JP2009100078A - Image data analysis device, image data analysis method, and program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To remove halftone dots of an image without vignetting the contour of a subject or a character. <P>SOLUTION: Spaces between halftone dots are divided like am equal-distance grid to generate halftone dot cells surrounding individual halftone dots, and generated halftone dot cells are grouped by a prescribed number to generate fundamental grids. It is checked whether a halftone dot is formed in each of halftone dot cells constituting a fundamental grid or not to acquire halftone dot arrangement in the position of this fundamental grid. Since it is considered that halftone dot cells wherein halftone dots are formed are located on the inside of the contour and halftone dot cells wherein halftone dots are not formed are located on the outside of the contour, it is possible to estimate the contour line in he fundamental grid on the basis of the acquired halftone dot arrangement. Since it is possible to extract the entire contour in image data by estimating the contour in positions of respective fundamental grids in accordance with this method, it is possible that only the inside of the contour is vignetted to remove halftone dots without vignetting the contour. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a technique for removing halftone dots used to represent the shading of an image from the image.

In the printing industry and the like, a method of expressing the density of an image using small dots called halftone dots is widely used. In this method, an image is printed by printing a large number of halftone dots while changing the size of the halftone dots according to the density to be expressed. Then, if this is viewed a little away, the individual halftone dots will not be noticeable, and it will appear as if they are filled with a certain darkness, so that it is possible to express the shading.

Here, in an image formed using halftone dots, adjacent halftone dots are formed at a constant interval. Therefore, when reading an image with halftone dots by a scanner, the period between the halftone dots and the reading resolution of the scanner are determined. A phenomenon called “moire” in which interference fringes occur due to interference may occur, and image quality may deteriorate. Alternatively, when trying to generate digital data of an image whose shading is represented by halftone dots, each halftone dot must be represented by digital data, so the period of halftone dots and the digital data Moire can occur due to interference with the resolution. Therefore, a technique for avoiding the generation of moire due to halftone dots has been proposed by performing a blurring process using a low-pass filter, removing halftone dots, and replacing the shades represented by halftone dots with shades of intermediate colors ( For example, Patent Document 1).

JP 7-95409 A

However, with such a technique, the low-pass filter blurs the characters in the image and the outline of the subject, so that there is a problem that the entire image becomes a blurred and low-quality image. Of course, if the image data is analyzed in detail and the outline of the character or subject is detected, it is possible to avoid the outline portion and perform the blurring process. However, in order to accurately detect the outline, complicated image processing is required. This demands a new problem of increasing the processing load.

The present invention has been made in order to solve the above-described problems of conventional techniques, and an object of the present invention is to provide a technique that can remove halftone dots with a simple process without blurring the outline of characters or subjects. .

In order to solve at least a part of the problems described above, the image data analysis apparatus of the present invention employs the following configuration. That is,
An image data analysis device that extracts an outline in image data obtained by reading an image represented by a plurality of halftone dots,
A halftone dot cell calculating means for calculating a plurality of halftone cells that divide the image into a regular grid between all the halftone dots constituting the image;
Basic lattice calculation means for calculating a basic lattice composed of a predetermined number of halftone cells adjacent to each other;
Halftone dot arrangement detecting means for detecting the halftone dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid;
A basic in-grid contour extracting means for extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
The gist of the present invention is to provide contour extracting means for extracting a contour in the image data by extracting a contour in the basic lattice for each of the plurality of basic lattices adjacent to each other.

In addition, the image data analysis method of the present invention corresponding to the above image data analysis apparatus,
An image data analysis method for extracting an outline in image data obtained by reading an image represented by a plurality of halftone dots,
Calculating a plurality of halftone cells that divide the image into equidistant grids among all the halftone dots constituting the image;
Calculating a basic grid composed of a predetermined number of the halftone cells adjacent to each other;
Detecting the dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid; and
Extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
And a step of extracting a contour in the image data by extracting a contour in the basic lattice for each of the plurality of basic lattices adjacent to each other.

In such an image data analysis apparatus and image data analysis method of the present invention, when image data expressed by a plurality of halftone dots is received, the image data is divided into equal intervals between the halftone dots to surround each halftone dot. A grid (halftone cell) is calculated. In the case of an image in which only a part of the image is represented by halftone dots (for example, a photograph or a picture represented by halftone dots and a text portion in which normal black characters are used instead of halftone dots are mixed. In the case of an image or the like), a halftone cell may be set in a portion represented by a halftone dot. When the halftone cells are calculated, a basic grid is generated by collecting adjacent halftone cells every predetermined number, and the halftone dot arrangement in each basic grid is detected. And, by extracting the outline for each basic grid from the detected halftone dot arrangement,
Extract contours in image data.

According to the present invention, "if a predetermined number of halftone cells adjacent to each other are put together in a basic grid,
Based on the arrangement of halftone dots in the basic grid, it was completed by finding a new finding that the position of the contour can be extracted by simple processing. So first of all,
Such knowledge found by the inventors of the present application will be described. In general, when an image is represented by halftone dots, halftone dots are formed inside the outline and halftone dots are not formed outside the outline. It can be considered that it means whether the point cell is located inside or outside the outline. Then, if there is a halftone cell in which a halftone dot is not formed immediately adjacent to a halftone dot cell in which a halftone dot is formed, it can be estimated that a contour line exists between these halftone dot cells. it can. For example, when a halftone dot is formed in the left halftone cell of two adjacent halftone cells and no halftone dot is formed in the right halftone cell, the left halftone cell is Since the right halftone cell is considered to be outside the contour, it can be estimated that a contour line exists between the two halftone cells. Of course, it is also possible to estimate the position of the contour line from the presence / absence of halftone dot formation of not only two halftone cells but also more halftone cells. Therefore, if a basic grid is formed by grouping halftone cells for each predetermined number, the contour lines in each basic grid are extracted by checking whether or not halftone dots are formed in each basic grid. It is possible. Then, by connecting the contour lines extracted from the basic grids, it becomes possible to extract the contour lines in the image data.

