JPH09191403A - Picture processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To execute the dither processing of high solution for a picture which requires resolution and to hold required gradation for a picture to which tone reproducibility is requested in a picture processor executing a binarization processing by a systematic dither method. SOLUTION: Plural matrixes of the same size different in resolution and plural γ conversion tables corresponding to the matrixes are prepared as threshold matrixes for the binarization processing. The specified matrix and the γconversion table corresponding to the matrix are selected and used in accordance with a characteristic which is requested for the picture. The matrix of higher resolution is selected for the picture which requires resolution and the matrix of lower resolution is selected for the picture which requires tone reproducibility, for example. At that time, a halftone line drawing detection part 103 detects whether the line picture of halftone exists in the input picture or not. Then, γ conversion is executed by using the γ conversion table corresponding to the matrix in a γ conversion part 106 through the use of the matrix generated in a threshold matrix generation part 104 in accordance with the result, and a dither processing part 105 executes a dither processing by using the matrix.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus using a systematic dither processing method applicable to a hard copy apparatus.

[0002]

2. Description of the Related Art FIG. 20 is a block diagram showing a schematic configuration of a conventional general image processing apparatus which performs binarization processing by the systematic dither method in a hard copy apparatus. In the figure, 1 is a γ conversion unit that performs γ correction, and 2 is a dither processing unit that performs dither processing (binarization processing).

In the above structure, when the density input value is given, the γ conversion unit 1 refers to the γ conversion table to convert the data. This conversion is performed in order to correct the non-linearity existing in the input / output relationship of the hard copy device and ensure the linearity between the input density value and the final output density value.

Next, the area ratio data output from the γ conversion unit 1 is sent to the dither processing unit 2 where binarization processing is performed. Specifically, the output pixel value is determined by comparing the area ratio with the corresponding element (value) in the preset threshold matrix.

[0005]

However, in the conventional image processing apparatus as described above, since the dither processing is performed using a single threshold matrix, there are the following problems. It was

(1) If a dot concentration type threshold matrix is used to obtain gradation, the resolution is lowered.

(2) If a dot dispersion type threshold matrix is used to increase the resolution, gradation reproducibility deteriorates.

These problems will be described with reference to FIG. FIG. 8 shows a typical input / output characteristic of the hard copy device when two threshold matrices different in dot growth method are used. In the figure, the horizontal axis is the normalized dot area ratio, which shows what percentage of the unit half-tone cell area formed by the threshold matrix is on, and the vertical axis is the output density normalized by the maximum density. Is shown.

Characteristic A shown in FIG. 8 represents a dot concentration type characteristic, and characteristic B represents a dot dispersion type characteristic. FIG. 9 shows an example of these threshold matrices. (A) is a fatting type (dot concentrated type) matrix, and (b) is a Bayer type (dot dispersion type) matrix.

As shown in FIG. 8, it is understood that the dot concentration type is superior in gradation reproducibility. This difference in gradation reproducibility is due to the fact that the hard copy device cannot completely reproduce isolated dots, and is generally greatly affected by the printing process of the hard copy device. Although it changes depending on the process conditions, it is particularly remarkable in the case of an apparatus using an electrophotographic method. Although the horizontal axis of FIG. 8 is defined as the dot area ratio here, as is clear from the above description, this is not the area ratio actually output, but the color material control amount given to the hard copy device. It is more accurate to interpret.

As described above, it is extremely difficult to solve the problems (1) and (2) at the same time with only one matrix.

FIG. 14 shows a typical input / output characteristic of the hard copy device when two threshold matrices having different sizes are used. Similar to FIG. 8, the horizontal axis of FIG. 14 is the normalized dot area ratio, which shows what percentage of the unit halftone cell area formed by the threshold matrix is on, and the vertical axis is the maximum density The converted output density is shown.

Further, characteristics A and B shown in FIG. 14 represent input / output characteristics of large size and small size, respectively. From the figure, it can be seen that when the size of the matrix is large, the rise in the low density part is faster, and considering the general digital system in which the input image is quantized, the output resolution in the low density part is higher. ing.

