JP4069768B2 - Image processing method and apparatus - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an image processing method and apparatus for performing pseudo-halftone expression by applying an error diffusion method.
[0002]
[Prior art]
Conventionally, in a laser printer or a copying machine, a low-value processing method such as a dither method or an error diffusion method is used to faithfully reproduce a halftone original image such as a photographic image.
[0003]
The error diffusion method is one of pseudo halftone representation methods that can reproduce a halftone original image such as a photographic image with a reduced value. In the error diffusion method, the gradation level of the original image is lowered to the reduced data by a certain threshold value, and the gradation value (density value or data value) of the pixel of interest and the density value of the corresponding reduced data are Are weighted and distributed to a plurality of surrounding pixels within a certain range. According to the error diffusion method, since the density of the original image is maintained, a relatively faithful image can be obtained. Compared with the dither method, both resolution and gradation are excellent.
[0004]
However, the error diffusion method has a drawback that the granularity of the image is poor. In particular, the poor granularity in the highlight portion (low density portion) is conspicuous, which reduces the image quality of the entire image. This is due to the occurrence of a texture in which dots are connected in a specific direction in the highlight portion. Such a texture is generally called worm noise.
[0005]
In order to compensate for such drawbacks, random noise is superimposed on input data or a sine wave is added.
[0006]
[Patent Document 1]
JP2000-287086
[0007]
[Patent Document 2]
JP 2001-257880 A
[0008]
[Problems to be solved by the invention]
According to the method described in Patent Document 1 described above, random noise is generated according to the gradation level of input pixel data, and the random noise is converted into pixel data before the pixel data is subjected to multilevel error diffusion processing. Superimpose on.
[0009]
According to the method described in Patent Document 2, an amplitude and frequency-modulated sine wave is added to a target pixel in accordance with an average value of the target pixel and peripheral pixels.
According to these conventional methods, a certain degree of improvement effect can be obtained, but high frequency noise such as a screen is applied to an image in which dots are sparsely distributed such as a highlight portion (low frequency image). By simply superimposing, a modulation effect cannot be obtained and worm noise cannot be eliminated.
[0010]
The present invention has been made in view of the above-described problems, and an object of the present invention is to prevent the occurrence of worm noise in a highlight portion or the like and improve graininess in image processing to which an error diffusion method is applied.
[0011]
[Means for Solving the Problems]
The method according to the present invention is an image processing method for outputting binary output data indicating dot on / off by applying an error diffusion method to multi-valued input data, the processing target being included in the input data Storing a binarization result for a target area including pixels in a predetermined range around the target pixel, a dot count value obtained from the binarization result in the target area, and a tone value of the target pixel Corresponding to the reference value related to the number of dots that should exist in the target region, and binarization of the pixel of interest according to the result of the comparison between the reference value and the dot count value. Forcing the result to be reversed.
[0012]
The apparatus according to the present invention comprises: binarization storage means for storing a binarization result for a target area composed of pixels in a predetermined range around a target pixel as a processing target in input data; and the binarization A dot count unit that obtains a dot count value related to the number of dots turned on or off based on the binarization result read from the storage unit, and should exist in the target region corresponding to the gradation value of the pixel of interest A reference value acquisition means for obtaining a reference value related to the number of dots, and a compulsory for reversing the binarization result for the pixel of interest according to the size of the dot count value and the reference value Reversing means.
[0013]
Preferably, the forced inversion means forcibly turns off the binarization result of the target pixel when the target pixel is in a highlight portion and the dot count value is larger than the reference value, Is in the shadow portion and the dot count value is smaller than the reference value, the binarization result of the target pixel is forcibly turned on.
[0014]
A weight storage unit configured to store a weight corresponding to a distance from the target pixel in the target region, wherein the dot count unit performs a product-sum operation on the binarization result and the weight to perform the dot calculation; Obtain the count value.