The present invention is based on such new knowledge, and it is possible to extract a contour in image data by extracting a contour line from each basic lattice based on the halftone dot arrangement of the basic lattice. If the contour in the image data can be extracted in this way, it is possible to remove halftone dots without blurring the contour by performing blurring processing on the inside of the extracted contour line. In addition, according to such a method, it is possible to extract a contour only by checking the presence or absence of halftone dot formation in each halftone cell, and it is not necessary to perform other complicated image analysis, so that the processing burden is excessively increased. It is possible to eliminate halftone dots.

In the above-described image data analysis apparatus of the present invention, when calculating the basic grid, four halftone cells arranged in a direction intersecting each other may be collectively used as the basic grid.

In order to determine the direction of the contour line traversing the basic grid (for example, whether the contour traverses the basic grid in the vertical or horizontal direction), a plurality of halftone cells arranged in at least a direction intersecting each other, It is necessary to put it together as a basic lattice. As an example, let us consider a case where a halftone dot is not formed only in the upper right halftone dot cell among the basic lattice composed of four halftone dot cells arranged vertically and horizontally. First, in the left-right direction, a halftone cell in which no halftone dot is formed exists in the right half. Next, in the vertical direction, halftone cells in which no halftone dots are formed exist in the upper half. By taking these two things into consideration, it is possible to estimate that the contour line passes obliquely from the lower right to the upper left in the upper right part of the basic lattice. As described above, in order to determine the direction of the contour line, a plurality of halftone cells arranged in a direction intersecting each other are required.
On the other hand, as the number of halftone cells collected as a basic grid increases, it becomes more difficult to determine the position of the contour line in the basic grid. In view of these points, it is possible to easily determine the contour in the basic grid by collecting two halftone cells arranged in a crossing direction as a basic grid.

In the image data analysis apparatus of the present invention described above, when checking the presence / absence of halftone dot formation in each halftone cell, whether or not there is a region having a density higher than a predetermined density in the halftone cell. It is also possible to investigate. When such a region exists and the size of the region is equal to or larger than a predetermined ratio compared to the size of the halftone cell, a halftone dot is formed in the halftone cell. It may be judged as a thing.

In general, noise or the like does not form a continuous area of a constant size, but tends to exist in a discrete manner. Even if it occurs in the cell, it is possible to distinguish between halftone dots and noise. As a result, it is possible to accurately determine the presence or absence of halftone dots without erroneously determining the presence or absence of halftone dots due to noise or the like. It becomes possible.

In the image data analysis apparatus of the present invention described above, in addition to checking the halftone dot arrangement in each basic grid, it is also possible to check whether or not the halftone dots in each basic grid are missing. Then, in addition to the halftone dot arrangement, the contours in the basic grid may be extracted by considering the presence or absence of a halftone dot.

If a halftone dot is missing, it is considered that the missing is caused by the outline,
It can be estimated that the contour line passes at least near the missing halftone dot. Therefore, if the presence or absence of halftone dots in the basic grid is taken into account, the position of the contour line in the basic grid can be estimated more accurately. As a result, the contour in the image data can be extracted more accurately. It becomes possible to do.

In the image data analysis apparatus of the present invention described above, the center position of the halftone dot formed in the halftone cell is obtained, and the deviation between the obtained center position of the halftone dot and the center position of the halftone cell is examined. It may be determined whether the halftone dot shape is missing.

Whatever the size of the halftone dot is, if the halftone dot is missing, the center position is biased by the missing amount. Therefore, if the deviation between the center position of the halftone dot and the center position of the halftone cell that is the center position when the halftone dot is not missing is examined, whether or not the halftone dot is missing regardless of the size of the halftone dot. Can be accurately determined, and as a result, the contour can be accurately extracted.

In the above-described image data analysis apparatus of the present invention, a low-pass filter may be applied to a region inside the extracted contour, and the obtained image data may be output.

In this way, only the inside of the contour is blurred by the low-pass filter, so that halftone dots can be removed without blurring the contour. As a result, it is possible to output a suitable image having no moire and a clear outline. Further, the process of applying an image processing filter such as a low-pass filter to the image data can be easily executed by a CPU or the like, so that halftone dots can be removed without increasing the processing load.

Furthermore, the present invention can also be realized using a computer by causing a computer to read a program for realizing the above-described image data analysis method and executing a predetermined function. Therefore, the present invention also includes the following aspects as a program. That is, the program of the present invention corresponding to the image data analysis method described above is
A program for realizing, using a computer, a method for extracting a contour in image data obtained by reading an image represented by a plurality of halftone dots,
A function of calculating a plurality of halftone cells that divide the image into equidistant grids among all the halftone dots constituting the image;
A function of calculating a basic lattice composed of a predetermined number of halftone cells adjacent to each other;
A function of detecting a halftone dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid;
A function of extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
The gist is to realize, by a computer, a function of extracting a contour in the image data by extracting a contour in the basic lattice for each of the plurality of basic lattices adjacent to each other.

If this program is read into a computer and the above functions are realized, it is possible to quickly remove halftone dots and obtain a suitable image without moire.

