JP2007188284A - Image processing apparatus, image processing method, image processing program, and recording medium recording the program - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing apparatus capable of suppressing the increase of a granular feeling by executing processing so that each dot is formed on a prescribed pixel position in accordance with an AM screen to be applied, and an image processing method, an image processing program, or a recording medium in which the program is recorded. <P>SOLUTION: A pixel group division part 22a divides an input image into cells on the basis of dot positions corresponding to the AM screen to be used in an image output apparatus 30. A centroid position calculation part 22b calculates a centroid position of a cell from the input gradation value and coordinates of each pixel in the cell. A centroid position determination part 22c determines a centroid position by integer number processing by adding or subtracting a prescribed value to/from the centroid position calculated in the prescribed cell. An arrangement part 22d arranges a center of a gradation transformation matrix on the determined centroid position. A gradation transformation part 22e refers to a gradation transformation table stored in the arranged gradation transformation matrix and converts the input gradation value of each pixel in the cell into an output gradation value. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

The present invention divides an input image into pixel groups each composed of a plurality of pixels, and applies a gradation conversion matrix to each of the divided pixel groups, whereby an input gradation value of each pixel included in the pixel group is obtained. The present invention relates to an image processing technique for converting a tone into an output tone value having two or more types of tone values.

Conventionally, in an image output apparatus having a print engine such as a printer, the gradation value of input image data is converted into two or more kinds of predetermined output gradation values by gradation conversion processing such as halftone processing for each pixel, Dots are formed according to the converted output gradation values, and printing is performed on printing paper with a halftone dot (AM screen) image generated from the formed dots, and an output image is obtained as a printed matter. .

As halftone processing, conversion processing by a dot concentration type dither method (multi-value dither method) is known. In the multi-value dither method, threshold values are distributed so that dots grow from the center in a matrix of a predetermined size, and gradation conversion processing is performed by comparison with the input gradation value of each pixel. is there.

However, when a matrix in which threshold values are distributed is applied to the input image, the input gradation values of the pixels are lost in areas where the threshold values are large in the matrix, and conversely, gradations which are larger than the input gradation values in areas where the threshold values are small. Since it is converted into a value, it may be a printed matter different from the input image. For example, in the case of an input gradation value having an intermediate size, such as a gray line, a printed product with a line cut off at a position corresponding to a pixel whose input gradation value is lost by comparison with a threshold value. As described above, the processing by the multi-value dither method has a problem in the image quality of the output image.

Furthermore, in the multi-value dither method, the dot positions are fixed at a predetermined interval according to the AM screen, so that the dots can be formed at the optimum positions for reproducing the distribution state of the input gradation values. In addition, there is a problem that the printed material is not faithful to the input image.

Therefore, in order to solve these problems, the present applicant divides the input image into pixel groups (also referred to as cells) composed of a plurality of pixels, and calculates the centroid from the input gradation value of each pixel in the divided cells. A prior invention has been proposed in which the position is obtained and a dot corresponding to the sum of the input gradation values of each pixel is generated at the position of the center of gravity (hereinafter referred to as “AAM (Advanced AM screen method)”).
Call it. Japanese Patent Application No. 2004-137326: Unpublished at the time of confirmation before filing this application ).

However, in the AAM method, a halftone dot (that is, a dot) is formed on the assumption that the center of gravity of the cell is located approximately at the center of any pixel in the cell. Therefore, even when the center of gravity is located at the boundary of the pixel, processing is performed so as to form a dot at the approximate center position of the pixel.

By the way, the AM screen used in the image output apparatus may be one having a halftone dot arrangement in which the center of a halftone dot is originally located at a pixel boundary. In such a case, AA
In the M method, as described above, dots are formed so that the center of gravity of the cell is located at the approximate center of any pixel in the cell. However, there is a problem that the graininess is increased due to this variation.

The present invention solves such problems, and an image processing apparatus and an image processing method for suppressing an increase in graininess by processing so that dots are formed at predetermined pixel positions in accordance with an AM screen to be used. An object of the present invention is to provide an image processing program or a recording medium on which the program is recorded.

An image processing apparatus of the present invention that achieves the above object divides an input image into a pixel group composed of a plurality of pixels, and applies a gradation conversion matrix to each of the divided pixel groups, thereby providing the pixel group with An image processing apparatus that performs tone conversion of an input tone value of each pixel included into an output tone value having two or more types of tone values corresponding to dots constituting a halftone image that is an output image, A pixel group dividing unit that divides the input image so that the shape of the divided pixel group becomes at least two different divided shapes based on the halftone dot positions of the halftone image. And

According to the image processing apparatus having such a configuration, when the input image is divided into a plurality of cells, the input image is divided into at least two types of divided shapes based on the halftone dot positions. Therefore, for example, the number of pixels constituting the cell can be increased or decreased according to the halftone dot position, or the cell shape can be changed, so that the dot formation position can be divided into a preferable cell shape so as to approach a predetermined pixel position. Therefore, an increase in graininess can be suppressed.

Furthermore, the image processing apparatus of the present invention includes a centroid position calculation unit that calculates a centroid position of the pixel group using an input gradation value of each pixel included in the pixel group, and the pixel group division unit includes: When each pixel included in the pixel group has the same input gradation value, the calculated gravity center position is divided so as to be the halftone dot position.

According to such a configuration, there is a high probability that the center of gravity position of the cell, that is, the predetermined pixel position matches the halftone dot position, and it becomes possible to output a halftone image in which the formed inter-dot distance is maintained. . Therefore, it is possible to suppress an increase in graininess that occurs when the dot formation position shifts.

Here, the pixel group dividing unit calculates the first pixel group in which the calculated barycentric position is the pixel center when the pixels included in the pixel group have the same input gradation value. The input image may be divided into a second pixel group in which the barycentric position is a pixel boundary.

Depending on the AM screen used by the image output apparatus, there are cases where the halftone dot position is not only located approximately at the center of the pixel but also located approximately at the boundary between the pixels. A like this
By dividing the cells according to the halftone dot positions on the M screen, it is possible to increase the probability that dots are formed at predetermined pixel positions.

Furthermore, the pixel group dividing unit may divide the input image so that the number of pixels included in the first pixel group and the number of pixels included in the second pixel group are the same. .

In this way, since the input tone values of the input image can be tone-converted without the sum of the input tone values in the divided cells differing greatly from cell to cell, the size of the dots formed is made uniform. , Increase in graininess can be suppressed.

