JPH1070656A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processor by converting multi-gradation image data into binary image data by the mean error minimum method or the error spread method without causing deterioration in the gradation. SOLUTION: A data correction means 2 multiplies an error component 8x+i ,y+1 spread by a binarized picture element with a weight coefficient wij of a normalized weight matrix and adds the product to multi-gradation image data dxy to obtain corrected image data Dxy . A binarization means 5 compares the calculated correction image data Dxy with a threshold level T of a threshold level buffer 6 to binarize the data. An error calculation means 7 calculates the error exy based on the correction image data Dxy from the data correction means 2, binarized data Oxy from the binarization means and an output dot conversion value B from an output dot conversion value storage buffer 8. The output dot conversion value B stored in the output dot conversion value storage buffer 8 is set optionally by a setting means.

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 for binarizing multi-tone image data using a minimum error method or an error diffusion method, and is applicable to printers, copiers, facsimile machines, and the like. Things.

[0002]

2. Description of the Related Art In general, when outputting multi-tone surface image data to a printer capable of binary output only, or reducing the data capacity to reduce the data storage capacity and increase the transfer speed. To reduce the number, a binarization process for reducing the number of gradations of each pixel is performed. Various methods are known for the binarization process. Among them, an error diffusion method and an average error minimization method, which have excellent characteristics that high resolution and continuous gradation control are possible, are widely used. Used.
In particular, in the case of inexpensive ink jet printers whose production volume has been greatly increased in recent years for personal use, the rate of adopting these methods as a binarization processing method is particularly high.

In the error diffusion method, a quantization error generated when a certain pixel is binarized is weighted and distributed to neighboring pixels that have not been binarized. Also,
The average error minimizing method corrects the data value of a pixel of interest with a weighted average value of quantization errors generated in neighboring binarized pixels. The error diffusion method and the average error minimization method are logically equivalent, except that only when the error diffusion operation is performed. The error diffusion method and the average error minimum method are sometimes employed in the above-described printers for personal use, and various patent applications have been filed in recent years. For example, in the example using the error diffusion method, Japanese Patent Laid-Open No.
There are inventions described in 284173, JP-A-7-226841, and the like.

A general method of the average error minimization method is to correct a multi-tone pixel of interest from a peripheral pixel that has already been binarized by a weighted error, and to correct the multi-tone pixel by a preset threshold.
Perform value conversion. Naturally, the error generated at that time is newly diffused to another peripheral pixel. A matrix obtained by assigning a weighting coefficient to a matrix of peripheral pixels that has already been binarized is called a weight matrix, and FIG. 9 shows an example of a weight matrix. “*” In FIG. 9 indicates the target pixel,
The error calculated from the pixel that has been converted into a value is multiplied by a weighting coefficient corresponding to FIG. 9 to calculate the total error received by the target pixel.

In this case, the weight of the weight matrix is
The weight is divided by the sum of all the weights. In other words, the processing is performed as a normalized error. This can be expressed by the following equation (1). D _xy = d _xy + Σ (ex _{+ i, y + j} · wi _{, j} / Σwi _{, j} ) (1) where x, y: target multi-tone pixel coordinates i, j: weight matrix coordinates (however, _Dxy : target multi-tone pixel data ex _{+ i, y + j} : error of peripheral pixels _wij : coefficient of weight matrix _Dxy : correction pixel data

The corrected pixel data _Dxy is binarized by a preset threshold value T. Binarizing method, "0" the output data if the corrected pixel data D _xy is smaller than the threshold T, the output data if the corrected pixel data D _xy is equal to or greater than the threshold value T is set to "1". Here, when D _xy = T, the output data is set to “1”. However, even if the output data is set to “0”, the output image hardly changes. These can be expressed as follows. IF D _xy <T THEN O _xy = 0 ELSE O _xy = 1 where T: threshold value O _xy : output binary data

Then, an error in the target pixel is calculated based on the corrected pixel data D _xy and the output binary data O _xy .
Here, generally, the value of the output binary data O _xy is “0”
, The corrected pixel data D _xy itself becomes an error, and if the value of the output binary data O _xy is “1”, the value obtained by subtracting the maximum value in the input image data from the corrected pixel data D _xy is It becomes an error. These are represented by the following equations. e _xy = D _xy −O _xy · d _max where d _{max is} the maximum value of the input image data.
No. 26841, and the like. Theoretically, according to these methods, it is possible to perform a binarization process in which the density information of the entire image is stored in the input image and the output image.