Hereinafter, in order to clarify the contents of the present invention described above, examples will be described in the following order.
A. Device configuration:
B. Halftone removal processing:
C. Variations:
C-1. First modification:
C-2. Second modification:

A. Device configuration :
FIG. 1 is a perspective view illustrating a printing apparatus 10 equipped with the image data analysis apparatus of this embodiment. As illustrated, the printing apparatus 10 includes a scanner unit 100, a printer unit 200, a control unit 300 that controls operations of the scanner unit 100 and the printer unit 200, and the like. The scanner unit 100 has a scanner function of reading a printed image and generating image data, and the printer unit 200 has a printer function of receiving image data and printing an image on a print medium. . If the image read by the scanner unit 100 is output from the printer unit 200, a copy function can be realized. That is, the printing apparatus 10 according to the present embodiment is a so-called scanner / printer / copy combined apparatus that can independently realize a scanner function, a printer function, and a copy function.

The control unit 300 includes a CPU, a ROM, a RAM, and the like.
In addition to the control of the printer unit 200, various processes for improving the image quality of the image output from the printer unit 200 can be executed. For example, when copying an image whose shading is represented by halftone dots, if the captured image is printed as it is, the resolution of the scanner unit 100 or the printer unit 200 interferes with the halftone dots, resulting in moire. In view of this, it has a function of removing halftone dots by performing blurring processing on the captured image data. However, if the outline of the character or subject is blurred due to the blurring process, the entire image becomes a blurred low-quality image. Therefore, the control unit 300 according to the present embodiment is equipped with a function that can remove halftone dots without blurring the outline by analyzing the image data and estimating the outline.

FIG. 1 conceptually illustrates a state in which the control unit 300 has a function of removing halftone dots without blurring the outline. When the control unit 300 receives from the scanner unit 100, the storage medium, or the like the image data whose density is represented by the halftone dots, the control unit 300 analyzes the image data to estimate the contour,
A halftone dot is removed by applying a blurring process only to the inside of the contour. Then, by supplying the image data from which the halftone dots are removed to the printer unit 200, a suitable image having no moire and a clear outline is printed. Hereinafter, such processing performed by the control unit 300 will be described in detail.

B. Halftone removal processing:
FIG. 2 is an explanatory diagram showing the flow of “halftone removal processing” in the present embodiment. Such processing is processing performed to remove halftone dots of the read image when the scanner unit 100 captures the image printed by the halftone dots and outputs the image from the printer unit 200.
Of course, similar processing is performed when image data captured by the scanner unit 100 is stored in a recording medium, and the image data is read out and printed by the printer unit. Further, when image data that has been subjected to so-called halftone dot processing and converted into a representation format using halftone dots is received and output from the printer unit 200, similar processing is performed in the control unit 300.

When the copy function of the printing apparatus 10 is executed, the scanner unit 100 starts reading an image and transfers the read image data to the control unit 300. In response to this, the control unit 300 starts the “halftone removal process” in FIG. 2, and acquires the image data sent from the scanner unit 100 on the RAM of the control unit 300 (step S100). In this embodiment, it is assumed that the image illustrated in FIG. 3 is read.

After the image data is read, the image data is analyzed to obtain a “halftone cell” (step S102). Here, the “halftone cell” is as follows. Usually, when expressing shading using halftone dots, the halftone dots are not arranged at irregular positions but are arranged at regular positions. For example, in the image illustrated in FIG. 3, the shading is expressed by arranging a large number of halftone dots, but each of these halftone dots is not irregularly arranged. It is regularly arranged in a grid. In general, not limited to the image of FIG.
In an image in which shading is expressed by halftone dots, the halftone dots are regularly arranged in a grid pattern, and a grid surrounding each halftone dot can be assumed. This lattice is called a “halftone cell”. In step S102, the halftone dots of the image are divided into grids, and processing for obtaining a grid (halftone dot cell) surrounding each halftone dot is performed. Various methods can be used to obtain a halftone cell, but it can be easily obtained by calculating a basic vector representing the repetition of a halftone dot.

FIG. 4 is an explanatory diagram conceptually showing how the basic vector is calculated. In order to calculate the basic vector, first, an arbitrary region is selected as a sample region from among regions in which halftone dots are drawn. In the example of FIG. 4A, an area indicated as A is selected as a sample area. When the sample area is selected, this time, as shown in FIG. 4B, the image of the sample area is overlapped with the surrounding images while being moved little by little, and an area that matches the image of the sample area is searched. If an area that matches the image of the sample area is found, the image of the sample area is repeated there, so it is repeated by calculating the vector of the difference between the position of the area and the position of the original sample area. (Basic vector) is obtained (see FIG. 4C).

When one basic vector is obtained in this way, this time, in order to obtain another basic vector, the image of the sample area is moved in a direction different from the previous basic vector. For example, FIG.
In the example of c), since a horizontal basic vector is obtained, the image of the sample area may be moved in the vertical direction (see FIG. 4D). Then, as in the case of obtaining the basic vector previously, an area that matches the image of the sample area is searched for. When a matching region is found, another basic vector is obtained by obtaining a difference between the position of the region and the position of the original sample region (see FIG. 4D). When two basic vectors are obtained in this way, a halftone cell is obtained based on the two basic vectors.

FIG. 5 is an explanatory diagram conceptually showing how halftone cells are obtained from two basic vectors. As shown in FIG. 5A, when a basic vector is drawn starting from a halftone dot, since the basic vector represents the repetition period of the halftone dot, the end point of the basic vector is just the next halftone dot. Come in position. Therefore, focusing on the midpoint of the basic vector, the midpoint of the basic vector is exactly in the middle of the distance between the halftone dots, so if you draw a straight line parallel to the other basic vector, A straight line that evenly divides between halftone dots can be drawn. Similarly, by drawing a straight line from the middle point for the other basic vector, a straight line that evenly divides between the halftone dots can be drawn. If two straight lines are drawn in this way, this time, the two straight lines are drawn using the basic vector to draw a straight line over the entire halftone dot (
(Refer FIG.5 (b)). By drawing a straight line through the halftone dot using the basic vector,
A halftone cell surrounding each halftone dot is determined (see FIG. 5B).