[0015]
Further, the binarization storage means includes a highlight storage unit for storing a binarization result when the target pixel is in the highlight portion and a binarization result when the target pixel is in the shadow portion. The dot count unit stores the dot count based on the binarization result read from the highlight storage unit when the pixel of interest is in the highlight unit. When the pixel of interest is in the shadow portion, the dot count value is obtained based on the binarization result read from the shadow storage portion.
[0016]
DETAILED DESCRIPTION OF THE INVENTION
[First Embodiment]
FIG. 1 is a block diagram showing an image processing apparatus 1 according to the first embodiment of the present invention, FIG. 2 is a diagram showing a target region TR that is a region around a pixel of interest, and FIG. 3 is an example of a reference value KT. FIG.
[0017]
The image processing apparatus 1 shown in FIG. 1 performs error diffusion processing on input data (input image data) D1 having 256 gradations (8 bits), and binary output data having lower gradations ( Output image data) D4 is output.
[0018]
Such an image processing apparatus 1 is used as a partial function of an image processing apparatus using, for example, a copying machine, a printer device, an image reading device, or a personal computer. That is, for example, error diffusion processing is performed on color input data D1 output via a CCD image sensor, an A / D conversion unit, a shading correction unit, and the like, and output data D4 is output. The output data D4 is then output to the print engine via, for example, a γ correction unit, a print position control unit, a D / A conversion unit, and the like. Or it is output to a display device.
[0019]
As shown in FIG. 1, the image processing apparatus 1 includes an addition unit 31, a comparison unit 32, a forced inversion unit 33, a subtraction unit 34, and an error integration unit 35.
The adder 31 adds error data to the input data D1. The comparison unit 32 compares the added input data D3 with the threshold value DS, and outputs an ON signal as comparison data DR when the input data D3 is large and an OFF signal when the input data D3 is small. The comparison data DR can also be represented by a 1-bit state value of “H” and “L” or “1” and “0”.
[0020]
The forcible inversion unit 33 forcibly inverts the comparison data DR when the binarization result in the target region TR becomes a predetermined state. Data output from the forced inversion unit 33 is output data D4. Details will be described later.
[0021]
The subtracting unit 34 calculates an error between the binarized output data D4 and the binarized input data D3. At this time, in order to match the gradation of the output data D4 and the input data D3, the output data D4 is multiplied by an appropriate multiple (256 times).
[0022]
The error integrating unit 35 accumulates the error and outputs the accumulated error data to the adding unit 31 so as to be distributed to the peripheral pixels of the target pixel.
Note that the error diffusion processing by the adding unit 31, the comparing unit 32, the subtracting unit 34, and the error integrating unit 35 is known per se, and various conventional methods or circuits can be applied. The image processing apparatus 1 according to the present embodiment is characterized by having a forced inversion unit 33. Next, the configuration and operation of the forced inversion unit 33 will be described.
[0023]
The forced inversion unit 33 includes a binarization storage unit 331, a dot count unit 332, a determination unit 333, and a forcible processing unit 334.
The binarization storage unit 331 stores the binarization result for a target area including pixels within a predetermined range around the target pixel as a processing target in the input data. In the present embodiment, the output data D4 from the forced inversion unit 33 is stored as a binarization result.
[0024]
As shown in FIG. 2, the target region TR is a region composed of 16 × 33 pixels that expand to the left and right above the target pixel TG. However, 16 pixels to the right of the target pixel TG are excluded. Since the target pixel TG itself is also excluded, after all,
16 × 33− (16 + 1) = 511
As a total area of 511 pixels. Therefore, the binarization storage unit 331 is configured by, for example, a line memory for 16 lines, and stores a binarization result for 16 lines that has been processed before the target pixel TG.
[0025]
The dot count unit 332 obtains a dot count value DT related to the number of dots on or off based on the binarization result read from the binarization storage unit 331.
