurement result on the monitor 4. At this time, the CPU 7 displays not only the area value and the volume value of the measurement result or their average value, the maximum value, the minimum value, and the standard deviation, but also the cell whose area and volume are measured. When displaying this cell, the CPU 7
Assigns a serial number to each cell. This serial number is
For example, it may be attached based on the position of the center of gravity of the cell. Here high ranking enough display position is left in the monitor 4, when the display position in the case of a lateral position are the same is assumed a higher rank as the left, to the extracted cell is shown by Q ₁ in FIG. 16 On the other hand, as shown in FIG. 19, the serial number 1 is assigned.

Although not shown in FIG. 19, the CPU 7
Also displays an area value, a volume value, and the like obtained by measurement on a screen displaying cells as shown in FIG. FIG. 20 shows an example of the display. In FIG. 20, "cell 1" indicates a cell numbered serially. The same applies to other cases. In practice, it goes without saying that the measured value is displayed in the "xxx" part of FIG. In FIG. 20, the area value, the volume value, and the like are displayed as numerical values, but may be displayed as an appropriate graph such as a bar graph. With such a display, the operator can easily grasp the degree of variation of the cell area and volume, and can judge the quality of the gravure plate.

The CPU 7 stores these measurement results in the storage unit 10. In the process of step S11, the measurement result may be automatically printed out by the printer 5.

As described above, according to this image processing system, after taking in the image picked up by the microscope 1, all the processes are performed fully automatically, so that the burden on the operator is very light. is there. Also, as mentioned above,
Since this image processing system can be configured with the microscope 1 and a personal computer or a workstation, it can be configured in a small size, and can easily measure cells formed on the gravure plate surface at a printing site. Further, the area and volume of the cell can be measured with high accuracy even for a channel generating cell having a broken bank.

In the above embodiment, the area of each cell is represented by the number of pixels in the cell.
If the actual size value of the display area of one pixel on the monitor 4 can be known from the magnification or the like, the measured area value or the like may be displayed as the actual size value.

The embodiments of the image processing method and the image processing system according to the present invention have been described above.
A usage form of such an image processing system will be described. FIG. 21 is a diagram illustrating an embodiment of a gravure plate engraving system to which the above-described image processing system is applied. In FIG.
3a, 23b, 23c, 23d, 23e and 23f are engraving heads of the gravure plate engraving device 21, 24 is a cell volume measuring device, 25 is an imaging means, 26 is an input unit including a keyboard and a mouse, and 27 is a display and a printer. An output unit 28 is a processing unit of the cell volume measuring device 24. The processing unit 28 is provided with an image input unit 30, an engraving condition input unit 31, and a cell volume calculation unit 32. 2
Reference numeral 9 denotes a storage unit of the cell volume measuring device 24, in which image data 33 and engraving condition data 34 are stored.

The gravure plate engraving device 21 is a gravure plate 22
And has a structure that rotates around its axis. In the example shown in FIG. 21, the engraving head 23 of the gravure plate engraving device 21 includes six pieces, and has a structure that moves while performing engraving in a direction parallel to the axial direction of the gravure plate 22. That is, the gravure plate 22 is rotated around its axis, and the engraving heads 23a to 23f are moved in the axial direction of the gravure plate 22, thereby engraving the entire surface of the gravure plate 22.

The imaging means 25 is the same as the microscope 1 in FIG. The cell volume measuring device 24 includes a processing unit 28 and a storage unit 2
9 is provided.

The processing unit 28 includes an image input unit 30 for inputting an image of a cell engraved on the gravure plate 22 picked up by the image pickup unit 25 by the image input unit 30, and an engraving condition input for inputting engraving conditions. Means 31 and a cell volume calculating means 32 for performing the above-described image processing on the image fetched from the image input means 30 and calculating the volume of the cell extracted from the image. Therefore, the image input means 30 corresponds to the video capture board 6 shown in FIG. Further, the cell volume calculation means 32 calculates the volume of the cell extracted from the image. As a method of calculating the volume of the cell, the simple method described above may be used, or a more accurate method may be used. May be used. As the engraving conditions input from the engraving condition input unit 31, the use time corresponding to the engraving needle angle and the engraving needle wear degree is input in the same manner as described above. The engraving conditions are input for each of the engraving heads 23a to 23f.