FIG. 6 is an explanatory diagram showing a state in which halftone cells are set in a part of the image of FIG. 3 (portion indicated by “A” in FIG. 3). As shown in FIG. 6, when a halftone cell is set by drawing a straight line, a halftone cell can be set not only in a portion where a halftone dot actually exists but also in a portion where there is no halftone dot. Accordingly, in the next step S104, it is checked whether or not there is actually a halftone dot in the halftone cell for all of these halftone cells. Whether or not a halftone dot exists can be easily determined by acquiring pixel data in the halftone cell. That is, since the pixels in the halftone dot portion are filled with a predetermined gradation value, it is only necessary to check whether or not a pixel having this gradation value is in the halftone cell. For example, in the case of a halftone dot filled in black, the luminance of each pixel in the halftone cell is compared with a predetermined threshold value to check whether there is a pixel corresponding to the black dot. If there is a pixel corresponding to a black dot, the halftone cell can be determined to have “halftone dot”, and conversely, if there is no pixel corresponding to a black dot, “no dot” can be determined. . In addition, since a pixel having a black gradation value may exist by chance due to noise or the like, in order to avoid the influence of such noise or the like, a certain number or more of pixels having a black gradation value are present. It may be determined that “there is a halftone dot” only when there is.

As described above, when “halftone” or “no halftone” is determined for all halftone cells, then, based on the arrangement pattern of “halftone” and “no halftone” in the halftone cells. ,
A process for setting a region for removing halftone dots is performed (step S106 in FIG. 2). That is, the contour line of the subject is estimated from the arrangement of “halftone dot” and “no halftone dot” in the halftone cell, and the inside of the contour line is set as a halftone dot removal region (region subjected to blur processing by a low-pass filter). It is. Here, to facilitate understanding, before describing the specific contents of the processing performed in step S106, first, the contour of the subject is determined based on the arrangement pattern of “with halftone dots” and “without halftone dots”. A method of estimating the will be described.

FIG. 7 is an explanatory diagram showing how the contour is estimated from the arrangement pattern of “with halftone dot” and “without halftone dot”. A halftone dot in FIG. 7 is a part of the image shown in FIG.
A portion indicated by “A”), a line drawn in a grid pattern is a halftone cell set in the previous step S102. First, paying attention to the four halftone cells indicated by “A” in FIG. 7, all of the four halftone cells have halftone dots. Since there should be no halftone dots outside the outline,
These four halftone dots are all located inside the contour line. Therefore, it can be estimated that the contour line does not exist inside these four halftone dots.

On the other hand, when attention is paid to the four halftone cells indicated by “B” in the figure, the upper two halftone cells have no halftone dots, and the lower two halftone cells have halftone dots. In this case, it is considered that the upper two halftone cells without halftone dots are located outside the contour, and the lower two halftone cells with halftone dots are located inside the contour. . Therefore, it can be estimated that the outline passes in the horizontal direction between the lower two halftone cells and the upper two halftone cells.

Further, when attention is paid to the four halftone cells indicated by “C” in the figure, the upper right halftone cell has no halftone dot and the other three halftone cells have halftone dots. From this, the upper right halftone dot cell without halftone dots is located outside the outline, while the three halftone dots other than the upper right (cells with halftone dots) are located inside the outline. It is thought that there is. Therefore, it can be estimated that the outline passes obliquely between the upper right halftone cell and the other three halftone cells.

In this way, by looking at the positional relationship between the halftone dot cell with “halftone dot” and the halftone dot cell with “no halftone dot”, it is possible to infer the presence or absence or position of the outline. Therefore, step S1 in FIG.
In 06, a contour line is estimated based on the arrangement relationship of “with halftone dot” and “without halftone dot” obtained in the previous step S104, and the inside of the estimated contour line is represented by a halftone dot removal region (blurring process using a low-pass filter). The process of setting to (region to perform) is performed. When estimating the contour line,
It is possible to estimate the contour line from a larger number of halftone cells, not limited to four halftone cells.
Alternatively, for example, it is also possible to estimate a contour line from three halftone cells substantially by dividing four halftone cells into two diagonal lines. In this way, the number of halftone cells is not limited to four in principle, but in this embodiment, the four halftone cells “with halftone dots” and “
The description will be made assuming that the contour line is estimated based on the positional relationship of “no dot”.

Further, when the contour line is estimated based on the positional relationship between the “halftone dot” and “no halftone dot” of the four halftone cells, the four halftone cells are “with halftone dot” or “without halftone dot”, respectively. Therefore, the arrangement pattern of “with halftone dots” and “without halftone dots” has only 2 4 ways (= 16 ways) in total. Therefore, in the halftone dot removal processing of the present embodiment, the estimated contour lines are obtained in advance for these 16 types of halftone dot arrangement patterns, and the estimated contour lines correspond to the halftone dot arrangement patterns. By attaching and memorizing it, it is possible to quickly estimate the contour line. Hereinafter, a process (step S106 in FIG. 2) of estimating a contour line by such a method and setting a halftone dot removal area inside the contour line will be described in detail.