The insensitivity characteristic in the low input region as described above is caused by the fact that the device cannot completely reproduce an isolated dot as in the above, and is greatly influenced by the printing process of the hard copy device, and the printing principle thereof. Although it varies depending on the set process conditions, it appears remarkably in the case of an apparatus using an electrophotographic method. It should be noted that it is more accurate to interpret that the horizontal axis of FIG. 14 is not the area ratio actually output, but the color material control amount given to the hard copy device.

Therefore, it is extremely difficult to solve the problems (1) and (2) at the same time with only one matrix as in the above.

The present invention has been made in view of the above problems, and maintains gradation for an image requiring gradation such as a natural image, and provides halftone characters or patterns. It is an object of the present invention to provide an image processing apparatus capable of performing high resolution dither processing on an image requiring high resolution.

It is another object of the present invention to provide an image processing apparatus capable of improving gradation reproducibility in a low density area while maintaining a required resolution in a density area requiring high resolution.

[0018]

The image processing apparatus according to the present invention is configured as follows.

(1) In an image processing apparatus that performs image processing using the systematic dither method, a plurality of threshold matrices having different resolutions for dither processing and a γ conversion table for performing a plurality of γ conversions corresponding thereto are provided. There is provided a control means which is prepared and which selects and uses a specific threshold matrix and a γ conversion table corresponding to the specific threshold matrix according to the characteristics required for the image.

(2) In the image processing apparatus of the above (1), two types of matrices of the same size having different resolving powers are prepared as threshold matrices, and a dot concentration type matrix is adopted as the lower resolving power matrix, The control means has a window having a size equal to the number of rows and columns of the threshold matrix, and when a line drawing having a halftone is detected in the window area, the matrix having the higher resolution is selected, and the other matrix is selected. In the case of an image, the dot-concentrated matrix having the lower resolution is selected.

(3) In the image processing apparatus of (1) or (2) above, the matrix having the higher resolution has dots arranged on a diagonal line and the number of pixels in the matrix is maintained at a pseudo level of 45. Provided a screen angle of degree.

(4) In the image processing device according to (1) or (2), the matrix having the higher resolution has an even number of rows and columns, and half of the pixels in the matrix are the basic pixels. The basic pattern is provided with a screen angle of 45 degrees.

(5) In an image processing apparatus that performs image processing using the systematic dither method, a plurality of threshold matrices having different sizes for dither processing and a γ conversion table for performing a plurality of γ conversions corresponding thereto are provided. There is provided a control means which is prepared and which selects and uses a specific threshold matrix and a γ conversion table corresponding to the specific threshold matrix according to the characteristics required for the image.

(6) In the image processing device according to (5), a plurality of matrices having the maximum number of rows as the threshold value matrix and the number of rows and the number of columns thereof are integer multiples of the other matrices and the number of columns are prepared. The control means has a window having a size equal to the number of rows and columns of the maximum size matrix, and selects successively smaller size matrices as the average density of the image detected in the window area increases. I chose

(7) In the image processing apparatus of (5) above, two types of matrices having different sizes are prepared as threshold matrices, and the number of rows of the smaller size matrix is the number of rows of the larger size matrix. And the number of columns of the larger matrix is an integer multiple of the number of columns of the smaller matrix,
The control means has a window consisting of the number of rows of the smaller matrix and the number of columns of the larger matrix, and the size is determined when the average density of the image detected in the window area is lower than a predetermined value. The larger matrix is selected, and the smaller size matrix is selected when the average density is higher than a predetermined value.

[0026]

BEST MODE FOR CARRYING OUT THE INVENTION

(First Embodiment) FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention. In the figure, 101 is a shifter for thinning out input values, 102 is a line buffer for storing image data for n rows, and 103 is data for a line drawing in which a predetermined area of the input image is a halftone display. A halftone line drawing detection unit for judging whether or not it is, a reference numeral 104 is a threshold matrix generation unit for selecting a predetermined threshold matrix according to the judgment result of the line drawing detection unit 103, and 106 is the line drawing detection unit 10
A γ conversion unit (control unit) that selects a γ conversion table based on the determination result of 3 and performs γ conversion using the γ conversion table, and a dither processing unit (control unit) 105 that performs dither processing using a given threshold matrix. Is.