The storage unit 29 has image data 33 in which an image captured by the image input means 30 is stored, and engraving condition data 34 in which engraving conditions input from the engraving condition input means 31 are stored.

FIG. 22 is a graph showing engraving characteristics when engraving a gravure plate. In FIG. 22, the output value on the horizontal axis is the value of data output to the gravure plate engraving device 21, and the cell volume on the vertical axis is the volume of cells formed on the plate surface of the gravure plate 22. Also,
The arrow shown in FIG. 22 indicates that the engraving characteristic A of the test cut value obtained by the test engraving is corrected to the engraving characteristic B of an appropriate reference value. That is, as described later, first, a test cut is performed, and the volume of the cell formed by the test cut is obtained by the above-described method, and the engraving characteristic obtained from the relationship between the output value at the test cut and the cell volume is obtained. In the case where it is as shown in a, the engraving conditions for correcting the engraving characteristics to the engraving characteristics of the reference value shown in b in the figure are obtained, and the engraving conditions for obtaining the appropriate engraving characteristics are changed for all the engraving heads 23a. 23b, 23c, 23d, 23e, 23f
The main engraving is performed by the gravure plate engraving device 21.

FIG. 23A shows a gravure plate 22 on which test engraving and main engraving have been performed, and FIG. 23B is a view showing a developed view of the peripheral surface of the gravure plate 22. FIG.
As shown in (B), the test engraving is performed for each engraving head 23a, 23b, 23c, 23d,
23e and 23f. In addition, the main sculpture of the picture is also performed by each of the sculpture heads 23a, 23b, 23c, 23d, and 23.
e, 23f. FIG. 23B shows the engraving heads 23a, 23b, 23c, 23d, 23e,
Each engraving area performed by 3f is shown.

As shown in FIG. 21, the fact that the gravure plate engraving device 21 has six engraving heads 23a to 23f means that the gravure plate engraving device 21 has six engraving heads as compared with the case of one engraving head. This means that the engraving of the entire plate can be completed six times faster. On the other hand, compared to the case of one engraving head, 6 engraving heads are used.
It is necessary to appropriately set engraving conditions for the double number of engraving heads 23a to 23f. This is because when engraving is performed using different engraving heads, a slight difference in engraving characteristics of the engraving head results in a difference in color tone between the patterns of the printed matter, and is conspicuous because the patterns are adjacent to each other. Therefore, it is necessary to set all appropriate engraving conditions for the engraving heads 23a to 23f in order to ensure print quality.

The differences between the engraving characteristics of the engraving heads 23a to 23f include the differences in the characteristics of the electric and mechanical systems that drive the engraving needles of the engraving heads 23a to 23f, as well as the differences in the sharpness and shape of the engraving needles. This is caused by the progress of wear of the engraving needle. In order to match the engraving characteristics as shown in the engraving characteristics A in FIG. 22 with the appropriate engraving characteristics B in FIG. 22, it is ideal to equalize all the items caused by the differences in the characteristics. But it is extremely difficult.

Therefore, normally, an appropriate conversion table that determines the input / output characteristics of image data is created, image data for gravure printing plate creation is input to the conversion table, and the output image data converted by the conversion table is converted. The main image engraving is performed based on the corrected image data. This conversion table is shown in FIG.
Naturally, it is possible to obtain corrected image data that achieves appropriate engraving characteristics as shown in FIG. However, differences in sculpture characteristics are caused by various factors,
In some cases, it is not possible to set proper engraving conditions only by setting the conversion table. For example, there are cases where the wear of the engraving needle is remarkably progressing, and the characteristics of the electric system and the mechanical system for driving the engraving needle have changed greatly. In that case, if you set the engraving conditions, including re-grinding or replacing the engraving needle, re-adjusting the electric system and mechanical system, replacing the entire engraving head itself with a good engraving head, etc. Good.

Hereinafter, the process from the test engraving to the main engraving in the gravure engraving system shown in FIG. 21 will be described together with the gravure engraving method. FIG.
5 is a flowchart showing a process of the processing.

First, test engraving is performed (step S2)
0). This test sculpture is based on the reference data.
3 (B). Thus, a test cell is formed on the gravure plate surface 22. Here, the reference data for performing the test engraving includes not only data of a specific print density but also data of various print densities. For example, scale data of a predetermined density of several steps (about 2 to 10 steps) including a maximum print density to a minimum print density may be used. This scale data may be a patch having a small area at each density, or at least one cell at each density. In the latter case, the rate at which the effective printing area of the gravure plate 22 is limited by the test engraving may be small.