[0026]
That is, when the dot count unit 332 obtains the dot count value DT, the dot count window DW corresponding to the target region TR is used. The dot count unit 332 covers the line memory of the binarization storage unit 331 so that the position of the pixel of interest TG coincides with the dot count window DW, and the range of the dot count window DW, that is, on or off in the target region TR. Count the number. If the number of one of on and off is counted, the other number is easily calculated. In the first embodiment, the number of ON or OFF in the target area TR is the dot count value DT.
[0027]
The determination unit 333 obtains a reference value KT that is a value related to the number of dots that should exist in the target region TR, corresponding to the gradation value of the target pixel TG (that is, the data value or density value of the input data D1). . Then, the obtained reference value KT and the dot count value DT obtained by the dot count unit 332 are compared, and a compulsory signal KS is output according to the magnitude. Next, how to obtain the reference value KT will be described.
[0028]
In the forced inversion unit 33 provided in the image processing apparatus 1 of the present embodiment, the actual dot density in the target region TR becomes substantially equal to the dot density theoretically derived corresponding to the gradation value of the target pixel TG. Thus, the comparison data DR output from the comparison unit 32 is forcibly inverted. The actual dot density in the target region TR is obtained as the dot count value DT described above. The reference value KT may be a dot density that theoretically corresponds to the gradation value of the target pixel TG.
[0029]
Therefore, for example, when the data value of 256-gradation input data D1 is “1”, the theoretical number of dots (on dots) in 512 pixels is “2”, so If the number of dots in the target region TR is “1”, the dots may be turned on in the next target pixel TG. Therefore, the reference value KT for the data value “1” is set to “1”. In addition, when the data value of the input data D1 is “2”, the theoretical number is “4”, so if the number of dots in the target region TR is “3”, A dot may be turned on in the target pixel TG. Therefore, the reference value KT for the data value “2” is set to “3”. In this way, the reference value KT in the highlight portion is determined. When the data value is “0”, the reference value KT is “0”.
[0030]
Regarding the shadow portion, when the data value of the input data D1 is the maximum “255”, the dot must be turned on also in the next target pixel TG. When the data value is “254”, the theoretical number of dots in 512 pixels is “510”, so if the number of dots in the target region TR of 511 pixels is “509” or less. The dot has to be turned on at the next pixel of interest TG. Therefore, the reference value KT for the data value “254” is set to “509”. Similarly, the reference value KT for the data value “253” is set to “507”. In this way, the reference value KT in the shadow part is determined. FIG. 3 shows the reference value KT determined in this way.
[0031]
The highlight portion is a portion where the data value of the input data D1 is “0” to “127”, and the shadow portion is a portion where the data value is “128” to “255”. On-dot is a data state corresponding to an operation of actually placing dots in a printer or the like.
[0032]
The determination unit 333 holds the reference value KT described above as a reference value table TB1. Further, instead of the reference value table TB1, it may be held as a program for obtaining by calculation.
[0033]
As can be understood from the above description, the determination conditions in the determination unit 333 are as follows.
When the target pixel TG is in the highlight portion and the dot count value DT is larger than the reference value KT, a compulsory signal KS that forcibly turns off the state of the comparison data DR for the target pixel TG is output. To do.
[0034]
Further, when the target pixel TG is in the shadow portion and the dot count value DT is equal to or less than the reference value KT, a compulsory signal KS for forcibly turning on the state of the comparison data DR for the target pixel TG is generated. Output.
[0035]
Note that the value of the reference value KT described above is an example, and a different value may be used. Further, in determining whether the dot count value DT and the reference value KT are large or small, whether or not they are equal may be appropriately processed according to how the reference value KT is determined.
[0036]
The forcible processing unit 334 forcibly turns on or off the comparison data DR based on the forcible signal KS output from the determination unit 333. That is, the comparison data DR is corrected by the forced signal KS. When there is no forcing signal KS, the comparison data DR is output as it is.