The pixel group dividing unit may divide the input image so that the divided shape of the first pixel group and the divided shape of the second pixel group are at least left-right symmetric or vertically symmetric. Good.

In this way, one gradation conversion matrix is prepared for the gradation conversion matrix to be applied at the time of gradation conversion, and this one gradation conversion matrix is turned over so that the first cell and the second cell are reversed. It becomes possible to apply in common. Accordingly, the storage capacity can be reduced.

Further, when the barycentric position of the pixel group calculated from the input gradation value of each pixel does not coincide with the center of the pixel, the barycentric position calculating unit uses the predetermined integer processing method to calculate the barycentric position. A center-of-gravity position determining unit that determines an integerized position as the center-of-gravity position of the pixel group may be further provided.

In this way, the center of gravity position can be made to coincide with the center position of the predetermined pixel, so that dots can be formed around the predetermined pixel. As a result, since a halftone dot image with little shift in the dot formation position is formed, an increase in graininess due to the dot position shift can be suppressed.

Furthermore, the center-of-gravity position determination unit has a half length of one pixel width for either one of the first pixel group or the second pixel group according to the predetermined integerization processing method used. A value corresponding to this may be added to or subtracted from the calculated barycentric position of the pixel group.

In this way, in the integerization process for matching the calculated barycentric position with the center of the pixel, the center of the dot formation position is predetermined for both the first pixel group and the second pixel group. It can be set as the center position of the pixel.

Furthermore, the image processing apparatus of the present invention includes an arrangement unit that arranges the center of the gradation conversion matrix at the center of gravity of the pixel group, and a predetermined gradation conversion table stored in the arranged gradation conversion matrix. Referring to FIG. 5, a gradation conversion unit that performs gradation conversion of an input gradation value of each pixel of the pixel group into an output gradation value may be further provided.

By doing this, the center of the gradation conversion matrix is moved to a predetermined pixel position corresponding to the center of gravity position of the cell, and gradation conversion is performed with reference to a predetermined gradation conversion table for each pixel in the cell. Dots can be formed centering on a predetermined pixel, and an output image faithful to the input image can be obtained while suppressing an increase in graininess due to the positional deviation of the dots.

In addition, the image processing method of the present invention that achieves the above object divides an input image into pixel groups composed of a plurality of pixels, and applies a gradation conversion matrix to each of the divided pixel groups.
An image processing method for gradation-converting an input gradation value of each pixel included in the pixel group into an output gradation value having two or more kinds of gradation values corresponding to dots constituting a halftone image serving as an output image The center-of-gravity position calculating step of calculating the center-of-gravity position of the pixel group using the input gradation value of each pixel included in the pixel group, and the same input gradation value for each pixel included in the pixel group A pixel group dividing step of dividing into a first pixel group in which the calculated centroid position is a pixel center and a second pixel group in which the calculated centroid position is a pixel boundary; When the barycentric position of the pixel group calculated from the input gradation value of each pixel does not coincide with the center of the pixel, a position obtained by converting the calculated barycentric position into an integer value using a predetermined integer processing method is used as the pixel group. A center of gravity position and a center of gravity position determining step for determining, The center position determining step corresponds to half the length of one pixel width for either one of the first pixel group or the second pixel group, depending on the predetermined integer processing method used. A value to be added is added to or subtracted from the calculated barycentric position of the pixel group.

Further, the present invention may be a computer program or a recording medium on which this program is recorded. That is, by dividing the input image into pixel groups composed of a plurality of pixels and applying a gradation conversion matrix to each of the divided pixel groups, the input gradation value of each pixel included in the pixel group is An image processing program for performing gradation conversion into an output gradation value having two or more kinds of gradation values corresponding to dots constituting a halftone dot image serving as an output image, the input of each pixel included in the pixel group The center-of-gravity position calculation function for calculating the center-of-gravity position of the pixel group using the gradation value and the calculated center-of-gravity position at the pixel center when each pixel included in the pixel group has the same input gradation value A pixel group dividing function for dividing the first pixel group into a second pixel group in which the calculated barycentric position is a pixel boundary, and the pixel group calculated from the input gradation value of each pixel If the center of gravity does not coincide with the center of the pixel, The computer realizes a centroid position determination function for determining the position obtained by converting the calculated centroid position into an integer value using a predetermined integer processing method and the centroid position of the pixel group, and the centroid position determination function is used In accordance with a predetermined integerization processing method, for the pixel group of either the first pixel group or the second pixel group, the calculated pixel group has a value corresponding to half the length of one pixel width. It is characterized by being added to or subtracted from the barycentric position.

As a recording medium on which this program is recorded, a flexible disk or a DVD-
Various media that can be read by a computer, such as a ROM, a CD-ROM, an IC card, and a punch card, can be used.

Next, an embodiment for carrying out the present invention will be described. FIG. 1 is a configuration diagram showing an example of an entire system including an image processing apparatus to which the present invention is applied. The image input device 10, the image processing device 20, and the image output device 30 are configured as a whole. These devices are basically computers or devices controlled by a computer, and operate by executing a predetermined program under a predetermined operating system.

The image input device 10 includes an application unit 11 and a rasterizing unit 12. The application unit 11 generates data to be printed such as character data, graphic data, and bitmap data. For example, the image input device 10 uses an application program such as a word processor or a graphic tool to generate character data, graphic data, and the like by operating a keyboard (not shown). The generated data is output to the rasterizing unit 12. Of course, the image input apparatus 10 receives data output from a device that acquires image data such as a scanner or a digital still camera (not shown), and outputs this data to the rasterization unit 12 as data generated by the application unit 11. There is no problem.

The rasterizing unit 12 converts the print target data output from the application unit 11 into image data of a plurality of colors for each pixel and outputs the image data. In this embodiment, R (red), G (
Image data represented by a multi-gradation value of 8 bits (24 bits in total) for each pixel is output for each color of green and blue (blue). Therefore, the output image data has 256 kinds of gradation values from 0 to 255 for each pixel. The image data generation processing in the rasterizing unit 12 is actually performed by a driver (not shown) installed in the image input device 10. Thus, the image data output from the rasterizing unit 12 is output to the image processing apparatus 20.

The image processing apparatus 20 is basically a computer and includes a color conversion unit 21 and a halftone processing unit 22. In the image processing device 20, color conversion processing and halftone processing are performed on the image data output from the image input device 10, and processing for outputting predetermined gradation data DR to the image output device 30 is performed.