Here, as a preparation before the description of the operation, the real coordinates (X, Y) and the block coordinates (U, V) are introduced, and the relationship between them will be described with reference to FIG. If the size of the threshold matrix used here is n rows and n columns (nxn), the figure shows the relationship between the real coordinates and the block coordinates when n = 4. As is clear from the figure, (U, V) and (X,
The relationship of Y) is

[0028]

[Equation 1]

Is given by However, floor (A) represents the maximum integer not exceeding A.

Based on the above description, the operation will be described below with reference to the flowchart of FIG. 3 and FIG.

The input image signal is thinned out by the shifter 101 and the bit width is compressed to a required output precision. This is performed in order to reduce the capacity of the line buffer 102 by dropping extra bits, because the maximum number of gradations that can be reproduced is defined by the size of the threshold matrix and the number of gradations that one pixel has. is there.

Here, when the gradation reproduction number of one pixel is p, the maximum number T of gradation reproduction is

[0033]

[Equation 2]

If it is determined that the condition is satisfied, it is possible to represent all the states of the output. (However, the number of reproducible gradations is usually smaller than T due to the non-linearity of the output characteristics.) The input value subjected to the thinning process is the line buffer 10
2 is stored. The line buffer 102 stores input values in an n × W rectangular area, where W is the matrix of the image. Subsequent processing is performed with the threshold matrix size n.
The block processing will be described for each xn.

In the flowchart of FIG. 3, step S10
The loop parameters U and V in 1 and S102 are the block coordinates described above. Steps S101 and S10
When the block coordinates are designated in 2, the data of the n × n rectangular area at the designated coordinates is sent to the halftone image detection unit 103 in step S103.

In step S104, a halftone image is detected by a predetermined method (a halftone image detecting method will be described later). In step S105, the result of halftone image detection is determined, and if the result is true, the process proceeds to step S106. In step S106, a dot dispersion type threshold matrix is selected, and a γ conversion table corresponding to the matrix is selected, and the process proceeds to step S108.

If the result is false in step S105, the process proceeds to step S107, where a dot concentration type threshold matrix is selected, a γ conversion table corresponding to the matrix is selected, and the process proceeds to step S108. move on.

FIGS. 4A and 4B show the dot concentrated type matrix and the dispersed type matrix, respectively. The lumped type matrix of FIG. 4A is of a type generally called a fatning type, and the distributed type matrix of FIG. 4B has a threshold distribution arranged diagonally. It has a screen angle of 2 degrees and doubles the apparent resolution in the vertical and horizontal directions.

In step S108, dither processing is performed using the selected γ conversion table and the created threshold matrix, and the result is output (note that the dither processing here is the γ of the γ conversion unit 106 in FIG. 1). The conversion and the dither processing of the dither processing unit 105 are included). Then, through steps S109 and S110, the above-described operation is performed by Uma.
The process ends by repeating up to x and Vmax.

Next, the above-described step S104 and the halftone line drawing detection processing in the line drawing detection unit 103 will be described. FIG. 5 is a flowchart showing an example of the halftone image detection processing.

This processing is premised on a line drawing such as a character especially as a halftone pattern requiring a resolution, and is basically a processing method belonging to pattern matching using a window.

First, in step S201, the pixel values in the nxn region are scanned, and the input values are made into a histogram within the range of D = 0 or Dl≤D≤Dh. Here, Dl is the characteristic B of FIG.
As shown in, the minimum density that can be represented by the distributed matrix used. Dh is an input value at the time when the target halftone image is output and it is almost indistinguishable from the distributed type even if the concentrated type matrix is used, and is determined by subjective evaluation. .

The reason Dl is needed is to avoid distributed matrix application for low input values.
Dh is provided because it is meaningless to perform the following processing on an input value higher than this. Further, D = 0 is necessary for extracting a white background line drawing embedded in the background, for example.

In step S202, the created histogram is inspected to select symbols having a frequency P satisfying Kmin <P <Kmax and tabulated. The number of symbols thus selected is W. The thresholds Kmin and Kmax are the minimum value and the maximum value of the number of “1” elements in the prepared matching template. This condition is necessary to efficiently perform the subsequent processing because Kmin and Kmax are relatively close to each other because the detection target is a line drawing.

Next, after setting False to the matching result in step S203, the steps S204 to S204 are performed.
A matching routine composed of 211 is entered.

The loop parameter i set in step S204 is the address of the table created by the above processing, and W is the number of elements in this table as described above.