[0037]
Note that the granularity actually becomes a problem in error diffusion processing because the data value is “16” or less in the highlight portion and the data value is “240” or more in the shadow portion. The signal KS may be output.
[0038]
Data output from the forcible processing unit 334 is input to the subtraction unit 34 and output as output data D4. Even if the comparison data DR of the comparison unit 32 is forcibly turned on or off by the forcible processing unit 334, an error is calculated using the output data D4 after such processing is performed. The gradation that D1 has is not lost.
[0039]
Next, an outline of processing by the image processing apparatus 1 will be described with reference to a flowchart shown in FIG.
In FIG. 4, first, data of the target pixel TG is input (# 11). Binarization is performed (# 12), and the binarization result is stored (# 13). The dot count value DT is acquired (# 14). The reference value KT is acquired (# 15). A discrimination is made by comparing them (# 16). When the forcible signal KS is output (Yes in # 17), the binarization result is forcibly inverted (# 18) and data is output (# 19).
[0040]
As described above, the binary value of the target pixel TG is set such that the actual dot density in the target pixel TG and the target region TR is equal to the theoretically derived dot density corresponding to the gradation value of the target pixel TG. As a result of the conversion is forcibly turned on or off, it is possible to prevent dots (on dots) from appearing continuously in the highlight portion and off dots from appearing continuously in the shadow portion. As a result, the occurrence of worm noise is prevented and the graininess is improved.
[Second Embodiment]
5 is a block diagram showing an image processing apparatus 1B according to the second embodiment of the present invention, FIG. 6 is a diagram showing the value of the weight WT, FIG. 7 is a diagram for explaining how to obtain the weight WT, and FIG. Is a diagram showing an example of the reference value KTB, and FIG. 9 is a diagram for explaining how to obtain the reference value KTB.
[0041]
The image processing apparatus 1B according to the second embodiment is the same as the image processing apparatus 1 according to the first embodiment except for the forced inversion unit 33B. Therefore, only the difference between the forced inversion unit 33B will be described.
[0042]
The forced inversion unit 33B includes a binarization storage unit 331, a dot count unit 332B, a determination unit 333B, a forcible processing unit 334, and a weight storage unit 335B.
That is, the weight storage unit 335B is provided in the forced inversion unit 33B. The weight storage unit 335B stores a weight WT corresponding to the distance from the target pixel TG in the target region TR.
[0043]
That is, as shown in FIG. 6, a weight WT is set for each pixel of the dot count window DW corresponding to the target region TR. The weight WT is set so that the weight WT of the pixel immediately before the pixel of interest TG, that is, the pixel of the pixel immediately before the pixel of interest TG in the horizontal scanning direction is set to “65”, and the layer surrounding the pixel of interest TG thereafter. The weights WT of “29” “17” “11” “8” “6” “5” “4” “3” “2” “1” are set. The weight WT is determined as follows.
[0044]
That is, as shown in FIG. 7, for each gradation value of 256 gradations, an area (number of pixels) necessary for expressing each gradation value with two gradations is obtained. Next, the length (number of pixels) of one side of the square for obtaining the area is obtained and set as the side length. Then, the portion below the decimal point of the length of one side is cut off, and this is used as the distance. The minimum gradation value among the gradation values having the same distance is set as the weight WT in the pixel group at the distance.
[0045]
For example, in FIG. 7, the area of the gradation value “6” is “42.67”, the side length is “6.53”, and the distance is “6”. The area of the gradation value “7” is “36.57”, the side length is “6.05”, and the distance is “6”. Since the gradation values “6” and “7” have the same distance “6”, the smaller gradation value “6” for the pixel group with the distance “6” is set as the weight WT. In FIG. 6, the weight WT of the column including the pixel having a distance of 6 pixels in the horizontal direction from the position of the target pixel TG is “6”. Further, the weight WT of the row including the pixel having a distance of 6 pixels in the vertical direction from the position immediately below the target pixel TG is “6”.