Next, the cell area and volume of the test engraved test cell are measured (step S21). The measurement of the cell area and the volume is performed by the engraving heads 23a to 23a.
The test is performed for test cells corresponding to all density scale data engraved with f. This means that the test cell is
5, the image is taken in by the image input means 30, and an instruction to start the volume measurement may be given to the cell volume calculation means 32.

As a result, the cell volume calculation means 32
By the above-described processing, a test cell is extracted from the captured image, and the area and volume of the extracted cell are calculated. As described above, the cell volume may be calculated by the simple method described above or by a more accurate method.

Then, the cell volume calculating means 32 calculates the area and volume of the cell extracted from the image, and outputs the result to a display and / or a printer as the output unit 27.

After measuring the area and volume of the test cell in this way, the measured volume data is compared with the reference volume data (step S22) to determine whether it is necessary to correct the engraving conditions. (Step S2
3). Here, the reference volume data is prepared in advance by collecting input data and measured volume data of cells for a good printed matter, that is, collecting engraving characteristics,
A predetermined allowable range is set for each reference value of the reference data. Therefore, in the determination of step S23, it is determined that correction of the engraving condition is unnecessary if the measured volume data is within the allowable range, and that the engraving condition is required if the measured volume data is out of the allowable range. .

If it is determined in step S23 that correction is necessary, the engraving conditions are corrected in step S24.

The engraving conditions are corrected by correcting the conversion table and setting appropriate data so that the measured volume data matches the reference volume data.
For example, the measurement volume data is a discrete measurement value given to each reference value of the discrete reference data used for performing the test engraving. Based on this measurement value, first, a curve is obtained. An approximation is performed to obtain continuous measured volume data. That is, data output to the gravure plate engraving device 21 (this is reference data input for engraving the test cell described above when viewed from the gravure plate engraving device 21 side). ) And the volume of the cells formed on the plate surface of the gravure plate 22 (this corresponds to the value on the vertical axis in FIG. 22), that is, the engraving characteristic is obtained as a continuous function. Then, in this engraving characteristic, input data for obtaining reference volume data for each input data is obtained.

The technique is roughly as follows. For example,
Now, gravure plate engraving equipment to engrave test cells
21 is supplied as I₁ , I_Two , I_Three 3
There are two reference data, which engrave the test cell
After measuring the volume of those test cells,
V₁ , V_Two , V_Three And the measurement points
After approximating the curve, engraving as shown by C in FIG.
It is assumed that the engraving characteristics are obtained. In addition, the standard engraving characteristics
As shown by d in FIG.₁ , I_Two, I_Three 
The appropriate cell volume for the three reference data is
Re V_R1, V_R2, V _R3And

At this time, based on the engraving characteristics of the actual engraving head indicated by C in FIG. 25A and the standard engraving characteristics indicated by B, when the data I ₁ is given, the cell having a volume of V _R1 is obtained. To sculpt, I ₁ is _first converted to I ₁ ′
It can be seen that the cell may be engraved based on the value of I ₁ ′. Similarly, the volume when given the data that I ₂ is engraving a cell of V _R2 is "converted into its I ₂ 'once I ₂ to I ₂ may be engraved cells by the value of , in the volume when given the data that I ₃ is engraving a cell of V _R3 is "converted into its I ₃ 'once I ₃ a I ₃ that may be engraved cells by the value of I understand.

Therefore, a graph showing the input / output characteristics of the data value is shown.
On the rough, I₁ The value of I₁′ _Two The value of I_Two′
I_Three The value of I_Three′ Are plotted,
Connect those points smoothly and show by E in FIG. 25 (B).
Create a conversion table with such input / output characteristics.
You.

Using such a conversion table, the value of the image data having the picture to be engraved is corrected and supplied to the gravure engraving apparatus 21 to engrave the cell. _The data having a value of ₁ is corrected to I ₁ ′ by the conversion table and supplied to the gravure engraving apparatus 21, so that a cell having a volume _VR1 is engraved. The same applies to data of other values, and as a result, a reference engraving characteristic as indicated by d in FIG. 25A can be obtained.