FIG. 10 is a schematic diagram when dots are printed by a printer device. The squares in the figure merely represent 1 as the resolution of the printer device for convenience, and black circles indicate dots. Generally, when dots are printed by a printer device, the dots have a circular shape or a shape close to a circle. In other words, the dot shape of the printer is designed to be as close to a perfect circle as possible.
Although there are various reasons for this, basically, when the printing dot is a perfect circle, the image quality is improved and the stability of the image is also increased.

[0009] Here, for the sake of convenience, the description will be made assuming that the dots of the printer device are perfect circles. Now the dot diameter is 1
If it is equal to, the output image is as shown in FIG. In this case, for example, even if all the images are printed (even if the recorded portion is completely filled with dots), a gap is opened between the dots and the image is not a solid black image. Further, in this case, problems such as deterioration of image quality of characters and the like also occur. For this reason, in a printer apparatus, usually, the dot diameter is made larger than 1 as shown in FIG. 10B to improve the reproducibility of solid black. In particular, the printing apparatus of the ink jet system tends not only to reproduce the solid black, but also to have a larger printed dot due to a problem such as bleeding of the ink on the recording paper.

[0010] A method using a minimum average error method or an error diffusion method
The gradation expression by the value image is an area gradation method. this is,
In this method, gradation is represented by the area ratio between black dots and white dots (non-printing areas). When gradation expression is accurately performed by the area gradation method, the area of the black dots is accurately determined by the resolution of the printer device. (1 × 1 in FIG. 10). However, for the above-described reason, in the conventional printer device, the area of the black dot is larger than the area of one regular dot.

FIG. 11 shows output density characteristics (relationship between input image data and optical density) when input image data is output by the average error minimization method without particularly performing tone correction (hereinafter referred to as γ correction). FIG. In the figure, the curve shown by the solid line is the measured value. From this graph, it can be seen that since the area of the black dot is large, the optical density is high in all the input image data and the output density is saturated in the region where the input image data is large. FIG. 12 is a block diagram of an image processing unit in a general binary printer. When multi-gradation image data is input to the image processing unit, the γ correction processing unit 11 performs γ correction processing, and then the average error minimum processing unit 12 performs average error minimum processing, Image data is output.

Generally, in a printer device, the relationship between the optical densities of input image data and output data is changed by gamma correction processing. This gamma correction processing is generally performed by using an LUT (Look) that is stored in a memory and has a one-to-one correspondence between input and output.
Up Table). For example, FIG.
If the output density characteristic indicated by the solid line in FIG. 1 is subjected to γ correction to make the optical density linear characteristic as indicated by the broken line in the same figure, the input image data “A” is input. In order to obtain the optical density of “C” at this time, the data of “B” may be output to the average error minimizing method processing unit 12 at the subsequent stage, and this relationship may be set in the LUT. If this is performed for all the input image data, the LUT in this gamma correction processing is completed.

[0013]

In this manner, the LUT
Is a problem when the image quality is completed. For example, when the number of bits of input data and the number of bits of output data in the γ correction processing of the γ correction processing unit 11 are the same (generally, this is almost the case), suppose that the number of bits is 8 bits. Input data must be 0
Numbers are linearly arranged up to 255. On the other hand, in the output data, in the region where the measured value graph of FIG. 11 rises steeply (the region of low density in the graph), even if the input data changes, the output data has the same number or the opposite. In the region where the rise of the graph is slowed down (the region of high density in the graph), the change in the output data is larger than the change in the input data, and the output data has discrete values. As a result, while the input data is data of 256 gradations, the output data has a smaller value. this is,
As a result, the number of gradations decreases, and the gradation deteriorates. In particular, in an ink jet printer, this tendency is remarkable because the size of the black dot is large.

In view of the above, an object of the present invention is to provide an image processing apparatus for converting multi-gradation image data into binary image data by a minimum average error method or an error diffusion method without deterioration in gradation. To provide.

[0015]

According to the first aspect of the present invention, the multi-gradation image data of the pixel of interest is corrected by adding an error diffused from neighboring binarized pixels, and the correction is performed. Image data correction means for outputting as image data;
A binarizing unit that compares the corrected image data output from the image data correcting unit with a preset threshold value and converts the image data into binary image data; and a corrected image data output from the image data correcting unit. An error calculating means for obtaining an error in the pixel of interest based on a set value set by a binarization result of the binarizing means;
The above object is achieved by providing a setting means for arbitrarily setting a set value set in accordance with a result of the binarization.

