JP3124589B2 - Image processing device - Google Patents

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 quantizing multivalued image data into binary data.

[0002]

2. Description of the Related Art As a pseudo intermediate processing system used in a digital copying machine, a facsimile machine, etc., for example, Japanese Patent Laid-Open No.
-210959 "has been proposed. In this image processing apparatus, the target pixel is binarized to black or white using binary data of a pixel which has already been binarized in the vicinity of the target pixel, and an error generated at the time of binarization is determined by the target pixel. The operation of adding to the multi-value data of neighboring non-binarized pixels is sequentially performed for each pixel.

That is, the average density is calculated using only the binary data after the binarization processing, and the input multi-value data is binarized using the average density as a threshold value. Therefore, binarization can be performed with a relatively small amount of processing, and an error between the input multi-valued data and the average density generated when the input multi-valued data is binarized is corrected. This is a binarization method that has the advantage of being obtained.

[0004]

However, in the above-described binarization method, the pixel of interest is binarized according to the conditions of the neighboring pixels, so that it is inevitable that dot connection or the like occurs in the highlight portion. There is a problem that the image quality of the highlight portion is reduced. In addition, there is a problem that image quality is deteriorated in a low density portion.

[0005]

SUMMARY OF THE INVENTION The present invention relates to the above-mentioned prior art.
Technique, and remove the lower bit data of the pixel of interest.
Data is binarized based on random numbers and converted to upper bit data.
In addition to binarizing the upper bit data by addition,
The average value used when binarizing the set data
By selecting from the values, the highlight of the image
The way dots are struck at the point
Image processing device that can obtain high-quality binary images
The purpose is to provide. For example, the following configuration is provided as one means for achieving such an object . That is, the multi-valued image data is 2
An image processing apparatus for processing binary to a value data of interest
Input means for inputting multi-valued image data corresponding to pixels,
Binary data of the pixels that have already been binarized to the pixel of interest before
An average value calculating means for obtaining a weighted average of a plurality of pixels in the pixel of interest near by referring to the multi-value image
Divide image data into upper bit data and lower bit data
Dividing means, and random number generating means for generating a random value ,
The division based on a random number value generated by the random number generating means;
And lower binarizing means for binarizing the lower bit data divided by means binarized with the lower binarizing means 2
Adding means for adding value data to the upper bit data divided by the dividing means, the lower bits by the adder means
Upper- level binarizing means for binarizing the higher-order bit data to which the binarized data of the data has been added based on the average value obtained by the average value calculating means; ,reference
Calculate multiple weighted averages with different pixels
The high-order binarization means calculates one average value from a plurality of average values.
Value, and use the selected average value as a threshold to
Data is binarized .

Further, for example, the higher-order binarizing means distributes a density error generated when the higher-order bit data of the target pixel is binarized to non-binarized pixels around the target pixel, thereby preserving the density. It is characterized by comprising a correcting means for performing the correction. In addition, for example, the upper binarizing means, before binarization
From the plurality of average values on the basis of the multivalued data of the pixel of interest
It is characterized in that one average value is selected. Or,
The average value calculating means calculates a plurality of weighted average values based on a plurality of weight masks having different sizes. Alternatively, the lower binarizing means is characterized by binarizing the generation random number of the random number generation means as a threshold value.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment according to the present invention will be described below in detail with reference to the drawings.

[0008]

FIG. 1 is a block diagram of an image processing apparatus according to an embodiment of the present invention. In FIG. 1, an input sensor unit 101 includes a photoelectric conversion element such as a CCD and a driving device for operating the photoelectric conversion element, and performs an operation of reading a document.
The document image data read by the input sensor unit 101 is multi-value analog image data. The multi-level analog image data is sequentially sent to the A / D converter 102, and the analog data of each pixel is converted into the corresponding multi-level digital data. Next, in the correction circuit 103, correction processing such as shooting correction is performed in order to correct the sensitivity unevenness of the CCD sensor and the illuminance unevenness due to the illumination light source.

The multivalued image data input to the subsequent binarization circuit 104 is quantized into binary data by a binarization method described later. The printer 105 is a printer configured by a laser beam or an inkjet method,
On / off control of the dots is performed based on the binary data sent from the binarization circuit 104, and the image is reproduced on the recording paper.