In step S205, the loaded nx
Scan the n region and set the pixel equal to the i-th table value to 1
And binarize unequal pixels to 0. Step S20
In 6 to S209, the prepared T types of template images are matched with the image binarized in step S205. If they are not matched, the process returns from step S211 to S204 and the same processing is repeated.
If it is determined in step S208 that a match has occurred, true is substituted in the matching result in step S210,
Exit the process.

Finally, in step S212, the matching result is output, and the process ends. The template image used here is an nxn binary image in which the target line drawing portion is configured by "1" and the other portions are configured by "0". The matching determination in step S208 is performed on the condition that all pixel values of the detected image and the template image are equal.

(Second Embodiment) This embodiment is applicable only when the size n of the area is an even number, and the structure of the apparatus is the same as that of the first embodiment described above, and therefore the description thereof is omitted. .

FIG. 6 is a flowchart showing the processing procedure of this embodiment. In the figure, portions having the same step numbers as those in FIG. 3 have the same functions as those described in the first embodiment, and therefore description thereof will be omitted, and only steps S301 and S302 that are functions unique to this embodiment will be described. .

The matrix selected when it is determined to be a halftone image in step S301 is, for example, as shown in FIG. 7 (FIG. 7 shows the case of n = 4). The matrix selected in step S302 is the matrix shown in FIG.

The matrix shown in FIG. 7 has a screen angle of 45 degrees to substantially double the resolution in the vertical and horizontal directions as compared with the matrix shown in FIG. 4A, and the matrix shown in FIG. It has twice the number of output states as compared to simply halving it in the direction. Further, since it is basically a dot concentration type, it is possible to accelerate the rising of the input / output characteristics in the low density region as compared with the matrix of FIG.

As described above, the matrix used when the halftone image is detected has half the number of pixels of the original matrix,
By adopting a screen with a screen angle of 45 degrees, a halftone image can be clearly expressed even at a lower density.

(Third Embodiment) FIG. 10 is a block diagram showing the configuration of the third embodiment of the present invention. In the figure,
201 is a shifter for thinning out input values, 202 is a line buffer for storing n lines of image data,
Reference numeral 203 is a highlight detection unit for determining whether or not a predetermined area of the input image is low density data, and 204
Is a threshold matrix generation unit that selects a predetermined threshold matrix according to the determination result of the detection unit 203, and 206 is the detection unit 203.
A γ conversion unit (control unit) that selects a γ conversion table based on the determination result of (1) and performs γ conversion using the γ conversion table, and 205 is a dither processing unit (control unit) that performs dither processing using a given threshold matrix. is there.

Next, assuming that the real coordinates (X, Y) and the block coordinates (U, V) have the relationship shown in FIG.
The operation will be described with reference to the flowchart of FIG.

The input image signal is thinned out by the shifter 201 and the bit width is compressed to the required output precision. This is because the maximum number of gray scales that can be reproduced is defined by the size of the threshold matrix and the number of gray scales that one pixel has, as in the above-described embodiment, so that extra bits are dropped to reduce the capacity of the line buffer 202. This is done in order.

Here, when the gradation reproduction number of one pixel is p, the maximum number T of gradation reproduction is

[0058]

(Equation 3)

If it is determined that the condition is satisfied, it is possible to express all the states of the output. (However, normally, the number of reproducible gradations is smaller than T due to the nonlinearity of the output characteristics.) The input value subjected to the thinning process is the line buffer 20.
2 is stored. The line buffer 202 stores input values in a rectangular region of n × W, where W is the image matrix. Subsequent processing is performed with the size n of the threshold matrix.
A description will be given assuming that block processing is performed for each xm.

In the flowchart of FIG. 11, step S4
The loop parameters U and V in 01 and S402 are
These are the block coordinates described above. Steps S401 and S4
When the block coordinates are designated by 02, step S403
Then, the data of the n × m rectangular area at the designated coordinates is sent to the highlight detector 203.

In step S404, highlight detection is performed by a predetermined method (the highlight detection method will be described later). In step S405, the highlight detection result is determined, and if the result is true, the process proceeds to step S406. In step S406, an nxm threshold matrix is created from the threshold data prepared in advance, and a γ conversion table corresponding to the matrix is selected, and step S408 is performed.
Proceed to.