According to a second aspect of the present invention, in the image forming apparatus according to the first aspect, the setting value set by the setting means is a value larger than a maximum value of the multi-tone image data. Achieve the goal. According to a third aspect of the present invention, in the image forming apparatus according to the first aspect, the set value set by the setting unit is changed according to a value of multi-tone image data of the pixel of interest to achieve the object. To achieve.

[0017]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Next, a preferred embodiment of the present invention will be described with reference to FIGS. FIG.
FIG. 3 is a block diagram of a minimum average error processing unit in the image processing apparatus according to the first embodiment of the present invention. As shown in FIG. 1, the average error minimum processing unit includes a multi-tone image data output unit 1, a data correction unit 2, a weighting coefficient generation unit 3, an error buffer 4, a binarization unit 5, a threshold buffer 6, It comprises an error calculation means 7, an output dot conversion value storage buffer 8, and a binary image data input means 9.

The multi-tone image data output means 1 outputs the input multi-tone image data _dxy to the data correction means 2. The data correction means 2 multiplies the error e _{x + i, y + j} diffused from the binarized pixel by the normalized weight matrix weighting coefficient w _ij , by adding the data d _xy, calculating the corrected image data D _xy. The weighting factor generator 3 stores a weight matrix and generates a weighting factor _wij . The error buffer 4 stores the error e _xy calculated by the error calculator 7.

The binarizing means 5 compares the corrected image data _Dxy calculated by the data correcting means 2 with the threshold value T stored in the threshold value buffer 6 and performs a binarizing process. The threshold buffer 6 stores a threshold T to be supplied to the binarizing means 5. The error calculating unit 7 is configured to output the corrected image data D _xy supplied from the data correcting unit 2 and the corrected image data D _xy supplied from the binarizing unit 2.
The error e _xy is calculated based on the digitized data O _xy and the output dot conversion value B supplied from the output dot conversion value storage buffer 8.
Is calculated. The output dot conversion value storage buffer 8 stores the output dot conversion value B to be supplied to the error calculation means 7. The output dot conversion value B can be arbitrarily set by setting means (not shown). Binary image data input means 9
Inputs the output binary image data _Oxy from the binarizing means.

FIG. 2 is a block diagram of the image processing unit according to the first embodiment of the present invention. This image processing unit
As shown in FIG. 2, it comprises only an average error minimizing method processing unit 13, and the gamma correction processing unit 11 shown in the conventional example shown in FIG. 12 is omitted. FIGS. 3A and 3B are configuration examples of the system according to the first embodiment of this invention. FIG. 7A shows an example in which an image processing unit is provided in the host computer 21. In this case, the image processing is performed by the host computer 21 by software, and the binary image data obtained as a result is obtained. The image data is transferred to the binary image output printer 22 to output a binary image. FIG. 2B shows an example in which the image processing unit is provided in the binary image printer 22. The host computer 21 communicates with the binary image output printer 22 using a page description language such as PostScript. Communication, binary image output printer 2
On the two sides, it is developed and subjected to image processing to output a binary image. The configuration shown in FIG. 7A is a method advantageous in cost and is suitable for an inexpensive inkjet printer or the like, and the configuration shown in FIG. 7B is an advantageous method in terms of high speed. It is suitable for a printer of a photographic system or the like. In addition, the system configuration according to the present invention is also possible,
The present invention is not limited to the configuration example shown in FIG.

Next, the operation of the first embodiment having such a configuration will be described with reference to the flowchart of FIG. First, before processing, a pixel counter x in the main scanning direction and a line counter y in the sub-scanning direction are initialized (step 1). In this embodiment, the processing of the image data proceeds in the main scanning (horizontal) direction starting from the upper left portion of the image data, and when the processing of one line is completed, the processing of the next line (the sub-scanning direction). Is performed. The weight matrix shown in FIG. 9 is used.

Next, when the value of the output dot conversion value B is set (step 2), this value B is stored in the output dot conversion value buffer 8. In this embodiment, when the maximum value of the input multivalue image data d _xy and d _max, the value of the output dot converted value B is set to B> d _max. For example, if the input multi-valued image data is 8 bits, the maximum value d _max is “25”.
5 ", so a number larger than" 255 "is set.