After correcting the engraving conditions in step S24 in this way, it is determined whether or not it is necessary to perform re-measurement (step S25). If the correction amount of the engraving condition in Step S24 is small, it is generally not necessary to perform re-measurement. Therefore, the main engraving in Step S26 may be performed immediately after correcting the engraving condition. However, when the correction amount of the engraving condition is not small, it is necessary to perform the re-measurement, so that the process returns to step S20 and the above-described process is repeated. When performing test engraving by returning from step S25 to step S20, test engraving is performed by correcting the reference data using the conversion table created in the engraving condition correction process of step S24.

If it is determined in step S23 that no correction is necessary, or if it is determined in step S25 that re-measurement is not necessary, engraving is performed (step S26). If it is determined in step S23 that the correction is not necessary, the image data of the pattern to be subjected to the gravure printing may be directly supplied to the gravure plate engraving device 21 for engraving. If it is determined in step S25 that re-measurement is not necessary, the value of the image data of the pattern to be subjected to gravure printing is corrected by the conversion table created in step S24 and supplied to the gravure plate engraving apparatus 21. And then do the sculpture. Then, by this main engraving, engraving is performed based on appropriate engraving conditions, and cells are formed on the plate surface of the gravure plate 22 to complete the gravure plate.

The above-described processing for engraving condition correction and comparison determination can be realized by an information processing device such as a personal computer having software for that purpose. Description is omitted.

As described above, according to this gravure plate engraving system, test engraving is performed prior to main engraving, the volume of the test cell formed by the test engraving is measured, and the current engraving is performed based on the volume. It is possible to determine whether the conditions are appropriate or not, and to correct the engraving conditions if not, so that a cell having a good volume can be formed regardless of the degree of wear of the engraving needle. Thus, gravure printing with good quality can be performed.

[Brief description of the drawings]

FIG. 1 is a diagram showing an embodiment of an image processing system according to the present invention.

FIG. 2 is a diagram showing a flow of processing in the image processing system shown in FIG.

FIG. 3 is a diagram for explaining a method of determining a threshold value in a binarization process in step S2 of FIG. 2;

FIG. 4 is a diagram showing an example of an image taken from a microscope 1.

FIG. 5 is a diagram showing an example of an image obtained by binarizing the image shown in FIG. 4;

FIG. 6 is a diagram illustrating an example of an image when noise is removed from the image illustrated in FIG. 4;

FIG. 7 is a diagram for explaining a thinning process in step S4 of FIG. 2;

FIG. 8 is a diagram for explaining a center line segment obtained by the thinning process in step S4 of FIG. 2;

9 is a diagram illustrating an example of an image of a center line of a bank obtained as a result of the processing in step S4 of FIG. 2;

FIG. 10 shows a step S5 for the center line shown in FIG. 9;
FIG. 8 is a diagram for explaining an interpolation straight line created by performing the processing of FIG.

FIG. 11 is a diagram illustrating an example of a bank image.

FIG. 12 is a diagram illustrating an example of a composite image obtained as a result of step S6 in FIG. 2;

FIG. 13 is a diagram showing an example of an image in a case where an outline extraction process has been performed on the image shown in FIG. 12;

FIG. 14 is a diagram for explaining the process of removing incomplete contours in step S8 of FIG. 2;

FIG. 15 is a diagram illustrating the cell division processing in step S9 of FIG. 2;

FIG. 16 is a diagram illustrating the cell division processing in step S9 of FIG. 2;

FIG. 17 is a diagram for explaining the process of removing incomplete cells in step S10 of FIG. 2;

FIG. 18 is a diagram for explaining processing of cell volume measurement in step S11 of FIG. 2;

FIG. 19 is a diagram showing an example of a display mode of a cell in step S11 of FIG. 2;

FIG. 20 is a diagram showing an example of display of measured area and volume.

FIG. 21 is a diagram showing an embodiment of a gravure printing engraving system according to the present invention.

FIG. 22 is a graph showing engraving characteristics when engraving a gravure plate.

FIG. 23 is a view showing a gravure plate on which test engraving and main engraving have been performed, and a development view of a peripheral surface of the gravure plate.

FIG. 24 is a flowchart showing processing from test engraving to main engraving in the gravure engraving system.

FIG. 25 is a diagram for explaining a process of correcting an engraving condition in step S24 of FIG. 24;

FIG. 26 is a diagram illustrating an example of a channel generation cell.