FIG. 8 is an explanatory diagram showing a state in which estimated contour lines and halftone dot removal regions are associated with the respective halftone dot arrangement patterns with respect to 16 halftone dot arrangement patterns. As described above, since the contour line can be estimated from the halftone dot arrangement pattern, it is possible to associate the contour lines with each of the 16 halftone dot arrangement patterns. For example, in the case of an arrangement pattern having all four halftone dots indicated by “A” in the figure, it is estimated that these four halftone dots are located inside the contour line as described above. Therefore, as shown in the figure, in this case, the inside of the halftone dot can be used as a halftone dot removal region. Similarly, in the case of an arrangement pattern without the upper right halftone dot (indicated by “B” in FIG. 8), the upper right halftone dot is located outside the outline, and the other three halftone dots are located inside the outline. Therefore, the contour line is at the position indicated by the thick line in the figure, and the lower left region of the contour line can be used as a halftone dot removal region. Similarly, other halftone dot removal regions can be associated with other arrangement patterns. The control unit 300 of the printing apparatus 10 of the present embodiment stores the 16 halftone dot arrangement patterns and the halftone dot removal areas in association with each other, and using these, the halftone dot removal area is stored as follows. Set it.

FIG. 9 is an explanatory diagram showing a state in which a halftone dot removal area is set in an image. The image shown in FIG. 3 is a part of the image shown in FIG. 3 (the part indicated by “A” in FIG. 3), and the broken line is a halftone dot cell. For this image, the control unit 300 of this embodiment sets a halftone dot removal area as follows. First, paying attention to the four halftone cells indicated by “A” in FIG. 9A, the lower two halftone cells have halftone dots and the upper two halftone cells have halftone dots. Because there is no
This halftone dot arrangement pattern is the halftone dot arrangement pattern indicated by “C” in FIG. Therefore, the control unit 300 reads the halftone dot removal area stored in association with the halftone dot arrangement pattern, and sets the halftone dot removal area in the image. When the halftone dot removal area is set by paying attention to the four halftone cells indicated by “A” in the drawing, the halftone dot removal area is set by paying attention to another four halftone cells. .

In FIG. 9B, the halftone dot removal area is set by paying attention to the four halftone cells indicated by “B” in the figure. These four halftone cells correspond to the halftone dot arrangement pattern indicated by “A” in FIG. 8 since all halftone cells have halftone dots. Therefore, the control unit 300 reads out the halftone dot removal area associated with the halftone dot arrangement pattern and sets the halftone dot removal area in the image.
Then, the halftone dot removal area is set so as to be connected to the halftone dot removal area set in FIG. 9A.

After the halftone dot removal area is set in this way, the halftone dot removal area is set by paying attention to another four halftone cells. For example, if a halftone dot removal area is set by paying attention to four halftone cells while moving downward by one halftone cell, halftone dot removal is performed as shown in FIG. The area can be set sequentially downward. Then, at the position indicated by “C” in the figure, this time, the arrangement pattern indicated by “D” in FIG. 8 is used, so halftone dot removal associated with “D” in FIG. 8 is performed. What is necessary is just to set an area | region. In this way, by focusing on the four halftone cells and reading the halftone dot removal area stored in association with the arrangement pattern, the halftone dot removal areas can be sequentially set.

FIG. 10 is an explanatory diagram showing a state in which a halftone dot removal region is set by performing such processing on all halftone cells. As described above, if a halftone dot removal area is set for each arrangement pattern of four halftone cells, a halftone dot removal area is set inside the contour estimated from the arrangement pattern, and is shown in the figure. Similarly, the halftone dot removal region is set only inside the estimated contour line. After the halftone dot removal area is set in this way, a process for removing halftone dots is performed by applying a low-pass filter to the halftone dot removal area (step S108 in FIG. 2).
.

FIG. 11 is an explanatory diagram illustrating a low-pass filter used for removing halftone dots. As shown in the figure, in the low-pass filter, the gradation value of the target pixel and the gradation values of the surrounding pixels are added with the same weight to obtain an average value, and the obtained average value is used for the gradation of the target pixel. Value.
If such a low-pass filter is applied to pixels in the halftone dot removal region as target pixels, the halftone dots are removed by averaging the gradation values in the halftone dot removal region. For example, in the case of a black halftone dot, the halftone dot removal region is composed of a halftone dot pixel (black) and a pixel other than the halftone dot (for example, white). By being replaced with the average gradation value, both the halftone pixel and the pixel other than the halftone are intermediate colors between black and white. Thus, the halftone dots are removed by making the halftone dot portion and the other portions in the same color indistinguishable. At this time, since the gradation value corresponding to the area ratio between the halftone dot portion and the portion other than the halftone dot (that is, the shading represented by the halftone dot) is obtained, not only the halftone dot is removed but also the halftone dot is removed. It is possible to replace the shades represented by dots with gradation values.

When the low pass filter is applied to the halftone dot removal area, the average gradation value is calculated using all the pixels within the range of the low pass filter (25 pixels of 5 vertical pixels × 5 horizontal pixels in the low pass filter in FIG. 11). The average gradation value may be calculated by using only the pixels in the halftone dot removal region among the pixels within the range of the low-pass filter. For example, when a low pass filter is applied using a pixel immediately inside the contour line as a target pixel, a pixel outside the contour line may fall within the range of the low pass filter. In such a case, a pixel outside the contour line may be included. The average gradation value may be calculated using only the pixels inside the contour line without using. By doing so, since there is no influence of the gradation value outside the contour line, a more accurate gradation value corresponding to the shading represented by the halftone dot can be obtained. In addition, since it is possible to avoid a situation in which the gradation value inside the contour line and the gradation value outside the contour line are slightly approaching, the contour can be made clearer.

When the halftone dot is removed by applying the low pass filter to the halftone dot removal region in this way, the halftone dot removal process in FIG. Based on the image data from which the halftone dots have been removed, the printer unit 20
If the image is printed at 0, since the halftone dots are removed, a good image without moire or the like can be obtained. At this time, since the low-pass filter is applied only to the inside of the contour line in the halftone dot removal processing of the present embodiment, the situation where the subject and the contour of the character are blurred due to the low-pass filter is avoided. As a result, it is possible to obtain a suitable image with a clear outline.