If the result of the above step S405 is false, the process proceeds to step S407, an nxm matrix is created with the threshold matrix of pxq prepared in advance as the kernel, and the γ conversion table corresponding to that matrix is created. Is selected and the process proceeds to step S408. Here, p and q are divisors of n and m, respectively, and the threshold matrix created in step S407 is a cycle (p,
q) is repeated. In FIG. 12, n = m = 8,
It is an example of a matrix created for the case of p = q = 4.

In step S408, the selected γ
Dither processing is performed using the conversion table and the created threshold matrix, and the result is output (note that the dither processing here includes the γ conversion of the γ conversion unit 206 and the dither processing of the dither processing unit 205 in FIG. Things). And
The above operation is performed by Uma through steps S409 and S410.
The process ends by repeating up to x and Vmax.

Next, the above-described step S404 and the highlight detection processing in the detection unit 203 will be described. FIG. 13 is a flowchart showing an example of highlight detection processing.

First, in step S501, the average value of the data in the sent area is calculated. The number of pixels N used for this averaging calculation is given by the following equation, where Mx is the number of pixels in the X direction and My is that in the Y direction.

[0066]

(Equation 4)

Next, in step S502, it is determined whether the calculated average value is smaller than the threshold Dth (Dth will be described later). At this time, if it is determined to be true, the process proceeds to step S503,
The maximum and minimum values in the area are calculated. Step S
At 504, the difference between the maximum and minimum values is checked to determine if it is less than threshold Ath.

Here, Ath is necessary for selecting an image that should not be regarded as a highlight even if the average value is small. For example, when a halftone fine line is present on a white background, the average value is expected to be small, but processing this with a matrix of a large size causes deterioration of the image due to a reduction in resolution. Therefore, it can be said that the image regarded as the highlight portion has a low average density and a low spatial frequency.

If it is determined to be true in step S504, it is regarded as a highlight portion, and step S5
At 05, true is substituted for the determination result. On the other hand, if it is determined to be false in step S502 or step S504, it is determined that it is not the highlight part, and step S
At 506, false is substituted for the determination result. Finally, step S
The determination result is output at 507, and the processing ends.

The highlight detection processing method described here is an example, and the condition "when the average density in the area is low and the spatial frequency of the image is sufficiently low is regarded as the highlight portion" is satisfied. It goes without saying that any method may be used as long as it is.

Although the series of operations has been described above, the Dth and γ conversion table will be finally described. Here, the characteristic A shown in FIG. 14 is regarded as the input / output characteristic when a large-sized matrix is used, and the characteristic B is regarded as the input / output characteristic when a small-sized matrix is used. Further, the output side quantization step is ΔD, the input side quantization steps of the characteristic A and the characteristic B are ΔKa and ΔKb, respectively, and an output is generated for an input smaller than Kth because the characteristic B is the insensitive characteristic. Instead, the output Dth is generated when the input is Kth.

[0072]

(Equation 5)

FIG. 15A shows a table obtained as the inverse conversion of the characteristic A, and FIG. 15B shows a table obtained as the inverse conversion of the characteristic B. In the same figure, when the input is smaller than Tth, the table of (a) is selected, so the state value for the large matrix is output, and when it is Tth or more, the table of (b) is selected, so the state value for the small matrix is Is output.

As described above, since the threshold matrix and the γ conversion table corresponding to the threshold matrix are used according to the determination result of the highlight detector 203, the output characteristic is kept linear.

Although it is assumed in the above description that there are two types of threshold matrices used, step S502 in FIG.
Needless to say, three or more types of threshold matrices can be used if the determination based on Dth is performed by a plurality of thresholds and the number of matrices corresponding to the determination outputs is prepared.

(Fourth Embodiment) FIG. 16 is a block diagram showing the configuration of the fourth embodiment of the present invention. In FIG.
The parts having the same reference numerals as have the same function. In the present embodiment, when the size of the large threshold matrix is nxm and the size of the small threshold matrix is pxq, p = n / 2 and n is limited to an even number.

When the image data is input in the configuration of FIG. 8, it is thinned out by the shifter 201, and this data is stored in the line buffer 202. The line buffer 202 stores the input value in a rectangular area of n / 2 × W, where W is the number of columns of the image. The subsequent processing will be described as a block processing for each size n / 2 × m of the threshold matrix.