FIGS. 2 and 3 show the binarizing circuit 1 in FIG.
04 shows the detailed block configuration. 2 and 3, reference numerals 1 and 2 denote binary image data subjected to the binarization processing, respectively.
Line memories for storing lines, 3 to 12, 21 are D-type flip-flops DF / F for delaying image data by one pixel, and 13, 14 are weighted averages of predetermined areas from binary image data around the target pixel. This is an arithmetic unit for obtaining a value.

Reference numeral 15 denotes a comparator for comparing the quantized data from the adder 22 with a predetermined value β. Reference numeral 16 denotes a plurality of weighted average values of a plurality of weighted average values from the computing units 13 and 14 based on a signal from the comparator 15. A selector for selecting one as a threshold,
A subtractor 17 calculates a difference between the threshold value output from the selector 16 and the quantized data of the pixel of interest.
A comparator 19 compares the threshold value output from 6 with the quantized data of the target pixel from the adder 22 to binarize the target pixel. Reference numeral 19 denotes the quantized data of the target pixel output from the subtractor 17 and An error ROM for calculating error data based on the difference with the threshold value, a line memory 20 for storing one line of error data from the error ROM 19, and a quantum memory 22 for adding density to the quantized data. An adder 23 adds error data from the line memory 20 and the DF / F 21 to the digitized data, and an adder 23 adds the binarized lower bit data to the upper bit data of the pixel and outputs quantized data. is there.

Numeral 24 denotes an arithmetic unit for dividing the multi-valued pixel data input from the correction circuit 103 into upper bit data and lower bit data, and 25 denotes a threshold value output from the random number generator 26 and a threshold value from the arithmetic unit 24. A comparator for comparing the lower bit data with the lower bit data and binarizing the lower bit data;
Is a random number generator that generates uniform random numbers. The principle of the binarization method in the present embodiment having the above configuration will be described below.

FIG. 4A shows the multi-value density of each pixel of the input image. In (1) of FIG. 4, f (i, j) indicates multi-valued image data of an input image at a target pixel position to be binarized, and takes a value of "0 to 255" in an 8-bit configuration. Further, the image data at the pixel position above the broken line has already been subjected to the binarization processing, and when the binarization of the pixel of interest is performed,
Thereafter, the same binarization processing is sequentially performed on f (i + 1, j), f (i + 2, j),.

FIG. 4B is a diagram showing the binarized image data. In (2) of FIG. 4, B (i, j) is the pixel of interest 2
The data (0 or 1) after quantification is shown. The portion surrounded by the broken line is pixel data that has already been subjected to the binarization processing at the time of processing the target pixel, and these are used in the binarization processing of the target pixel. FIG. 4C is a diagram illustrating weighting. R ₁ R ₂ is an example of a weighting mask for obtaining an average density and is represented by a matrix of 5 × 3 and 3 × 2 sizes.

Here, the weights for the non-binarized pixels are as follows: R ₁ (0,0) = R ₁ (1,0) = R ₂ (2,0) = 0, R ₂ (0,0) = R ₂ Used as (1,0) = 0. Then, first, 8-bit input image data f (i,
j) is divided into upper 6-bit data f _H (i, j) and lower 2-bit data f _L (i, j). The lower data is binarized to 0 or 1 using a uniform random number in which values from 0 to 3 occur at the same probability as a threshold. Binarized lower data B _L (i, j) is upper data
It is added to f _H (i, j) to obtain quantized data h (i, j).

Next, the weighted average densities m ₁ (i, j) and m ₂ (i, j) in the vicinity of the pixel of interest are obtained from the following equation from the pixel data that has already been binarized. _One of m ₁ (i, j) and m ₂ (i, j) is determined by the quantized data h (i, j) of the target pixel and the binarization correction value E (i, j) assigned to the pixel. select. Here, β is a predetermined value for judging whether or not it is a low density portion, and is set to about 10% of the maximum density value. In this embodiment, since the upper 6 bits are used, the maximum density value is 64. Therefore, a value corresponding to 10% is used. That is, β =
It becomes 6.

The quantized data h (i, j) of the pixel of interest is the selected m (i, j) and the binary correction value assigned to the pixel of interest.
It is binarized using E (i, j) according to the following equation. The correction value of the binarization error generated at this time is calculated in the same manner.