The color conversion unit 21 includes RGB having 256 types of gradation values output from the image input device 10.
Each 8-bit (24 bits in total) image data is received and color conversion processing is performed in accordance with the color components output by the image output device 30. In the present embodiment, RGB to CMYK image data DT
It is assumed that the color conversion process is performed. Here, C is cyan, M is magenta, Y is yellow, K
Indicates black. Of course, in the case of monochrome image data, such color conversion processing need not be performed.

The halftone processing unit 22 uses the CMYK image data DT output from the color conversion unit 21 as an input image, and converts this input image into quantized data (two or more kinds of multi-values such as binary and 128 values) for each pixel. The predetermined gradation data DR, which is the converted quantized data, is output to the image output device 30 as an output gradation value for each pixel. The halftone processing unit 22 performs various image processing for achieving the object of the present invention. Specific image processing contents will be described later.

The image output apparatus 30 includes a pulse width modulation unit 31 and a print engine 35. The print engine 35 includes a laser driver 36 and a laser diode (LD) 37.
In the image output device 30, the print engine 35 is driven in accordance with the output gradation value of each pixel output from the image processing device 20, and the dot used for each color component dot is used on a recording medium such as print paper by the print engine 35. It is formed based on a predetermined halftone dot position according to the screen,
Further, a halftone image is generated by the formed dots, and the image is output as a printed matter.

The pulse width modulation unit 31 inputs the output gradation value output from the halftone processing unit 22,
From this output gradation value, a laser driving pulse having a predetermined number of pulses and a pulse width is generated as driving data. Then, this drive data is output to the laser driver 36.

The laser driver 36 generates control data for driving the laser diode 37 from the input drive data, and outputs the control data to the laser diode 37. The laser diode 37 is driven based on the control data output from the laser driver 36, and further, a photosensitive drum and a transfer belt (not shown) are driven, and finally a recording medium such as a printing paper has a pulse width and a pulse number. In response, a predetermined number of dots of a predetermined color component are output at a predetermined position. In this way, dots of each color are formed on the recording medium, and the data to be printed generated by the image input apparatus 10 is printed.

Next, a specific hardware configuration of the image processing apparatus 20 including the halftone processing unit 22 that performs image processing to which the present invention is applied will be described with reference to FIG. Here, the color conversion unit 21 and the halftone processing unit 22 constituting the image processing apparatus 20 of FIG.
, Corresponding to RAM 27 and ROM 28.

The image processing apparatus 20 includes an input / output interface (I / F) 25, a CPU 26, and a RAM 2.
7, ROM 28, hard disk 29, etc., which are connected to each other via a bus. The input / output I / F 25 serves as an interface between the image input apparatus 10 and the image processing apparatus 20 and an interface between the image processing apparatus 20 and the image output apparatus 30. The input / output I / F 25 has an RG from the image input device 10 transmitted by a predetermined transmission method.
B image data is input and converted into data that can be processed by the image processing apparatus 20. RGB
Is temporarily stored in the RAM 27. Then, the output gradation value for each pixel is sent from the input / output I / F 25 to the image output device 30 using a predetermined transmission method.

The CPU 26 reads out a program stored in the hard disk 29 or the ROM 28 via the bus and executes the read program under a predetermined operating system, thereby functioning as the image processing device 20 and executing a predetermined process. To do. In particular,
A pixel group dividing unit 22a, a centroid position calculating unit 22b, a centroid position determining unit 22c,
It functions as an arrangement unit 22d and a gradation conversion unit 22e.

The pixel group dividing unit 22a divides the input image into cells based on the halftone dot position corresponding to the AM screen used in the image output device 30. The center-of-gravity position calculation unit 22b calculates the center-of-gravity position of the cell from the input gradation value and coordinates of each pixel in the cell. The center-of-gravity position determination unit 22c adds or subtracts a predetermined value to the center-of-gravity position calculated for a predetermined cell, and then performs integer processing to determine the center-of-gravity position. The placement unit 22d places the center of the gradation conversion matrix at the determined barycentric position. The gradation converting unit 22e refers to the gradation conversion table stored in the arranged gradation conversion matrix and converts the input gradation value to the output gradation value for each pixel of the cell. In this way, each unit manages the processing performed by the above-described halftone processing unit 22, and specifically executes the processing shown in the flowcharts of FIGS. 6 and 7 described later.

The program in which these processes are recorded may be stored in advance in the hard disk 29 or the ROM 28. For example, the program is supplied from the outside by a computer-readable recording medium such as a DVD-ROM or a CD-ROM, and is not illustrated. The program may be stored by being stored in the hard disk 29 via the program reading unit. Of course, the hard disk 29 or RAM is accessed by accessing a server or the like that supplies a program via a network means such as the Internet and downloading data.
27 may be stored.

As described above, in this embodiment, the input image composed of the CMYK image data DT output from the color conversion unit 21 is data represented by multi-gradation values of 8 bits for each color, Assume that each color has “0” to “255” and 256 types of gradation values. In the following description, it is assumed that the halftone processing unit 22 performs a quantization process for controlling the size of a dot to be output for each color component. As described above, the image output device 30 generates the pulse width and the number of pulses corresponding to the gradation-converted predetermined gradation data, that is, the output gradation value for each pixel,
Printing is performed by controlling the output of dots of each color component.

Next, the outline of the tone conversion processing performed by the halftone processing unit 22 is shown in FIG.
Will be described with reference to FIG. The image processing of the present invention is performed in order to facilitate understanding of the processing flowcharts shown in FIGS. 6 and 7 to be described later.

FIG. 3 shows an input image in which square pixels are arranged in a matrix and 9 pixels in the horizontal (X) direction,
It is an explanatory diagram showing a state in which each pixel of the input image 100 is divided into cells, assuming that each pixel is an image whose center position is represented by X and Y coordinates. is there. As shown in FIG. 3, halftone dot positions AMT (black circles in the figure) corresponding to the AM screen to be used are regularly set for the input image 100, and the X, Y coordinates are (1, 1) or A total of 12 halftone dot positions AMT, such as points located at (5, 3), etc., are set to coincide with the center of the pixel. The input image 100 is divided into cells around this halftone dot position AMT (
In the figure).

Here, it is assumed that the cell shape is symmetrical with respect to the halftone dot position, and the cell shape is divided so as to be all the same shape. This is for facilitating understanding of processing contents to be described later with respect to the processing when the divided shapes of the cells, which are the object of the present invention, are different.