[0046]
The dot count unit 332B obtains a dot count value DT by performing a product-sum operation between the ON or OFF state of the target region TR read from the binarization storage unit 331 and the weight WT read from the weight storage unit 335B. .
[0047]
In the determination unit 333B, as in the case of the determination unit 333 of the first embodiment, a reference value that is a value related to the number of dots that should exist in the target region TR corresponding to the gradation value of the target pixel TG. Get KTB. In the second embodiment, the method for obtaining the reference value KTB is different from that in the first embodiment. Next, how to obtain the reference value KTB will be described.
[0048]
That is, FIG. 9 is an extraction of the left half of the dot count window DW shown in FIG. Here, as an example, consider a case where the data value of the input data D1, that is, the gradation value is “4”. When the gradation value is “4”, as can be seen from the table of FIG. 7, the area is “64” and the side length is “8”. In FIG. 9, when the lower right 8 × 8 pixel region is viewed, if the reference value is “4”, the dot in the case where there is a dot in the 7 × 7 pixel region indicated by hatching Since the generation of (on dot) is prohibited, it can be seen that there is a dot modulation effect. Here, the modulation is to change the arrangement of dots with or without generating dots.
[0049]
However, since the dots should or may exist even outside the 8 × 8 pixel area in the dot count window DW, the number of dots needs to be included. Since the number of dots is “4.3”, the provisional reference value is
4 + 4.3 = 8.3
Accordingly, by rounding down the decimal point to “8”, an appropriate modulation effect can be obtained. In the dot count window DW, since there are monitoring areas on the left and right sides of the target pixel TG, twice the temporary reference value becomes the reference value KTB.
[0050]
By performing processing in the same way, a reference value KTB for each gradation value is obtained as shown in FIG. Although not shown in FIG. 8, the reference value KTB is also obtained for the gradation values “240” to “255” based on the same concept.
[0051]
In short, the reference value KTB is determined in accordance with the weight WT of the dot count window DW so as to obtain an appropriate modulation effect. Due to the nature of the modulation effect, the value of the reference value KTB does not have to be a strictly accurate value.
[0052]
The determination unit 333 compares the obtained reference value KTB with the dot count value DT obtained by the dot count unit 332B, and outputs a forced signal KS according to the magnitude. According to the image processing apparatus 1B of the second embodiment, the occurrence of worm noise or the like is more effectively prevented.
[Third Embodiment]
FIG. 10 is a block diagram showing an image processing apparatus 1C according to the third embodiment of the present invention, and FIG. 11 is a diagram showing an example of a binarization result state for an edge portion of an image.
[0053]
In the image processing apparatus 1C of the third embodiment, when the binarization result is stored for the target region TR, the highlight portion and the shadow portion are stored in different regions. Then, the dot count value DT is obtained separately in the dot count unit 332C for the highlight part and the shadow part. This prevents a unique pattern from occurring at the edge portion of the image.
[0054]
That is, in FIG. 11, the upper half of the target region TR is a shadow portion, and the binarization results are all “1”. The lower half is a highlight portion, and the binarization results are all “0”. Their boundaries are edge EGs. In such a state, whatever the data value of the target pixel TG to be input next is, in many cases, the number of dots corresponding to the data value (gradation value) and the dots in the target region TR It's far from the number. Therefore, the binarization result may be forcibly turned off or forcibly turned on until the edge portion is out of the target region TR. In addition, as soon as the edge portion deviates from the target region TR, dots occur continuously due to errors accumulated so far. Therefore, a unique pattern may occur.
[0055]
For example, even if the gradation value of the target pixel TG is “5”, for example, the binarization result is forcibly turned off for a while. There is a possibility that the on-dots appear continuously from the area where the shadow part is out of the target area TR.