Reference numeral 301 is a history buffer, which stores the type of matrix selected in the previous block processing.
Reference numeral 302 is a threshold matrix generation unit, and reference numeral 303 is a two-state switch connected to either contact according to the contents of the buffer 301. Hereinafter, the processing after the line buffer 202 will be described with reference to the flowcharts of FIGS.

First, when the loop parameter V is designated in step S601, its even number,
An odd number is judged. When it is determined in step S602 that V is an even number, the process branches to the loop including steps S603 to S612.

When block coordinate U in the horizontal direction is set as a loop parameter in step S603, step S603 is executed.
At 604, the pixel value of the n / 2xm area is highlighted by the highlight detection unit 2.
03, and highlight detection is performed in step S605. The processing in step S605 is the same as that described in the third embodiment. Next, if it is considered to be a highlight in step S606, the process branches to step S607 to create an n / 2xm matrix composed of the upper n / 2 row components of the large threshold matrix. At the same time, a γ conversion table corresponding to a large matrix is selected.

Next, in step S608, 1 is set at the position of the address U in the history buffer 301. On the other hand, if it is determined in step S606 that it is not the highlight portion, the process branches to step S609, where (n / 2
A small threshold matrix having a size of (xq) is laterally repeated with a cycle q to create a threshold matrix of n / 2xm, and a γ conversion table corresponding to the threshold matrix is selected. Then, in step S610, 0 is set to the position of the address U in the history buffer 301.

In step S611, the dither processing is performed using the threshold matrix created or selected by the above-described processing and the γ conversion table corresponding thereto. Then, the process returns from step S612 to S603, and the above processing is repeated.

When it is determined in step S602 that V is an odd number, the process branches to the loop composed of steps S613 to S619. In step S614,
The value stored in the address U of the history buffer 301 is referred to. This value represents the type of threshold matrix used in the process at the block coordinates (U, V-1).

Next, when the value is judged to be 1 in S615, an n / 2xm matrix composed of the components of the lower n / 2 rows of the large threshold matrix is created, and the γ conversion table corresponding thereto is created. Is selected. In addition, step S615
If it is determined that the value is 0, the matrix is created and the table is selected by the same method as in step S609. Next, in step S618, dither processing is performed, the process returns from step S619 to S613, and the above processing is repeated.

Then, after the above-described processing is repeated between steps S620 and S601, the processing of the last block is performed and this processing is terminated.

As described above, two kinds of threshold matrixes having different sizes and the γ conversion table corresponding to the threshold matrices are prepared, and when the number of columns of the smaller matrix is ½ of that of the larger matrix, even blocks are used. Highlight detection is performed, the result is recorded in the history buffer 301, and in the processing of odd-numbered blocks, the detection result in the same block column of the immediately preceding block row is used to process using the same matrix and table as the immediately preceding block row. The line buffer 20
The capacity of 2 can be cut in half.

[0087]

As described above, according to the present invention,
Prepares multiple threshold matrices with different resolutions and corresponding γ conversion tables to detect the characteristics of the image in the input area, and has excellent gradation reproducibility for images that require gradation reproducibility. By selecting a matrix with excellent resolving power for images that require high resolving power, the gradation can be maintained for images that require tonality, such as natural images, and For images requiring high resolution, such as characters and patterns, high resolution processing can be performed and dither processing with less image degradation can be performed.

Further, according to the present invention, a plurality of threshold matrixes having different sizes and a γ conversion table corresponding thereto are prepared to discriminate the density range of the input area. For example, a large threshold matrix is set in the low density range. By selecting and selecting a small threshold value matrix in the region of medium density or higher, it is possible to improve the gradation reproducibility in the low density region while maintaining the resolution in the region where the resolution is required.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a first exemplary embodiment of the present invention.

FIG. 2 is an explanatory diagram showing a relationship between real coordinates and block coordinates.

FIG. 3 is a flowchart showing the operation of the first embodiment.

FIG. 4 is an explanatory diagram showing an example of dot concentration type and dot dispersion type threshold matrices.

FIG. 5 is a flowchart showing an example of halftone image processing.