Cell division is performed in raster order, and as shown in FIG. 3, cells C (1) to C
It is divided into 12 cells C (N) (N = 1 to 12) up to (12). By the way, cell C
The pixels included in (12) are the X and Y coordinates (7, 6), (8, 6), (6, 7), (7, 7
), (8, 7), (6, 8), (7, 8).

Next, FIG. 4A shows an example of a gradation conversion matrix 200 applied when gradation conversion is performed on the input gradation values of the pixels included in each divided cell C (N). In addition, in the applied gradation conversion matrix 200 of FIG. 4A, the table number of the gradation conversion table to be referred to at the time of gradation conversion is stored at the position corresponding to each pixel in the cell. To obtain an output gradation value corresponding to the input gradation value of each pixel. An example of the gradation conversion table KHT stored in the gradation conversion matrix 200 is shown in FIG. The input tone value of each pixel is tone-converted to each output tone value with reference to the table number of the corresponding tone conversion table.

The conversion target pixels to which the gradation conversion matrix 200 is applied and are subjected to gradation conversion are shown in FIG.
The pixels indicated by shading in FIG. Accordingly, pixels other than the shaded pixels in the input image 100 are treated as non-conversion target pixels that are not subjected to gradation conversion processing in the present embodiment. In this case, the input gradation value of the non-conversion target pixel may be distributed to adjacent pixels prior to gradation conversion.

The stored gradation conversion table KHT converts the halftone dot position AMT into the gradation conversion matrix 200.
The table number is set so that dots can be easily formed at the tone conversion matrix center position HMC. That is, dots are easily formed at the halftone dot position AMT.

Here, the pixel corresponding to the position where the table number “3” is stored is divided such that half of the pixel is included in the cell, as indicated by the broken line in FIG. Therefore, in this embodiment, when the gradation conversion matrix 200 is applied to such a pixel, gradation conversion is performed on the input gradation value of the pixel, and a gradation value that is half of the obtained value is obtained. It is assumed that the output gradation value of the gradation conversion target pixel is added every time gradation conversion is performed. Further, the gradation conversion process may be performed for the gradation value half of the input gradation value. Of course, not half,
A predetermined ratio may be set according to the division position.

Next, the state of gradation conversion processing performed for each pixel of the cell will be described with reference to FIG. In FIG. 5A, the calculated gravity center GC (black circle in the figure) of the cell C (1) is the coordinates (2, 1), and the gradation conversion matrix 200 is displayed at the halftone dot position AMT1 of the cell C (1). The center position HMC of FIG. The numbers in the gradation conversion matrix 200 indicate the table numbers of the stored gradation conversion table KHT. Next, as shown in FIG. 5B, the center position HMC of the gradation conversion matrix 200 is shifted to the pixel located at the center of gravity. in this case,
The gradation conversion matrix 200 for one pixel is shifted to the right.

Around the shifted gradation conversion matrix 200, as shown in FIG. 5B with one of the two-dot chain lines, the gradation conversion matrix having the same shape is continuously surrounded so as to be adjacent to each other. Treat as existing. In the gradation conversion of each pixel in the cell, the gradation conversion table KHT stored in this position at a matrix position where each pixel overlaps this gradation conversion matrix or the continuously existing gradation conversion matrix. Based on the number, the input gradation value of each pixel is converted into an output gradation value.

Incidentally, the table number referred to by the pixel (2, 1) where the center of gravity is located is “1” before the shift.
”, But“ 0 ”after the shift. By shifting the gradation conversion matrix in this way, it becomes easier to form dots at the center of gravity, and an output image that is as faithful as possible to the input image can be obtained. Further, as described above, since the cell shape is divided so that the halftone dot position becomes the center of the cell, the probability that the center of gravity position becomes the halftone dot position is high.

Now, the outline of the processing is finished, and the gradation conversion processing performed by the halftone processing unit 22 will be sequentially described with reference to the flowcharts of FIGS. 6 and 7. In each process shown in these flowcharts, the CPU 26 (FIG. 2) appropriately reads a program and an arithmetic expression for executing this process from, for example, the ROM 28, or stores and reads necessary process data in, for example, the RAM 27 as appropriate. It is what you do.

When the processing shown in FIG. 6 is performed, first, in step S201, processing for reading an input image to be subjected to halftone processing, that is, gradation conversion processing is performed. As described above, the input image is assumed to be composed of square dot matrix pixels, and here, as an example, image data of the color component cyan is read as the input image. Naturally, the gradation conversion process is performed on the image data of all color components in addition to the color component cyan for the input image. In the case of a monochrome image, it is not necessary to perform gradation conversion processing for each color component.

In step S202, the read input image is divided into pixel groups (cells). The division of the cells is performed around the halftone dot positions arranged at predetermined intervals. In the example shown in FIG. 3, an AM screen whose halftone dot position substantially coincides with the center of the pixel is used. However, depending on the image output device, the AM screen to be used may be positioned at the center of the pixel. Instead, it may be located at the boundary of the pixel. Further, as described above, when processing a lot of color component data such as CMYK, for the purpose of preventing the occurrence of moire due to the mutual interference of each color component, the AM screens used for each color component are relatively relative to each other. It may rotate or shift, or use a different AM screen. In this case as well, the halftone dot position often occurs at the pixel boundary.

Therefore, in this embodiment, as shown in FIG. 8A, an AM screen is used in which a halftone dot is located at the center of the pixel and a halftone dot located at the pixel boundary is regularly arranged. And handle it as an output image. In addition, in order to simplify the description, it is assumed that the input image 100 is horizontal (X) 9 pixels × vertical (Y) 8 pixels, similarly to the description in FIG. 3. Then, when the input image has a uniform input gradation value, the halftone dot position is divided so as to be the point symmetry position of the cell so that the center of gravity of the cell coincides with the halftone dot position.

An example of the division performed in step S202 is shown in FIG. In the present embodiment, the first cell shape CK1 in which the halftone dot position (that is, the centroid position that becomes the centroid of the cell when the input image has a uniform input gradation value) is the center of the pixel, and the halftone dot position is the pixel. Second cell shape CK that becomes the boundary of
It shall be divided into two. Of course, the number of pixels constituting the cell is increased / decreased according to the halftone dot position so that the halftone dot position becomes the point symmetry position of the cell, and the dots are easily formed around the predetermined pixel position. In FIG. 8B, the number of pixels in the first cell shape CK1 after division is different from the number of pixels in the second cell shape CK2 in order to divide the cell shape. ing.