[0056]
As described above, for example, when the target pixel TG is a highlight portion, the dot generated in the shadow portion is influenced by the control of the dot that matches the highlight portion. On the other hand, when the target pixel TG is a shadow part, it is affected by the off-dot generated in the highlight part, although the dot control suitable for the shadow part is being performed.
[0057]
Therefore, the binarization result of the highlight portion is used for discrimination in the highlight portion, and the binarization result of the shadow portion is used for discrimination in the shadow portion.
As shown in FIG. 10, in the image processing apparatus 1C of the third embodiment, instead of the binarization storage unit 331 of the first and second embodiments, a binarization result control unit 331Ca, for highlighting A storage unit 331Cb and a shadow storage unit 331Cc are provided.
[0058]
Both the highlight storage unit 331Cb and the shadow storage unit 331Cc can store the binarization result for the target region TR. The binarization result control unit 331Ca writes the binarization result when the target pixel TG is in the highlight portion to the highlight storage unit 331Cb, and the binarization result when the target pixel TG is in the shadow portion. Is written to the shadow storage unit 331Cc. At that time, data is also written to the other storage unit.
[0059]
Specifically, the following processing is performed.
(1) When the target pixel TG is a highlight portion and the binarization result is on
On is written in the highlight storage unit 331Cb.
[0060]
On is written in the shadow storage unit 331Cc.
(2) When the target pixel TG is a highlight portion and the binarization result is off
OFF is written in the highlight storage unit 331Cb.
[0061]
OFF is written in the shadow storage unit 331Cc (anomaly processing).
(3) When the target pixel TG is a shadow portion and the binarization result is on
Off is written in the highlight storage unit 331Cb (anomaly processing).
[0062]
On is written in the shadow storage unit 331Cc.
(4) When the target pixel TG is a shadow part and the binarization result is off
OFF is written in the highlight storage unit 331Cb.
[0063]
On is written in the shadow storage unit 331Cc.
Then, when the pixel of interest TG is a highlight portion, the dot count unit 332C performs a product-sum operation on the data read from the highlight storage unit 331Cb and the weight WT to obtain the dot count value DT, and the pixel of interest TG Is a shadow part, a product-sum operation is performed on the data read from the shadow storage part 331Cc and the weight WT to obtain the dot count value DT.
[0064]
In this way, dots are connected by using the binarization result for the highlight portion when processing the highlight portion and using the binarization result for the shadow portion when processing the shadow portion. And generation of a peculiar pattern can be prevented.
[0065]
In the above embodiment, the example in which the target region TR has a region of 511 pixels has been described, but it may be a region larger or smaller than this. Although the case where the input data D1 has 256 gradations has been described, the present invention can also be applied to input data D1 having other gradation numbers.
[0066]
In the above embodiment, the data subjected to the forced inversion in the forcible processing unit 334 is stored in the binarization storage unit 331. However, the data before the storage inversion can be stored in the binarization storage unit 331. is there. When the input data D1 is color data, the above process may be performed for each color. In that case, the processing described above may be applied only to a specific color, not to all the colors.
[0067]
In the embodiment described above, the configuration, the circuit, the number, the processing content, and the like of each part or the whole of the image processing apparatuses 1, 1B, and 1C can be appropriately changed in accordance with the gist of the present invention.
[0068]
In the above embodiment, the processing contents of the respective units of the image processing apparatuses 1, 1B, 1C are performed by hardware circuits, by software using a CPU and a memory storing an appropriate program, or by a combination thereof. It is possible to realize. In addition, the configuration, the circuit, the number, the processing content, and the like of each part or the whole of the image processing apparatuses 1, 1B, and 1C can be appropriately changed in accordance with the gist of the present invention.
[0069]
The present invention includes inventions described in the following supplementary notes.
[Appendix 1]
Calculating a quantization error in the error diffusion process based on data output from the forced inversion means;
The image processing apparatus according to claim 2 or 3.
[0070]
【The invention's effect】
According to the present invention, in image processing to which the error diffusion method is applied, it is possible to prevent occurrence of worm noise in a highlight portion or the like and improve graininess.