Next, the data correcting means 2 multiplies the error e _{x + i, y + j} diffused from the binarized pixel by the weighted coefficient w _{ij of} the normalized weight matrix. Thereafter, the input multi-tone image data d _xy is added to calculate the corrected image data D _xy (step 3). The calculation of the corrected image data _Dxy is performed by the above equation (1). When this equation (1) is expanded using the weight matrix of FIG. 9, the following equation is obtained. D _xy = d _xy + (ex _{-1, y-1} × 1 + ex _{, y-1} × 2 + e
_{x + 1, y-1} × 1 + ex _{-1, y} × 2) × _1/6

Next, the obtained corrected pixel data _Dxy is binarized by a binarizing means 5 using a preset threshold value T. Binarization method, when the correction pixel data D _xy is smaller than the threshold T, the output binary image data O _xy is "0", when the correction pixel data D _xy is equal to or greater than the threshold value T is output 2 The value image data _{Oxy is set} to “1” (step 4
~ Step 6). These can be expressed as follows. IF D _xy <T THEN O _xy = 0 ELSE O _xy = 1

Next, the error calculating means 7 stores the corrected image data D _xy obtained by the data correcting means 2, the binary image data O _xy obtained by the binarizing means, and the output dot converted value storage buffer 8. Based on the output dot conversion value B, an error _exy at the pixel of interest is calculated (step 7). When these are expressed by equations, the following equation (2) is obtained. _{e xy = D xy -O xy ·} B (2) Next, the pixel counter x in the main scanning direction, is incremented (step 8). Thus, the processing for one pixel is completed.

When the processing for one pixel is completed as described above,
It is determined whether the processing for one line in the main scanning direction has been completed (step 9), and if not completed (step 9;
N) Then, the processing of the next (right adjacent) pixel is performed. On the other hand,
If the result of this determination is that processing for one line in the main scanning direction has been completed (step 9; Y), the line counter y is incremented (step 10). Subsequently, it is determined whether or not the processing in the sub-scanning direction has been completed (step 1).
1) If this processing is not completed (step 11;
N), the processing of the next (lower) line is performed, and if the processing has been completed (step 11; Y), the processing of the series of images described above is completed.

FIG. 5 shows the first embodiment.
6 is a graph showing output density characteristics when the value of an output dot conversion value B is changed. The curve g1 in the figure is the same as the curve in FIG. 11, and is the output density characteristic when the output dot conversion value B = _dmax . In the drawing, the value of the output dot conversion value B is g1 <g2 <g3 <g4. Naturally, if the value of the output dot conversion value B is increased, the optical density is reduced in the entire area of the input image data.

In the first embodiment, the average error minimizing method is used. However, if the error diffusion method is used instead, the effect is theoretically equivalent. It is not restricted by the difference between the method and the error diffusion method. In the mean error minimum method and the error diffusion method, a method of superimposing random noise on a threshold value, a method of changing the threshold value using a dither matrix, a method of alternately processing lines, and the like have been devised. However, the present invention corresponds to either case, and the present invention is not limited by a method of changing those thresholds.

As described above, according to the first embodiment, the error calculating means calculates the error e in the target pixel.
_xy is represented by the corrected image data D _xy , 2
The value dot data _Oxy and the output dot conversion value B are obtained, and the output dot conversion value B can be set arbitrarily. Accordingly, since the binarization process adapted to the size of the output dot can be performed, the output density characteristic of the printer device can be changed without using an LUT or the like, and γ correction can be performed without deterioration in gradation. It becomes possible. In addition, the number of γ correction processing units can be reduced, which can also have the effect of simplifying the configuration and speeding up the processing.

Next, a second embodiment of the present invention will be described with reference to FIGS. FIG. 6 is a block diagram of the processing unit for minimizing the average error according to the second embodiment. In the second embodiment, as shown in FIG. 6, input multi-tone image data d _xy output from multi-tone image data output means 1 is converted into output dot conversion value B in which output dot conversion value B is stored. This is supplied to the value storage buffer 8A. Output dot conversion value storage buffer 8A
Is a buffer capable of storing data corresponding to all data of the input multi-tone image data d _xy . In other words, the output dot conversion value B is one pair when the input multi-tone image data d _xy is input. LUT whose value is determined by 1 (Loo)
k Up Table).

For example, when the input multi-tone image data d _xy is composed of 8 bits, this LUT is 256 bytes (B
yte). Further, the LUT of the output dot conversion value B can be arbitrarily set by setting means (not shown). Since the configuration of the other parts of the second embodiment is the same as that of the first embodiment of FIG. 1, the same parts are denoted by the same reference numerals and description thereof is omitted.