As a result of such division, the input image 100 has a cell C (1) having two different cell shapes.
) To 12 cells of cell C (12). Therefore, the maximum value N of the cell number N
The max is “12”. Further, as described above, an image indicated by a halftone dot in the figure is a pixel for gradation conversion.

Next, in step S203, a process of applying the gradation conversion matrix 200 to the upper left cell of the input image 100 is performed. As shown in FIG. 8B, cell C (1) with cell number N = 1.
Since the tone conversion matrix 200 is applied to the cell number “
“1” is stored in “N”.

Next, in step S204, a process of calculating the total input gradation value of the pixels in the cell is performed, and it is determined in the next step S205 whether or not the total is larger than “0”.

When the sum is not greater than “0” (S205: NO), the input tone value of all the pixels in the cell is “0”, and therefore tone conversion processing for dot formation is performed. Not
Proceed to step S250. In this case, the input gradation value becomes the output gradation value as it is. By doing so, the load related to the gradation conversion process can be reduced.

When the sum is larger than “0” (S205: YES), a process of calculating the barycentric position of the pixel in the cell (step S206) is performed. This process is necessary for the shift process of the center of the gradation conversion matrix in step S221.

When the coordinates of the center of gravity of the cell are (XG, YG), each coordinate is calculated using the following formulas (1) and (2).
XG = [Σ {(X coordinate of each pixel) × (gradation value of each pixel)}] / total gradation value of cell (1)
YG = [Σ {(Y coordinate of each pixel) × (gradation value of each pixel)}] / total gradation value of the cell (2)

Next, in step S215, it is determined whether or not the center-of-gravity position is a pixel boundary when the input gradation value is uniform. If the cell is a pixel boundary (S215: YES)
), The position subjected to the integer processing as it is is determined as the barycentric position (step S216), and if it is not a cell that becomes the pixel boundary (S215: NO), the value “0.5” is added to the X direction to perform the integer processing The position is determined as the center of gravity position (step S217). The processes performed in step S216 and step S217 will be supplementarily described with reference to FIG.

FIG. 9 is an explanatory diagram for explaining the integer processing for the X coordinate. Here, attention is focused on the cell C (4) having the second cell shape in which the halftone dot position becomes the boundary of the pixel. And this cell C (
Assume that the center-of-gravity position calculated for 4) is XG = 2.5, which is the pixel boundary position as shown by the solid circle in the figure.

As described above, in the subsequent processing, gradation conversion is applied to each pixel and gradation conversion is performed. At this time, the data is stored in each matrix position of the gradation conversion matrix corresponding to each pixel position. Since the gradation conversion table is referred to, next, an integer process is performed in order to make the center of gravity position coincide with the center position of the pixel.

At this time, if it is assumed that the integer processing is performed by rounding, XG as shown in FIG.
If the value of “2.5” is, the center of gravity position is the position indicated by the dotted circle in the figure, and the X coordinate is “
3 "pixel position. On the other hand, if the value is slightly below the value “2.5”, the position of the center of gravity becomes the position indicated by the black circle in the figure, and the X coordinate is determined to be the pixel position “2”. That is, the value "
The position of the center of gravity moves by one pixel depending on whether or not it exceeds 2.5 ". Therefore, in the case of an input image in which the input tone value is slightly changed, the calculated center of gravity position may slightly shift depending on the cell. As a result, the dot formation position is shifted for each cell, so that noise is generated in the output image and the graininess is increased. Further, as described above, since the cell having the second cell shape CK2 is originally divided into cell shapes in which the position of the center of gravity is easily calculated in the vicinity of the boundary of the pixel, this is likely to occur.

Therefore, as shown in FIG. 9, in the present embodiment, in the case of a cell having the second cell shape CK2,
It is assumed that integer processing is performed by rounding off the decimal point so that the center of gravity position is determined at a predetermined pixel position. For example, for the cell C (4), if the calculated centroid position is in the range of 2 or more and less than 3, the centroid position is determined as the pixel position where the X coordinate is “2”. For cells having the shape CK2, when the centroid position is in the vicinity of the halftone dot position, the centroid position is always determined at the pixel position on the left side of the boundary position. Therefore, in the process of step S216 (FIG. 6), an integerization process is performed for the calculated barycentric position by rounding off the decimal part.

By the way, the integer conversion processing method that rounds off the decimal places as described above is normally used for all divided cells in order to reduce the processing load. At this time, the integer processing (FIG. 6, step S217) performed for the cell having the first cell shape CK1 will be described.

As shown in FIG. 9, in the cell C (1) having the first cell shape CK1, when the calculated range of the centroid position is 0.5 or more and less than 1.5, the determined centroid position is Since the pixel position closest to the halftone dot position (black circle in the figure) is desirable, the integer processing is performed so that the X coordinate “1”, which is the pixel position where the halftone dot position exists, is located. However, if the integer processing for rounding down the decimal point is used as it is, if the X coordinate is 0.5 or more and smaller than 1, the pixel becomes “0” instead of the pixel of X coordinate “1”. .

Therefore, after adding a half length with one pixel as a unit length, that is, “0.5” to the calculated center of gravity position, an integerization process is performed to round off the decimal point. By doing so, for example, in the cell C (1), the calculated range of the center of gravity is 0.5 or more and 1.5.
If it is smaller, the X coordinate is 1 or more and a value smaller than 2 can be used to perform integer processing similar to that of the cell C (4). As a result of this integer processing, the barycentric position is the pixel of the X coordinate “1”. Determined as position.

In this way, by determining the center of gravity at a predetermined pixel position according to the shape of the cell, dots can be formed near the predetermined position in the cell. Therefore, for example, when the input image is an image having substantially the same input gradation value, the probability that dots are regularly formed based on the halftone dot position set according to the AM screen to be used increases. It is possible to output a halftone dot image in which the variation in the distance is suppressed. Therefore, it is possible to suppress the increase in graininess that occurs as a result of the dot formation position shifting due to a slight shift in the center of gravity position.

Returning to FIG. 6, next, in step S220, gradation conversion processing of pixels in the cell is performed. By this process, a gradation conversion matrix is applied to the cell, and the input gradation value of each pixel is output to form a dot by referring to the table number of the gradation conversion table corresponding to each pixel in the cell. Tone conversion to tone value. This process will be described later with reference to the flowchart of FIG.