As described above, in the halftone dot removal process of the present embodiment, the contour line is estimated based on the arrangement pattern of the halftone dots, and the low pass filter is applied only to the inside of the estimated contour line.
It is possible to remove halftone dots without blurring the outline of the subject or characters. of course,
Since the contour estimated from the arrangement of the halftone dots does not exactly match the original contour, the original contour cannot be accurately reproduced. However, as long as the halftone dots are arranged according to the outline, the outline estimated from the arrangement of the halftone dots is not significantly different from the original outline, so by performing halftone dot removal based on the estimated outline Thus, it is possible to print a suitable image in which the contour is almost correctly reproduced. In particular, with regard to the outline of a character, even if the estimated outline is slightly different from the original outline, it does not feel unnatural. Therefore, by removing halftone dots based on the estimated outline, it is possible to obtain sufficient quality. It is possible to print high-quality characters.

Further, in the halftone dot removal process of the present embodiment, the halftone dot placement pattern and the halftone dot removal area are stored in advance in association with each other with respect to the 16 halftone dot placement patterns (see FIG. 8). Halftone dot removal is possible. That is, when setting a halftone dot removal region in an image, it is only necessary to match the halftone dot arrangement pattern of the image with the 16 halftone dot arrangement patterns to find a matching halftone dot arrangement pattern. In the pattern matching process, there are 16 patterns to be matched.
Since it only passes, it can be processed extremely quickly. For example, when collation processing is performed using a CPU, the data size required for the collation processing is only 4 bits (that is, 16 types), so a CPU that can process 4 bits in one cycle (for example, a general 8-bit CPU)
Etc.), high-speed processing is possible in only one cycle. In this way, when setting the halftone dot removal area, it is only necessary to perform a halftone dot pattern matching process that can be performed at high speed, and no other complicated image analysis processing is required. It can be removed. In addition, since collation processing can be performed with only one CPU cycle, even a general-purpose low-speed CPU can perform sufficiently high-speed processing, so that a high-speed LSI dedicated to image processing is not required. It is also possible to keep the device configuration simple.

When setting the halftone dot removal area from the halftone cell arrangement pattern, as described above, the halftone dot removal area is not set by using the four halftone cells, but more halftone dot cells are set. It is also possible to set using a small number of halftone cells. However, when four halftone cells are used as in this embodiment, it is possible to set the halftone removal area particularly easily. This point will be supplementarily described. That is, a large number of halftone dots drawn in an image can be regarded as a lattice having each halftone dot as a lattice point, if the way of viewing is changed. Then, the four halftone dots constitute just a unit cell of this lattice. Here, since the unit cell has the property (translational symmetry) that can form the entire cell by arranging the unit cells in parallel, the halftone dot removal region can be associated with the four halftone dots that are the unit cell. For example, it is possible to set a halftone dot removal region for the whole halftone dot by simply arranging the halftone dot removal regions while being translated. For this reason, as shown in FIG. 9, it is possible to set a halftone dot removal area for the whole halftone dot by simply executing a process for setting a halftone dot removal area for each of the four halftone dot arrangement patterns. It has become.

C. Modified example:
C-1. First modification:
In the above-described embodiments, the description has been made on the assumption that the contour line is estimated based on the arrangement pattern of the halftone cells with “halftone” and the halftone cells with “no halftone”. However, “with halftone dots” and “
In addition to the arrangement pattern of “no halftone dots”, it is possible to estimate the contour line more accurately by considering whether or not halftone dots are missing.

FIG. 12 is an explanatory diagram showing a state in which the contour line is estimated in consideration of the lack of halftone dots. Here, FIG. 12A is an outline estimated from the arrangement pattern in the case of not considering the lack of halftone dots described in the above-described embodiment (see FIG. 8). In the case of the halftone dot arrangement pattern in FIG. 12 (a), as described above, there is no halftone dot in the upper right, and there are other halftone dots, so that there is a contour line at the position indicated by the solid line in the figure. It is estimated to be. On the other hand, when a halftone dot is missing, the lack is caused by a contour line, so it can be estimated that the contour line passes at least in the immediate vicinity of the missing halftone dot. For example, in the arrangement pattern shown in FIG. 12B, since the upper left halftone dot is missing, it can be estimated that the contour line passes through the immediate vicinity of this halftone dot. Therefore, when there is a missing halftone dot, it is possible to estimate a more accurate contour line by bringing the outline closer to the missing halftone dot. FIG. 12B shows how a more accurate contour line is estimated in this way. In FIG. 12B, the contour line is made to approach the center of the halftone cell. However, the present invention is not limited to this, and any form may be used as long as the position of the contour line is brought closer to the missing halftone dot from the state shown in FIG.

Further, as shown in FIG. 12 (c), when the lower right halftone dot is missing, it can be estimated that the outline passes near the lower right halftone dot. By bringing the outline of (a) closer to the lower right halftone dot, a more accurate outline can be obtained. Furthermore, as shown in FIG. 12D, when both the upper left halftone dot and the lower right halftone dot are missing, the contour line passes near the two halftone dots. Therefore, a more accurate contour line can be obtained by bringing the contour line in FIG. 12A closer to the two missing halftone dots.