E ₁ (i + 1, j) = E ₂ (i, j + 1) = {h (i, j) + E (i, j) -m (i, j)} / 2 (4 E ₁ (i + 1, j) is a correction value assigned to the pixel (i + 1, j) next to the target pixel, and E ₂ (i, j + 1) is a pixel one line after the target pixel. This is a correction value assigned to (i, j + 1). The binarization correction value E (i, j) assigned to the target pixel is obtained by the following equation, and the pixel which is one pixel before the target pixel by the above-described method.
An error E ₁ (i, j) generated when binarizing (i-1, j) and a pixel (i, j-1) one line before the target pixel are generated. This is the sum with the error E ₂ (i, j).

E (i, j) = E ₁ (i, j) + E ₂ (i, j) (5) The above operation is sequentially performed for each pixel to binarize the entire image. In this way, by quantizing the lower bits using the random number as a threshold, the density is appropriately dispersed, and there is an effect that the connection of dots in the low density portion can be effectively prevented. Further, since the density becomes the quantization probability, the density is statistically stored.

In the upper bits, binarization is performed based on peripheral pixel information, and density is preserved by correcting a binarization error, so that an image having excellent gradation and resolution can be obtained. At low densities, widening the average density calculation area has the effect of preventing the connection of dots. Together with the effect of using the random numbers, high-quality image quality in the highlight portion can be obtained. In this case, a line memory is required for a wide average density calculation area for the low density portion, but by using the above method, the memory for storing the correction amount of the binarization error is a higher-order bit. Since only minutes are required, it can be realized without increasing the total memory.
The binarization processing of the pixel of interest (i, j) based on the principle of the above-described binarization method in the configuration of the present embodiment shown in FIGS. 1 to 3 will be described below.

The line memories 1 and 2 shown in FIG. 2 store, in the comparators 18 shown in FIG.
Is stored. When the target pixel is binarized, the line memory 2 outputs the binary data B (i + 2, j-1) one line before, and the line memory 1 outputs the binary data B (i + 2, j-2) is output. Each of the DF / Fs 3 to 12 outputs data delayed by one pixel. That is, D
F / F 3 is B (i + 1, j-2), DF / F 4 is B (i, j-2), and DF / F 5 is
B (i-1, j-2), DF / F 6 is B (i-2, j-2), DF / F 7 is B (i + 1,
j-1), DF / F 8 is B (i, j-1), DF / F 9 is B (i-1, j-1), D
F / F 10 is B (i-2, j-1), DF / F 11 is B (i-1, j), DF / F
12 outputs B (i-2, j), respectively.

Each of the binary data from the DF / Fs 3 to 12 is input to the arithmetic unit 13. Among them, DF / F 7
B (i + 1, j-1), B (i, j-1) from DF / F 8 and DF / F 9
B (i-1, j-1) and B (i-1, j) from the DF / F 11 are also input to the arithmetic unit 14. The arithmetic unit 13 and the arithmetic unit 14 calculate the weighted average density value m ₁ (i, j) indicated by the code A and the weighted average density value indicated by the code B from the binary data in the vicinity of the target pixel based on the respective weight masks. Calculating m ₂ (i, j)
To the selector 16.

On the other hand, the multi-valued data f (i, j) of the pixel of interest sent from the correction circuit 103 is input to the calculator 24 in FIG. Calculator 24 the 8-bit multi-value data f of the pixel of interest is input (i, j), is divided into the upper 6 bits data f _H (i, j) and low-order 2 bits data f _c (i, j) output I do. The random number generator 26 generates a uniform random number having a value of 0 to 3 for each pixel.

The comparator 25 uses the values of random numbers 0 to 3 generated by the random number generator 26 as a threshold value to binarize the low-order 2-bit data f _L (i, j) and lower-order binary data B _L Outputs (i, j). The adder 23 stores the bit data from the arithmetic unit 24.
f _H (i, j) and the lower-order binarized data B _L (i, j) binarized to 1 or 0 from the comparator 25 to obtain quantized data
Outputs h (i, j).

The adder 22 adds error correction data E ₁ from the line memory 20 and DF / F 21 to the quantized data h (i, j).
(i, j) and E ₂ (i, j) are added to output h (i, j) + E (i, j). The comparator 15 compares the error-corrected quantized data h (i, j) + E (i, j) with a predetermined value β, and outputs a select signal to the selector 16 based on the comparison result. Output.