Next, in step S250, it is determined whether or not the cell number N is Nmax. In the example of FIG. 8, Nmax = 12. If the cell number N is not Nmax (S250: NO)
After performing the process of setting N to N + 1 (step S251), the process returns to step S204,
The processing from step S204 to S250 described above is repeated for the next cell to be subjected to gradation conversion.

On the other hand, when the cell number N is Nmax (S250: YES), the gradation conversion to the output gradation value has been completed for all the pixels to be subjected to gradation conversion, and the halftone processing unit 22 The process ends.

Now, the gradation conversion processing of the in-cell pixel performed in step S220 of FIG. 6 will be described with reference to the flowchart of FIG. When this process is started, first, in step S221, a process of shifting the center of the gradation conversion matrix 200 to the pixel where the center of gravity is located is performed.
This processing is performed in order to obtain an output image faithful to the input image by making the dot formation position not the halftone dot position based on the AM screen to be used, but the center of gravity position of the cell, as previously described in FIG. To do.

In this embodiment, since the input image is divided into two types of cell shapes, two types of gradation conversion matrices are applied. The gradation conversion matrix 200 applied to the first cell shape CK1
FIG. 10 shows a state where a is applied to the cell C (1), and a state where the gradation conversion matrix 200b applied to the second cell shape CK2 is applied to the cell C (4).

FIG. 10 shows that the determined barycentric position is the pixel position at coordinates (1, 1) for cell C (1) and the pixel position at coordinates (2, 3) for cell C (4). Yes. Of course, when shifting the gradation conversion matrix to the determined center of gravity position, the gradation conversion matrix having the same shape is present in the periphery as described above, and two types of floors having different shapes as shown in FIG. Even in the case of the tone conversion matrix, it is assumed that there is a tone conversion matrix in which the tone conversion matrix 200a and the tone conversion matrix 200b are continuously repeated in the periphery, and these repeated tone conversion matrices are integrated. Can be handled as a shift.

Next, in step S222, in the state where the gradation conversion matrix 200a or the gradation conversion matrix 200b is applied, the gradation conversion unprocessed pixel closest to the centroid position is searched, and in step S223, the searched floor is searched. A tone conversion process is performed on the tone conversion unprocessed pixels with reference to the tone conversion table. As described above, each gradation conversion matrix stores the table number of the gradation conversion table. Then, a table number “0” is stored at the center of the gradation conversion matrix so that dots are easily formed, and each matrix position is subjected to gradation conversion so that dots are formed around this. Stores the table number to be referenced.

For example, in FIG. 10, in the cell C (1), the pixel at the coordinate (1, 1) is first searched as an unprocessed pixel to be subjected to gradation conversion, and the gradation at the position corresponding to the pixel at the coordinate (1, 1). Gradation conversion is performed by obtaining an output gradation value corresponding to the input gradation value of the searched pixel based on the table number “0” of the gradation conversion table stored in the conversion matrix.

In step S224, the searched pixel is subjected to a process of setting the gradation value after gradation conversion as an output gradation value. By this processing, an output gradation value corresponding to a dot formed at the pixel position is obtained for the pixel in the cell.

Next, in step S230, it is further determined whether or not all the pixels in the cell have been processed. If not (S230: NO), the process returns to step S222, and the above-described step S222 is performed.
To S225.

And when it determines with having processed all the pixels in a cell in step S230 (S230:
YES), the gradation conversion process of the pixel in the cell performed in step S220 (FIG. 6) is terminated.

As described above, according to the present embodiment, when the input image is divided into a plurality of cells, when a halftone dot is not located at the center of the pixel, such as when the halftone dot is located at the pixel boundary, Accordingly, division is performed into at least two types of divided shapes. Then, in the cell in which the halftone dot is located at the pixel boundary, the barycentric position can be determined as a predetermined pixel position. Therefore, the variation in the distance between dots formed by reducing the variation in the dot formation position can be suppressed, and the increase in graininess can be suppressed.

As mentioned above, although this invention was demonstrated using one Embodiment, this invention is not limited to such embodiment at all, Of course, it can implement with various forms within the range which does not deviate from the meaning of this invention. It is. Hereinafter, a modification will be described.

(First modification)
In the embodiment, when the input image is divided into a plurality of cells, the first cell shape in which the halftone dot is located at the center of the pixel and the second cell shape in which the halftone dot is located at the boundary of the pixel are mutually connected. Division into two types of division shapes with different numbers of pixels in the cell was performed. Thus, when the number of pixels constituting the cell is different, for example, when the input image has a uniform input gradation value, the output gradation value after gradation conversion differs for each cell. Accordingly, where dots of uniform size should be formed, two types of dots having different sizes are formed. Therefore, as a first modification, 2
The input image may be divided so that the number of pixels included in each cell having different types of division shapes is the same.

An example of cell division according to the first modification is shown in FIG. As shown in the figure, with respect to the first cell shape CK1a, for the pixels at the upper and lower ends of the cell, half of the input gradation values are to be subjected to gradation conversion as described above. It becomes a pixel in the cell. As a result, for the first cell shape CK1a, the number of pixels constituting the cell is six, which is the same as the six pixels that are the number of pixels in the cell for the second cell shape CK2a. As a result, it is possible to suppress the formation of non-uniformly sized dots.

(Second modification)
Further, in the above-described embodiment, it has been described that the cell is divided so that the halftone dot position is the point symmetry position of the cell, and as a result, the cell is divided into two types of cells having different pixel numbers and shapes. In this case, as described above, two kinds of gradation conversion matrices are required, and the process of switching and applying the gradation conversion matrix for each cell having a different division shape is performed, resulting in a heavy processing load. Therefore, as a second modification, the first cell shape and the second cell shape may be divided so as to be line symmetric.

An example of cell division according to this modification is shown in FIG. As shown in the figure, the dot position AM
The first cell shape CK1b in which T is the center of the pixel and the second cell shape CK2b in which the dot position AMT is the pixel boundary are divided so as to be symmetrical. In the example of division in FIG. 12, as in FIG. 8, since the cells are divided with the halftone dot positions being point-symmetrical positions, the vertical and horizontal directions are also symmetrical. Therefore, at the time of division, the division processing may be performed so that at least one of left and right or upper and lower sides is symmetrical. Of course, for example, when the number of pixels per cell is large, it is conceivable to divide symmetrically only in the left-right direction or in the vertical direction.