[0071]
According to the invention of claim 5, it is possible to prevent the generation of a unique pattern.
[Brief description of the drawings]
FIG. 1 is a block diagram illustrating an image processing apparatus according to a first embodiment of the present invention.
FIG. 2 is a diagram illustrating a target region that is a region around a target pixel.
FIG. 3 is a diagram illustrating an example of a reference value.
FIG. 4 is a flowchart illustrating an outline of processing by the image processing apparatus.
FIG. 5 is a block diagram showing an image processing apparatus according to a second embodiment of the present invention.
FIG. 6 is a diagram illustrating weight values.
FIG. 7 is a diagram for explaining how to obtain weights;
FIG. 8 is a diagram illustrating an example of a reference value.
FIG. 9 is a diagram for explaining how to obtain a reference value;
FIG. 10 is a block diagram showing an image processing apparatus according to a third embodiment of the present invention.
FIG. 11 is a diagram illustrating an example of a binarization result state for an edge portion of an image.
[Explanation of symbols]
1,1B, 1C Image processing device
331 Binarization storage unit (binarization storage means)
331Cb Highlight storage unit (binarization storage means)
331Cc Shadow storage unit (binarization storage means)
332, 332B, 332C dot count unit (dot count means)
333, 333B, 333C Discriminator (reference value acquisition means)
334 Forced processing part (forced inversion means)
335B, 335C Weight storage unit (weight storage means)
TR target area
KT standard value
TB1 reference value table (reference value acquisition means)

Claims (7)

An image processing method for applying binary error data to multi-valued input data to output binary output data indicating dot on / off,
Storing a binarization result for a target area composed of pixels in a predetermined range around a target pixel to be processed in the input data;
Comparing a dot count value obtained from the binarization result in the target area with a reference value related to the number of dots that should exist in the target area corresponding to the gradation value of the target pixel;
Forcibly inverting the binarization result for the pixel of interest according to the result of the comparison between the reference value and the dot count value;
An image processing method comprising: An image processing apparatus that performs error diffusion processing on multi-valued input data and outputs binary output data indicating dot on / off,
Binarization storage means for storing a binarization result for a target area composed of pixels in a predetermined range around a target pixel to be processed in the input data;
Dot count means for obtaining a dot count value related to the number of dots on or off based on the binarization result read from the binarization storage means;
Reference value acquisition means for obtaining a reference value related to the number of dots that should exist in the target region in correspondence with the gradation value of the target pixel;
Forced inversion means for comparing the dot count value with the reference value and forcibly inverting the binarization result for the pixel of interest according to the magnitude thereof;
An image processing apparatus comprising: The forced inversion means is
When the target pixel is in the highlight portion and the dot count value is larger than the reference value, the binarization result of the target pixel is forcibly turned off,
When the target pixel is in a shadow portion and the dot count value is smaller than the reference value, the binarization result of the target pixel is forcibly turned on.
The image processing apparatus according to claim 2. Weight storage means for storing a weight according to the distance from the target pixel in the target region;
The dot count means obtains the dot count value by performing a product-sum operation on the binarization result and the weight;
The image processing apparatus according to claim 2 or 3. The binarization storage means includes
A highlight storage unit for storing the binarization result when the target pixel is in the highlight unit, and a shadow storage unit for storing the binarization result when the target pixel is in the shadow unit; Have
The dot count means obtains the dot count value based on the binarization result read from the highlight storage unit when the target pixel is in the highlight portion, and when the target pixel is in the shadow portion. Calculates the dot count value based on the binarization result read from the shadow storage unit,
The image processing apparatus according to claim 2. The reference value is set based on multi-value input data of a target pixel.
The image processing method according to claim 1. The reference value acquisition means obtains the reference value based on multi-value input data of a target pixel; The image processing apparatus according to claim 2.

Images