12 (e) and 12 (f) show a state in which a more accurate contour line is estimated from the lack of halftone dots when there are halftone dots only in the left two of the four halftone cells. It is shown. In the case of FIG. 12E, since there are no two right halftone dots as described in the above-described embodiment, the illustrated contour line can be estimated. On the other hand, when the upper left halftone dot is missing as shown in FIG. 12 (f), it is considered that an outline exists near the missing halftone dot. ) Is closer to the upper left halftone dot, a more accurate contour line can be obtained. Similarly, for other halftone dot arrangement patterns, it is considered that there is a contour line near the missing halftone dot, so the contour line can be estimated more accurately by bringing the contour line closer to the missing halftone dot. It is possible.

In this way, when a halftone dot is missing, it is considered that the outline passes through the immediate vicinity of the missing halftone dot. Therefore, a more accurate outline can be obtained by bringing the outline close to the missing halftone dot. A line is obtained. Therefore, in the above-described embodiment, the estimated contour line and the halftone dot removal area are stored in advance in association with the arrangement pattern of “with halftone dot” and “without halftone dot” in the four halftone cells. As in (see FIG. 8), in the first modified example, the estimated contour lines for the arrangement patterns of “with halftone dots”, “without halftone dots”, and “missing halftone dots” in the four halftone cells. The halftone dot removal area is stored in association with each other. Here, for the arrangement pattern including “halftone dot missing”, as described above, it is only necessary to store the outline and halftone dot removal area obtained by bringing the outline close to the missing halftone dot, For an arrangement pattern that does not include “halftone dot missing”, the outline and halftone dot removal region shown in FIG. 8 may be stored. Then, similarly to the above-described embodiment (see FIG. 9), the halftone dot removal region is set in the image by collating the halftone dot arrangement pattern of the image with the stored arrangement pattern. If the halftone dot removal area is set in this way, it is possible to estimate the contour line more accurately based on the lack of halftone dots and set the halftone dot removal area inside.

In the first modification, since the four halftone dots can take three ways of “with halftone dot”, “no halftone dot”, and “missing halftone dot”, respectively, a total of 3 to the 4th power (= 81 ways) Since it is necessary to store the halftone dot placement pattern and the halftone dot removal area in association with each other, the number of placement patterns to be stored is larger than in the above-described embodiment. However, as described in the above-described embodiment, since the process of collating the stored arrangement pattern with the halftone dot arrangement pattern of the image can be executed at high speed, the number of arrangement patterns to be stored increases. However, the halftone dot removal region can still be set at high speed.

The determination of whether or not the halftone dot in the halftone cell is missing can be made by various methods. However, the position of the center of gravity of the halftone dot is calculated and the position of the center of the halftone cell is calculated. It can be easily judged by examining the difference. FIG. 13 shows how a halftone dot is determined by examining the difference between the center of gravity of the halftone dot and the center of the halftone cell. Pixels indicated by hatching in the figure are pixels constituting halftone dots. Pixels constituting a halftone dot can be extracted by examining whether or not all pixels in the halftone cell have a gradation value corresponding to the halftone dot. When the pixels constituting the halftone dots are extracted in this way, the barycentric positions of those pixels are calculated. Here, if the halftone dot is not missing, as shown by the dotted line in FIG. 13, the pixels constituting the halftone dot should exist in a substantially circular shape at the center of the halftone cell. Matches the center of the halftone cell. On the other hand, if the halftone dot is missing, the center of gravity is biased by the lack of the halftone dot, so that a deviation occurs between the center of the halftone cell and the center of gravity of the halftone dot. In this way, it is possible to determine whether or not a halftone dot is missing by examining whether or not there is a deviation between the position of the center of gravity of the halftone dot and the center of the halftone dot cell. If a halftone dot is determined by such a method, the center of gravity will be shifted if there is a halftone dot of any size, so the shading represented by the halftone dot (ie, the size of the halftone dot) will be reduced. Regardless, it is possible to easily detect missing dots.

FIG. 14 is an explanatory diagram showing a state in which a halftone dot removal region is set in consideration of the lack of halftone dots in the arrangement pattern of “with halftone dots” and “without halftone dots”. As described above, the lack of a halftone dot means that there is a contour line near the halftone dot, so by considering the lack of halftone dot, the contour line is more accurately shown as shown in the figure. It is possible to estimate. If a low-pass filter (see FIG. 11) is applied to the halftone dot removal region set in this way, a suitable image in which the original contour is reproduced more accurately can be output.

C-2. Second modification:
In the first modified example described above, contour lines estimated for the halftone dot arrangement patterns of “with halftone dot”, “no halftone dot”, and “halftone dot missing” are stored in advance in association with each other. Was described as an accurate estimate. However, it is possible not only to consider the presence or absence of halftone dots, but also to estimate the outline more accurately by adjusting the approach of the outline more finely in consideration of the amount of lack. is there.

FIG. 15 is an explanatory view exemplifying how the outline is finely adjusted for the missing halftone dot. The broken line in the figure is the contour estimated by the method of the first modification example, and as described above, it is close to the missing halftone dot (in the example shown, the halftone dot cell The outline is drawn (towards the center). However, it can be seen that the outline may be brought closer to the halftone dot by looking at the lack of the halftone dot.

Here, when adjusting the contour line, the shape of the missing halftone dot may be examined in detail, and the contour line may be adjusted so that the shape of the part lacking the halftone dot and the contour line are smoothly connected. Alternatively, the outline may be adjusted so as to pass through the pixel located at the extreme end among the pixels constituting the missing halftone dot (the rightmost pixel in the example of FIG. 15). . Note that the process of making these fine adjustments is
Since only the missing halftone dots need to be processed, the processing load is not excessive. For this reason, also in the second modification, the halftone dot removal region can be set still at high speed.

FIG. 16 is an explanatory view exemplifying a state in which a halftone dot removal region is set with respect to a contour line subjected to fine adjustment. If the outline is fine-tuned based on the missing halftone dot, the outline can be drawn closer to the original outline, so that the outline can be estimated more accurately as shown in the figure. It becomes.