Next, what kind of gradation conversion matrix is applied to the two cell shapes thus divided will be described with reference to FIG. On the left side of FIG. 13 (a), the first cell shape CK
The target pixel (shaded) of gradation conversion processing is shown for 1b, and the gradation conversion matrix 201 applied to the first cell shape CK1b is also shown on the left side of FIG. 13 (b). At each matrix position of the gradation conversion matrix, the table number of the gradation conversion table referred to by the target pixel of the gradation conversion process corresponding to that position is stored as shown in the figure.

On the other hand, on the right side of FIG. 13A, the target pixel (shaded) of the gradation conversion processing is shown for the second cell shape CK2b, and the same applies to the second cell shape CK2b on the right side of FIG. 13B. A gradation conversion matrix 202 is shown. At each matrix position of the gradation conversion matrix, the table number of the gradation conversion table referred to by the target pixel of the gradation conversion process corresponding to that position is stored as shown in the figure.

As is apparent from the figure, the gradation conversion matrix 201 and the second applied to the first cell shape.
However, since the first cell shape CK1b and the second cell shape CK2b are bilaterally symmetrical, that is, reverse, the two tone conversions are different. The matrix can also be reversed so that it can be shared.

FIG. 13C shows an example of the gradation conversion matrix 203 that can be commonly applied to these two different cell shapes. As shown in the figure, with respect to the gradation conversion matrix 202 applied to the second cell shape CK2b, one matrix having the table number “3” stored in the lower right in the figure as shown by the broken line. It is added.

As is apparent from the figure, it can be seen that the matrix portion shown by shading in FIG. 13C is symmetrical (inverted) with the gradation conversion matrix 201 applied to the first cell shape CK1b. Therefore, the gradation conversion matrix 203 is converted into the first cell shape CK.
When applied to 1b, only the shaded matrix portion is turned over and used, and when applied to the second cell shape CK2b, if one matrix shown by the lower right broken line is used, it is common to each cell shape. The gradation conversion matrix thus made can be applied. Accordingly, it is possible to reduce the processing load related to memory capacity and gradation conversion.

(Third Modification)
In the embodiment and the first and second modified examples, as shown in FIG. 8B and FIG. 11, one pixel is divided to form a cell. Therefore, as described above, the divided pixels are shared pixels of cells that are treated as gradation conversion target pixels of all the cells including the pixel at the time of gradation conversion. Therefore, the gradation conversion process is performed on the input gradation value divided according to the pixel division ratio according to the number of shared cells. As a result, the number of gradation conversions increases as the number of shared pixels increases, and the load related to gradation conversion processing increases.

Therefore, as a third modified example, in order to prevent shared pixels from being generated when the input image is divided, 2
It is good also as what divides | segments into several or more cell shape. By dividing in this way, one tone conversion is sufficient for one pixel. As a matter of course, since the gradation conversion matrix is prepared in accordance with the type of cell division shape, the storage amount increases according to the number of types of gradation conversion matrix. Therefore, the number of shared pixels and the number of types of gradation conversion matrix may be determined based on the balance between the gradation conversion processing load and the storage amount.

FIG. 14 shows an example of cell division shapes according to the third modification. In the cell division example shown in FIG. 8B, the shared pixels in the second cell shape are divided so as to be included in only one cell. To divide. In this way, the number of pixels constituting the cell is increased or decreased according to the halftone dot position, or the cell shape is changed to divide the dot formation position into a preferred cell shape so that it approaches the predetermined pixel position. it can.

(Fourth modification)
In the above embodiment, as shown in FIG. 9, in the case of the second shape cell, the decimal point is rounded down to perform the integer processing, and the center of gravity position is in the range around the halftone dot position. The center of gravity position is always determined at the pixel position on the left side of the boundary position. In the present modification, conversely, the barycentric position may always be determined at the pixel position on the right side of the boundary position.

In the case of the fourth modification, in the process performed in step S216 (FIG. 6), an integerization process is performed on the calculated barycentric position by rounding up the fractional part instead of rounding down. At this time, in order to correspond to the integer processing by rounding up in the processing performed in step S217, the integer processing may be performed by subtracting 0.5 in the X direction, contrary to the above embodiment.

Of course, when the rounding method is rounded off, 0.5 is not added to the first cell shape by adding 0.5 in the X direction (step S217), but conversely the second cell shape is X direction. Then, 0.5 may be subtracted and an integer processing may be performed (step S216). Thus, 0.5 may be added or subtracted in the X direction for the first cell shape or the second cell shape, depending on the integer processing method.

Further, the image processing apparatus 20 to which the present invention is applied is configured to perform the processing in the previous embodiment and the modified example as an image output on a printing paper by a laser printer and displayed for reproduction.
The present invention is not limited to this, and various printers such as a printer that prints by ejecting ink onto a print medium such as paper, a printer that prints by thermally transferring ink, and the like Processing may be performed on the assumption that an image is reproduced and displayed on a photograph or a display.

Although the image processing apparatus 20 to which the present invention is applied is assumed to be independent from the image input apparatus 10 and the image output apparatus 30, it may be incorporated into the image output apparatus 30 such as a laser printer. The image input device 10 may be incorporated. Alternatively, the image processing apparatus according to the above-described embodiment is configured to be incorporated in various electronic devices having a computer function such as a copier, a computer device, a word processor, a fax machine, and a mobile phone in which an image output device such as a laser printer is incorporated. You may do it.