FIG. 6 is a flowchart showing the operation of the second embodiment of the present invention.

FIG. 7 is an explanatory diagram showing an example of a threshold matrix used in the second embodiment.

FIG. 8 is an explanatory diagram showing an example of input / output characteristics of a hard copy device.

FIG. 9 is an explanatory diagram showing an example of a fatting type matrix and a Bayer type matrix.

FIG. 10 is a block diagram showing a configuration of a third exemplary embodiment of the present invention.

FIG. 11 is a flowchart showing the operation of the third embodiment.

FIG. 12 is an explanatory diagram showing an example of a matrix created in the third embodiment.

FIG. 13 is a flowchart showing an example of highlight detection processing.

FIG. 14 is an explanatory diagram showing another example of the input / output characteristics of the hard copy device.

FIG. 15 is an explanatory diagram showing γ conversion characteristics.

FIG. 16 is a block diagram showing a configuration of a fourth exemplary embodiment of the present invention.

FIG. 17 is a flowchart showing the operation of the fourth embodiment.

FIG. 18 is a flowchart showing the operation of the fourth embodiment.

FIG. 19 is a flowchart showing the operation of the fourth embodiment.

FIG. 20 is a block diagram showing a schematic configuration of a conventional example.

[Explanation of symbols]

 Reference Signs List 101 shifter 102 line buffer 103 halftone image detection unit 104 threshold matrix generation unit 105 dither processing unit (control means) 106 γ conversion unit (control means) 201 shifter 202 line buffer 203 highlight detection unit 204 threshold matrix generation unit 205 dither processing Section (control means) 206 γ conversion section (control means)

Claims (7)

[Claims] 1. An image processing apparatus for performing image processing using the systematic dither method, wherein a plurality of threshold matrices having different resolving power for dither processing and a γ conversion table for performing a plurality of γ conversions corresponding thereto are prepared. An image processing apparatus is provided with a control unit that selects and uses a specific threshold value matrix and a γ conversion table corresponding to the specific threshold value matrix according to the characteristics required for the image. 2. A threshold matrix is provided with two types of matrices of the same size having different resolving power, and a dot concentration type matrix is adopted as the lower one of the resolving power, and the control means is the number of rows and columns of the threshold matrix. If a line drawing with a window size equal to the number of lines and halftones is detected in the window area, the matrix with the higher resolution is selected, and with other images, the dot with the lower resolution is selected. The image processing apparatus according to claim 1, wherein a centralized matrix is selected. 3. A matrix having a higher resolution has dots arranged diagonally, and a screen angle of 45 degrees is artificially provided while keeping the number of pixels in the matrix. Alternatively, the image processing device according to item 2. 4. The matrix having a higher resolution has an even number of rows and columns, and is composed of a basic pattern having a screen angle of 45 degrees with half the number of pixels in the matrix as basic pixels. The image processing apparatus according to claim 1, wherein the image processing apparatus is provided. 5. An image processing apparatus for performing image processing using the systematic dither method, wherein a plurality of threshold matrices having different sizes for dither processing and a γ conversion table for performing a plurality of γ conversions corresponding thereto are prepared. An image processing apparatus is provided with a control unit that selects and uses a specific threshold value matrix and a γ conversion table corresponding to the specific threshold value matrix according to the characteristics required for the image. 6. A plurality of matrices having the maximum number of rows and the number of columns of the matrix having the maximum size are integer multiples of the other matrix and the number of columns are prepared as the threshold matrix, and the control means sets the rows of the matrix of the maximum size. 6. The image processing according to claim 5, further comprising a window having a size equal to the number of columns and the number of columns, and successively selecting a matrix of a smaller size as the average density of the image detected in the window region increases. apparatus. 7. A matrix of two types having different sizes is prepared as the threshold matrix, and the number of rows of the smaller matrix is half the number of rows of the larger matrix, and the matrix of the larger matrix is The number of columns is an integer multiple of the number of columns of the smaller size matrix, and the control means has a window consisting of the number of rows of the smaller size matrix and the number of columns of the larger size matrix, and its window. When the average density of the image detected in the area is lower than a predetermined value, the larger size matrix is selected, and when the average density is higher than the predetermined value, the smaller size matrix is selected. Item 5. The image processing device according to item 5.