The selector 16 is based on the select signal sent from the comparator 15 and has a weighted average density value m ₁ (i, j) denoted by reference symbol A and a weighted average density value m ₂ denoted by reference symbol B.
(i, j) is selected and output as a threshold value m (i, j). The subtracter 17 calculates the difference between the quantized data h (i, j) + E (i, j) from the adder 22 and the threshold value m (i, j). The comparator 18 binarizes the error-corrected quantized data h (i, j) + E (i, j) using m (i, j) as a threshold value, and outputs a code C
And outputs binary data B (i, j). This binary data B
(i, j) is sent to the printer 105 and also input to the line memory 2 and the DF / F 11 shown in FIG. 2, and becomes peripheral pixel information for pixels to be binarized in the future.

On the other hand, h (i, j) + E output from the subtractor 17
(i, j) -m (i, j) is input to the error ROM 19. Error R
The OM 19 calculates the error data E ₁ (i
+ 1, j), E ₂ (i, j + 1) is output. E ₁ (i + 1, j) is DF / F 21
Is delayed by one pixel and assigned to pixel (i + 1, j), and E ₂ (i, j + 1) is stored in line memory 20 and assigned to pixel (i, j + 1) one line later. .

By repeating the series of processes described above, the binarization process of the image data is sequentially performed for each pixel. By performing the above-described binarization process and stochastically binarizing the lower-order data of the target pixel, the manner in which dots are formed in the highlight portion is appropriately dispersed, and the image quality can be improved.

[0029]

[Second Embodiment] In the first embodiment described above, only the lower two bits of the pixel of interest are binarized using random numbers.
The binarized data and the upper bits are added. However, the present invention is not limited to the above example, and control may be performed so that random numbers are processed according to the level of the upper bit of the target pixel. The second embodiment of the present invention controlled in this way
Examples will be described below.

The second embodiment has substantially the same configuration as that of the above-described first embodiment. The configuration of the first embodiment shown in FIG. 3 is slightly different from that of the first embodiment, and is the configuration shown in FIG. FIG.
2 are the same as those in the first embodiment. In other configurations, the same components as those in the first embodiment are denoted by the same reference numerals, and detailed description is omitted. 5 is different from the first embodiment shown in FIG.
5, the arithmetic unit 426 and the random number generator 427 only.

In the second embodiment, as shown in FIG. 5, the multi-valued data f (i, j) of the 8-bit target pixel sent from the correction circuit 103 is transferred by the arithmetic unit 424 to the upper bit data. And lower bit data. The upper bits and the lower bits may be distributed to the upper 4 bits and the lower 4 bits, or may be distributed to the upper 2 bits and the lower 6 bits contrary to the first embodiment.

Thereafter, in the second embodiment, the higher-order bit data is input to the arithmetic unit 426 together with the random number value generated by the random number generator 427. The output from the arithmetic unit 426 is
This is a threshold value for binarizing the lower bit data in the comparator 425. The arithmetic unit 426 outputs an output value (2 bits of lower bit data) in a low density area where the upper bit data from the random number generator 427 becomes 0 according to the value of the upper bit data.
By increasing the value of a random number that is a threshold for value conversion), the dot printing probability is reduced, and the connection of dots in the highlight portion is suppressed.

By controlling as described above, the way of transmitting the peripheral pixel information can be changed by changing the average density calculation mask in the low density portion, and it is possible to prevent the dot from being imprinted on a close place. Also, it is possible to improve the image quality in the low density portion.

[0034]

Third Embodiment In the first embodiment described above, a plurality of weighted average values are calculated, and one of the plurality of weighted average values is selected as a threshold value based on the quantization level of the pixel of interest. However, the same effect can be achieved even if the threshold value is directly calculated based on the quantization level of the pixel of interest. A third embodiment according to the present invention in which a threshold value is directly calculated based on the quantization level of a target pixel will be described below.

FIG. 6 is a block diagram of a binarizing circuit according to a third embodiment of the present invention, and other parts are the same as those of the first embodiment shown in FIG.
The configuration is the same as that of the embodiment. The same components as those in the first embodiment are denoted by the same reference numerals, and detailed description is omitted. In the third embodiment, as shown in FIG. 6, a select signal from the comparator 14 is directly input to the arithmetic unit 600 instead of the selector 16, and a weighted average value serving as a threshold is calculated. The arithmetic unit 600 has a function of calculating a weighted average value serving as a threshold value including the select signal from the comparator 14 as well as the functions of the arithmetic units 13 and 14 and the selector 16 of the first embodiment. A very high-speed operation can be performed by using the arithmetic device 600 in a table configuration.