The block diagram which shows the whole system in one Embodiment with which this invention is applied. 1 is a schematic diagram illustrating a hardware configuration of an image processing apparatus according to an embodiment. The schematic diagram explaining a mode that an input image is divided | segmented into a cell in this embodiment. (A) is explanatory drawing for demonstrating the mode of a gradation conversion matrix. FIG. 4B is an explanatory diagram of table numbers to be referred to for the gradation conversion table. (A) is explanatory drawing explaining a mode that a gradation conversion matrix is applied to a cell. FIG. 6B is an explanatory diagram for explaining a gradation conversion matrix shifted to the position of the center of gravity. 6 is a flowchart for explaining gradation conversion processing in the present embodiment. The flowchart explaining the gradation conversion process routine of the pixel in a cell. (A) is explanatory drawing explaining the relationship between a dot position and the pixel position of an input image. (B) is an example of cell division based on halftone dots located at the center of a pixel and the boundary of the pixel. The schematic diagram for demonstrating the integer-izing process which determines a gravity center position. An example of a gradation conversion matrix applied to a first cell shape and a second cell shape. Explanatory drawing which shows an example of the division | segmentation of the cell in a 1st modification. Explanatory drawing which shows an example of the division | segmentation of the cell in a 2nd modification. (A) is explanatory drawing of the pixel which comprises the cell divided | segmented in the 2nd modification. FIG. 5B is an explanatory diagram of a gradation conversion matrix applied to divided cells. (C) is an explanatory diagram of a tone conversion matrix that can be used in common. Explanatory drawing which shows an example of the division | segmentation of the cell in a 3rd modification.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 ... Image input device, 11 ... Application part, 12 ... Rasterization part, 20 ... Image processing apparatus, 21 ... Color conversion part, 22 ... Halftone processing part, 22a ... Pixel group division part, 22b
... centroid position calculation unit, 22c ... centroid position determination unit, 22d ... arrangement unit, 22e ... tone conversion unit, 2
DESCRIPTION OF SYMBOLS 5 ... Input / output I / F, 26 ... CPU, 27 ... RAM, 28 ... ROM, 29 ... Hard disk, 30 ... Image output device, 31 ... Pulse width modulation part, 35 ... Print engine, 36 ... Laser driver, 37 ... Laser Diode: 100: Input image, 200 to 203: Gradation conversion matrix, HMC: Gradation conversion matrix center, KHT: Gradation conversion table, AMT: Halftone dot position, GC: Center of gravity of cell, CK1: First cell Shape, CK2 ... second cell shape.

Claims (11)

By dividing the input image into pixel groups composed of a plurality of pixels, and applying a tone conversion matrix to each of the divided pixel groups, the input tone value of each pixel included in the pixel group is output to the output image. An image processing apparatus that performs tone conversion into an output tone value having two or more types of tone values corresponding to dots constituting a halftone dot image,
A pixel group dividing unit that divides the input image so that the shape of the divided pixel group becomes at least two different divided shapes based on the halftone dot positions of the halftone image. An image processing apparatus. The image processing apparatus according to claim 1,
A centroid position calculation unit that calculates a centroid position of the pixel group using an input gradation value of each pixel included in the pixel group;
The pixel group dividing unit divides the calculated gravity center position so as to be the halftone dot position when each pixel included in the pixel group has the same input gradation value. Processing equipment. The image processing apparatus according to claim 2,
The pixel group dividing unit includes a first pixel group in which the calculated centroid position is a pixel center when the pixels included in the pixel group have the same input gradation value, and the calculated centroid position. An image processing apparatus that divides the input image into a second pixel group that becomes a pixel boundary. The image processing apparatus according to claim 3,
The pixel group dividing unit divides the input image so that the number of pixels included in the first pixel group and the number of pixels included in the second pixel group are the same. Processing equipment. The image processing apparatus according to claim 3, wherein:
The pixel group dividing unit divides the input image so that a divided shape of the first pixel group and a divided shape of the second pixel group are at least left-right symmetric or vertically symmetric. Image processing device. An image processing apparatus according to any one of claims 3 to 5,
When the barycentric position of the pixel group calculated from the input tone value of each pixel does not coincide with the center of the pixel, the barycentric position calculating unit calculates the calculated barycentric position as an integer value using a predetermined integerization processing method. An image processing apparatus, further comprising: a center-of-gravity position determination unit that determines the converted position as the center-of-gravity position of the pixel group. The image processing apparatus according to claim 6,
The center-of-gravity position determination unit sets the length of one pixel width to half the length of one of the first pixel group and the second pixel group according to the predetermined integer processing method used. An image processing apparatus, wherein a corresponding value is added to or subtracted from the calculated barycentric position of the pixel group. An image processing apparatus according to any one of claims 3 to 7,
An arrangement unit that arranges the center of the gradation conversion matrix at the center of gravity of the pixel group;
A gradation converter that refers to a predetermined gradation conversion table stored in the arranged gradation conversion matrix and converts the input gradation value of each pixel of the pixel group into an output gradation value; ,
An image processing apparatus further comprising: By dividing the input image into pixel groups composed of a plurality of pixels, and applying a tone conversion matrix to each of the divided pixel groups, the input tone value of each pixel included in the pixel group is output to the output image. An image processing method for performing gradation conversion into an output gradation value having two or more kinds of gradation values corresponding to dots constituting a halftone dot image,
A centroid position calculating step of calculating a centroid position of the pixel group using an input gradation value of each pixel included in the pixel group;
When each pixel included in the pixel group has the same input tone value, the calculated first barycentric position where the calculated barycentric position becomes the pixel center and the calculated barycentric position become a boundary between the pixels. A pixel group dividing step of dividing into two pixel groups;
When the barycentric position of the pixel group calculated from the input gradation value of each pixel does not coincide with the center of the pixel, a position obtained by converting the calculated barycentric position into an integer value using a predetermined integer processing method is used as the pixel group. A center of gravity position and a center of gravity position determination step for determining,
The center-of-gravity position determination step has a half length of one pixel width for one of the first pixel group and the second pixel group according to the predetermined integerization processing method used. An image processing method comprising adding or subtracting a corresponding value to the calculated barycentric position of the pixel group. By dividing the input image into pixel groups composed of a plurality of pixels, and applying a tone conversion matrix to each of the divided pixel groups, the input tone value of each pixel included in the pixel group is output to the output image. An image processing program that performs gradation conversion into an output gradation value having two or more kinds of gradation values corresponding to dots constituting a halftone dot image,
A center-of-gravity position calculation function for calculating a center-of-gravity position of the pixel group using an input gradation value of each pixel included in the pixel group;
When each pixel included in the pixel group has the same input tone value, the calculated first barycentric position where the calculated barycentric position becomes the pixel center and the calculated barycentric position become a boundary between the pixels. A pixel group dividing function for dividing into two pixel groups;
When the barycentric position of the pixel group calculated from the input gradation value of each pixel does not coincide with the center of the pixel, a position obtained by converting the calculated barycentric position into an integer value using a predetermined integer processing method is used as the pixel group. The computer realizes the center of gravity position and the center of gravity position determination function to determine
The center-of-gravity position determination function has a length of half of one pixel width for either one of the first pixel group or the second pixel group according to the predetermined integerization processing method used. An image processing program, wherein a corresponding value is added to or subtracted from the calculated barycentric position of the pixel group. A computer-readable recording medium on which the image processing program according to claim 10 is recorded.

Images