FIG. 27 is a diagram illustrating an example of a channel generation cell in which a bank portion has been removed.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Microscope 2 ... Image processing apparatus 3 ... Input part 4 ... Monitor 5 ... Printer 6 ... Video capture board 7 ... CPU 8 ... ROM 9 ... RAM 10 ... Storage part 21 ... Gravure plate engraving device 22 ... Gravure plate 23a, 23b, 23c, 23d, 23e, 23f: engraving head, 24: cell volume measuring device, 25: imaging unit 26, input unit 27, output unit 28, processing unit 29, storage unit 30, image input unit 31, engraving condition input unit 32 ... Cell volume calculation means 33 ... Image data 34 ... Engraving condition data

Claims (6)

[Claims] An image of a channel generation cell is captured and binarized, and a center line segment of a bank portion of the channel generation cell is extracted based on the binarized image. An interpolation straight line is obtained by creating a straight line connecting the center line segments where the distance is closest and the direction at the end point is opposite within a predetermined angle range, and an image of a broken bank portion is created based on the interpolation straight line. Then, the image of the broken bank portion and the binarized image are combined by combining black and white colors, an outline forming a channel generation cell is extracted from the synthesized image, and a cell is formed at the channel portion of the outline. Is divided into
An image processing method comprising measuring at least an area and a volume of the divided cells. 2. An image pickup means for obtaining an image of a channel generating cell engraved on a gravure plate surface of a cylinder, an image taken from the image pickup means, binarized, and a channel is formed based on the binarized image. The center line segment of the bank portion of the generated cell is extracted, and a straight line connecting the center line segments in which the distance between the end points of the center line segments is the shortest and the directions at the end points oppose each other within a predetermined angle range is created. To obtain the interpolation straight line,
An image of the interrupted bank portion is created based on the interpolation straight line, and the image of the interrupted bank portion and the binarized image are combined by combining black and white colors, and a channel generation cell is generated from the combined image. An image processing system, comprising: extracting an outline that forms, dividing a cell at a channel portion of the outline, and measuring at least an area and a volume of the divided cell. 3. An image pickup means for picking up an image of a channel generating cell formed on a gravure printing plate to obtain image data, and an engraving condition inputting means for inputting engraving conditions when forming the channel generating cell and obtaining engraving condition data. Taking the image from the imaging means, binarizing it,
The center line segment of the bank portion of the channel generation cell is extracted based on the binarized image, and the center line where the distance between the end points of the center line segment is the shortest and the direction at the end point is opposed within a predetermined angle range. An interpolation line is obtained by creating a straight line connecting the minutes, an image of the interrupted bank portion is created based on the interpolation straight line, and the image of the interrupted bank portion and the binarized image are converted to black and white. Cell volume calculating means for combining colors and synthesizing them, extracting a contour forming a channel generation cell from the synthesized image, dividing the cell at a channel portion of the contour, and measuring a volume of the divided cell. Gravure plate engraving system. 4. The gravure plate according to claim 3, wherein said engraving condition data is an engraving needle angle which is an angle of an engraving needle attached to an engraving head, and an engraving needle wear degree of said engraving needle. Engraving system. 5. A test engraving process for forming a test engraving channel generating cell on a gravure plate by performing test engraving based on reference data, and taking an image obtained by capturing the test engraving channel generating cell to obtain a binary image. The center line segment of the bank portion of the channel generating cell is extracted based on the binarized image, and the distance between the end points of the center line segments is the shortest, and the directions at the end points oppose each other within a predetermined angle range. A straight line connecting the center line segments is created to obtain an interpolation straight line, an image of the broken bank portion is created based on the interpolation straight line, and the image of the broken bank portion and the binarized image are formed. Are combined with black and white colors, the contour forming the channel generation cell is extracted from the combined image, the cell is divided at the channel portion of the contour, and the volume is calculated for the divided cell. A cell volume calculation process, a measurement volume data obtained in the cell volume measurement process, a comparison determination process of comparing the reference volume data to determine the necessity of correcting the engraving condition, and a correction required by the comparison determination process. Engraving condition correction process of correcting the engraving condition when the determination is made, and, when the necessity of correction is determined by the comparison determination process, main engraving is performed based on the engraving condition and a channel is formed on the gravure plate surface. A gravure engraving method, comprising: a main engraving process of forming a generating cell to complete a gravure plate. 6. The gravure engraving method according to claim 5, wherein when the engraving condition correcting step is performed, the test engraving step, the cell volume measuring step, and the comparing and judging step are repeated.