Next, a second embodiment configured as described above will be described with reference to the flowchart of FIG. The difference between the operation of the second embodiment and the operation of the first embodiment is that the output dot conversion value storage buffer 8A
The value of the output dot conversion value B stored in the input multi-gradation image data output means 1
_The other operations are the same _except that they change according to _xy (see step 22 in FIG. 7). This is for obtaining a desired output density characteristic, since by each of the input multi-tone image data after d _xy values, can select a value of output dot converted value B, and input multivalue image data The output density value can be selected with a very large degree of freedom, and the gradation characteristics do not deteriorate. In the flowchart of FIG. 7, processing other than step 22, ie, step 2
1, and the respective processes from step 23 to step 31 correspond to the respective processes from step 1 and steps 3 to 11 in FIG. 4, and therefore, the description of these processes will be omitted.

FIG. 8 is a graph showing output density characteristics when the contents of the LUT of the output dot conversion value B are changed in the second embodiment. The curve g1 in the graph is the first
The same as FIG. 1, the value of the output dot converted value B corresponding to all of the input multivalue image data d _xy is the output density characteristic when d _max. The curves g2, g3 and g4 in the graph represent the LU of the output dot conversion value B in order to obtain a desired output density characteristic.
This is the result of creating T.

Note that the second embodiment is a method performed by the average error minimum method as in the first embodiment, but it is theoretically possible to use the error diffusion method instead. The effects are equivalent and the invention is not limited by the difference between the mean error minimization method and the error diffusion method. In the mean error minimum method and the error diffusion method, a method of superimposing random noise on a threshold value, a method of changing a threshold value using a dither matrix, a method of alternately processing lines, and the like have been devised. However, the present invention corresponds to either case, and the present invention is not limited by a method of changing those thresholds. As described above, according to the second embodiment, the value of the output dot converted value B is changed by the input multi-gradation image data d _xy, optimum output for the input multi-gradation image data d _xy Since the image density can be selected, a desired output density characteristic can be obtained without deterioration in gradation.

[0035]

According to the first and second aspects of the present invention, based on the corrected image data and the set value set by the binarization result, an error in the pixel of interest is determined, and the set value is arbitrarily set. Configurable. Therefore, since the binarization process adapted to the size of the output dot can be performed, the output density characteristic of the printer device can be changed without using an LUT or the like, and γ correction can be performed without deterioration in gradation. It becomes possible. Further, since the γ correction is not required, the effects of simplifying the configuration and speeding up the processing can also be obtained. According to the third aspect of the present invention, the set value is changed by the input multi-tone image data, and the optimum output image density can be selected for the input multi-tone image data. It is possible to obtain a desired output density characteristic without accompanying.

[Brief description of the drawings]

FIG. 1 is a block diagram of a processing unit for minimizing an average error according to a first embodiment of the present invention.

FIG. 2 is a block diagram of an image processing unit according to the first embodiment.

FIG. 3 is a diagram illustrating a configuration example of a system according to the first embodiment;

FIG. 4 is a flowchart illustrating an operation example of the first embodiment.

FIG. 5 is a graph showing output density characteristics when the value of an output dot conversion value B is changed in the first embodiment.

FIG. 6 is a block diagram of an average error minimum processing unit according to a second embodiment.

FIG. 7 is a flowchart illustrating an operation example of the second embodiment.

FIG. 8 is a graph showing output density characteristics when the contents of an output dot conversion value B LUT are changed in the second embodiment.

FIG. 9 is a diagram illustrating an example of a weight matrix.

FIG. 10 is a schematic diagram when dots are printed by a printer device.

FIG. 11 is a graph showing output density characteristics.

FIG. 12 is a block diagram of an image processing unit in a general binary printer device.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Multi-tone image data output means 2 Data correction means 3 Weighting coefficient generation means 4 Error buffer 5 Binarization means 6 Threshold buffer 7 Error calculation means 8 Output dot conversion value storage buffer 9 Binary image data input means

Claims (3)

[Claims] 1. An image data correction means for correcting by adding errors diffused from neighboring pixels which have already been binarized to multi-tone image data of a target pixel, and outputting the corrected data as corrected image data. A binarizing unit that compares the corrected image data output from the image data correcting unit with a preset threshold value and converts the image data into binary image data; and a corrected image data output from the image data correcting unit. Error calculating means for obtaining an error in the pixel of interest based on the set value set by the binarization result of the binarization means; and arbitrarily changing the set value set by the binarization result of the binarization means. An image processing apparatus comprising: setting means for setting. 2. The setting value set by the setting means is:
2. The image processing apparatus according to claim 1, wherein the value is larger than a maximum value of the multi-tone image data. 3. The image processing apparatus according to claim 1, wherein a setting value set by said setting means is changed in accordance with a value of multi-tone image data of said pixel of interest.