With the above configuration, by stochastically binarizing the lower-order data of the target pixel, the manner in which dots are formed in the highlight portion is appropriately dispersed, and the image quality can be improved. According to each of the above-described embodiments described above, the lower bits are stochastically quantized and the calculation area of the weighted average density value is widened in the low density portion. In addition, it is possible to prevent the connection of dots occurring in a low density portion, and to obtain a high quality image.

In the above description, in the first embodiment, only the lower two bits of the pixel of interest are calculated using random numbers.
The binarization processing is performed, and the binarized data and the upper bits are added. In the second embodiment, an example has been described in which control is performed so that random numbers are processed in a low-density region in accordance with the level of the upper bit of the target pixel. However, these are not provided separately and independently. The function of the first embodiment is combined with the function of the second embodiment. The highlight portion is the same as the first embodiment, and the low density region is the same as the second embodiment.
A configuration for performing a value conversion process may be adopted. By performing the optimum binarization processing by switching in this manner, a very high-quality binarized image can be obtained.

The present invention may be applied to a system including a plurality of devices, or may be applied to an apparatus including a single device. Needless to say, the present invention can also be applied to a case where the above is achieved by supplying a program to a system or an apparatus.

[0039]

As described above, according to the present invention, the lower bit data of the target pixel is binarized based on a random number.
Is added to the upper bit data to binarize the upper bit data and used to binarize the upper bit data.
Is selected from multiple average values.
This makes it possible to appropriately disperse the way dots are struck in the highlight portion of the image and obtain a high-quality binary image.
An image processing apparatus capable of performing the above can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an image processing apparatus according to an embodiment of the present invention.

FIG.

FIG. 3 is a detailed circuit diagram of the binarization circuit shown in FIG. 1;

FIG. 4 is a diagram illustrating an example of a multi-valued image, a binarized image, and an average density calculation weighting mask for each pixel according to the present exemplary embodiment.

FIG. 5 is a circuit diagram showing a portion of a binarization circuit according to a second embodiment of the present invention which is different from the configuration of the first embodiment;

FIG. 6 is a detailed circuit diagram of a binarization circuit according to a third embodiment of the present invention.

[Explanation of symbols]

 1, 2, 20 Line memory 3-12, 21 D type flip-flop DF / F 13, 14, 24, 424, 426 Arithmetic unit 15, 18, 25, 425 Comparator 16 Selector 17 Subtractor 19 Error ROM 23 Adder 26,427 Random number generator 101 Input sensor unit 102 A / D converter 103 Correction circuit 104 Binarization circuit 105 Printer 600 An arithmetic unit.

Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 1/40-1/409 H04N 1/46 H04N 1/60

Claims (5)

(57) [Claims] 1. A picture processing apparatus for processing binary the multivalued image data into binary data, input means for inputting multi-valued image data corresponding to the pixel of interest
When binary data of the pixels that have already been binarized to the pixel of interest before
Means for calculating a weighted average value of a plurality of pixels in the vicinity of the pixel of interest by referring to FIG.
Dividing means for dividing the over data, a random number generating means for generating a random number based on said random number value generated by said random number generation means divides
And lower binarizing means for binarizing the lower bit data divided by the unit, the partial binarized binary data is in the lower binarizing means
Adding means for adding the high-order bit data divided by the dividing means, the binary data of the lower bit data by the addition means
Have a summed high-order bit data to an upper binarizing means for binarizing based on the average value calculated by the average value calculating means
And the average value calculating means sets a plurality of pixels by referring to different pixels.
, And the upper binarization means includes a plurality of
Select one average value from the average values of the selected average
An image processing apparatus for binarizing upper bit data using a value as a threshold value . 2. The method according to claim 1, further comprising: distributing a density error generated when the high -order binarizing means binarizes the high-order bit data of the pixel of interest to non-binarized pixels around the pixel of interest. 2. The apparatus according to claim 1, further comprising a correction unit.
The image processing apparatus according to any one of the preceding claims. Wherein the upper binarizing means, the before binarization interest
One out of a plurality of average values on the basis of the multivalued data of the pixel
2. The image processing apparatus according to claim 1, wherein an average value is selected. 4. The image processing apparatus according to claim 1, wherein said average value calculating means calculates a plurality of weighted average values based on a plurality of weight masks having different sizes. 5. The image processing apparatus according to claim 1, wherein said lower-order binarizing means binarizes the random number generated by said random number generating means as a threshold.