The image data analysis apparatus according to the present embodiment has been described above. However, the present invention is not limited to all the embodiments described above, and can be implemented in various modes without departing from the spirit of the present invention.

For example, in the present embodiment, the case where the image data analysis apparatus is mounted on the printing apparatus has been described. However, the image data analysis apparatus is not limited to the case where the image data analysis apparatus is mounted on the printing apparatus. It can be mounted on an image display device,
It can also be mounted on a portable display device such as a photo viewer or a mobile phone terminal. Further, it can be mounted on an unmanned photo printing terminal placed on a street corner or in a public place.

It is explanatory drawing which shows the printing apparatus carrying the image data analysis apparatus of a present Example. It is the flowchart which showed the flow of the halftone dot removal process of a present Example. It is explanatory drawing which illustrated the image which performs the halftone dot removal process of a present Example. It is explanatory drawing which showed a mode that the basic vector for calculating | requiring a halftone cell was calculated. It is explanatory drawing which showed a mode that a halftone cell was calculated | required based on a basic vector. It is explanatory drawing which showed a mode that the halftone cell was set to the image. It is explanatory drawing which showed a mode that a contour line was estimated based on the arrangement pattern of "with halftone dot" and "without halftone dot". It is explanatory drawing which showed a mode that the halftone dot arrangement | positioning pattern and the halftone dot removal area | region were matched. It is explanatory drawing which showed notionally the mode that the halftone dot removal area | region is set for every four halftone cells. It is explanatory drawing which showed a mode that the halftone dot removal area | region was set. It is explanatory drawing which illustrated the low pass filter used for a halftone dot removal. In a 1st modification, it is explanatory drawing which showed notionally that a contour line was estimated in consideration of the lack of halftone dots. It is explanatory drawing which showed notably the mode that the missing | missing of a halftone dot was detected by investigating the gravity center of a halftone dot in the 1st modification. FIG. 10 is an explanatory diagram showing a state in which a halftone dot removal area is set in consideration of missing halftone dots in the first modification. In the 2nd modification, it is explanatory drawing which showed notionally that a contour line was fine-tuned in consideration of the lack of each halftone dot. It is explanatory drawing which showed a mode that the halftone dot removal area | region was set in the 2nd modification.

Explanation of symbols

10: Printing device,
DESCRIPTION OF SYMBOLS 100 ... Scanner part 200 ... Printer part 300 ... Control part

Claims (8)

An image data analysis device that extracts an outline in image data obtained by reading an image represented by a plurality of halftone dots,
A halftone dot cell calculating means for calculating a plurality of halftone cells that divide the image into a regular grid between all the halftone dots constituting the image;
Basic lattice calculation means for calculating a basic lattice composed of a predetermined number of halftone cells adjacent to each other;
Halftone dot arrangement detecting means for detecting the halftone dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid;
A basic in-grid contour extracting means for extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
An image data analyzing apparatus comprising: a contour extracting unit configured to extract a contour in the image data by extracting a contour in the basic lattice for each of the plurality of basic lattices adjacent to each other. The image data analysis device according to claim 1,
The basic grid calculation means is an image data analysis device which is a means for using the four halftone cells arranged in a direction crossing each other as the basic grid. The image data analysis device according to claim 1,
The halftone dot detection means has a size equal to or larger than a predetermined area ratio with respect to the halftone dot cell, and the halftone dot cell in which a continuous region having a concentration higher than a predetermined value is detected includes An image data analysis device which is means for determining that halftone dots are formed and detecting the halftone dot arrangement. The image data analysis device according to claim 1,
In addition to the halftone dot arrangement in the basic grid, the halftone dot arrangement detecting means is a means for detecting whether or not the halftone dot shape is missing for each of the halftone dots constituting the halftone dot arrangement. Yes,
The basic lattice outline extracting means is a means for extracting a contour in the basic grid based on the halftone dot arrangement in the basic grid and the presence / absence of the lack of the halftone dots constituting the halftone dot arrangement. An image data analysis apparatus. The image data analysis device according to claim 1,
The halftone dot arrangement detecting means detects the presence or absence of a lack of the halftone dot shape based on a deviation between the center position of the halftone dot detected in the halftone dot cell and the center position of the halftone dot cell. An image data analysis apparatus that is means for performing The image data analysis device according to claim 1,
After applying a low pass filter to the area inside the contour extracted in the image data,
An image data analysis device comprising image data output means for outputting the obtained image data. An image data analysis method for extracting an outline in image data obtained by reading an image represented by a plurality of halftone dots,
Calculating a plurality of halftone cells that divide the image into equidistant grids among all the halftone dots constituting the image;
Calculating a basic grid composed of a predetermined number of the halftone cells adjacent to each other;
Detecting the dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid; and
Extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
Extracting a contour in the image data by extracting a contour in the basic lattice for each of the plurality of basic lattices adjacent to each other. A program for realizing, using a computer, a method for extracting a contour in image data obtained by reading an image represented by a plurality of halftone dots,
A function of calculating a plurality of halftone cells that divide the image into equidistant grids among all the halftone dots constituting the image;
A function of calculating a basic lattice composed of a predetermined number of halftone cells adjacent to each other;
A function of detecting a halftone dot arrangement in the basic grid by detecting the presence or absence of the halftone dots in each halftone cell in the basic grid;
A function of extracting a contour in the basic grid based on the halftone dot arrangement detected in the basic grid;
A program for realizing a function of extracting a contour in the image data by extracting a contour in the basic lattice for each of a plurality of the basic lattices adjacent to each other.

Images