JPH1188693A - Pseudo-gradation processing unit - Google Patents

Abstract

PROBLEM TO BE SOLVED: To generate in simulating manner an error with respect to a pixel on a same line as a processing pixel by a random number. SOLUTION: This processing unit is provided with a random number generator 110, an error-weighting arithmetic section 111, an adder 112, a comparator 113, an output section 114, an error memory 115, and a subtractor 116. A random number, denoting a weighted error and a simulating error of a pixel preceding to a noted pixel, is added to input data 117 and the comparator 113 compares the resulting data with a threshold level 118. The comparison result is subjected to N-value processing and the result is outputted from the output section 114. Furthermore, a difference between an output value and data before entering the comparator 113 is used for an error, it is stored in the error memory 115 and used for processing a pixel of a succeeding line.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a digital copying machine,
Among image processing devices that are applied to an image processing unit such as a digital facsimile and a printer and that process and output image data from an image data input device such as a scanner that reads an original image, the M value input data is particularly converted into an N value input data. To (M>
N), by storing the errors that occur,
BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a pseudo gradation processing device that performs pseudo gradation expression while preserving the density of the entire document.

[0002]

2. Description of the Related Art When performing various types of image processing on image data read by a scanner, a pseudo gradation expression is performed to create an output image. The pseudo gradation processing includes a one-dot multi-value processing for converting one dot into an N value according to a certain threshold value, a dither processing for performing area gradation expression using a dither matrix, and a one-dot value as an N value using a threshold value. A typical example is an error diffusion process in which the difference (error) between the input value and the output value generated when the image is converted and stored is distributed to other surrounding pixels, thereby preserving the density energy of the entire document.

As an image processing apparatus for performing an error diffusion process,
For example, a technique described in Japanese Patent Application Laid-Open No. 9-65129 has been proposed. This technique calculates a local differential value for the converted N-value data, corrects the input M-value image data using the local differential value, obtains an edge enhancement effect, and obtains a low density image. It is intended to improve the reproducibility of reproduction.

[0004]

SUMMARY OF THE INVENTION In the error diffusion method,
At least the error that occurred in the previous line pixel of the pixel of interest,
In addition, the error generated in the pixel before the pixel of interest greatly affects the processing pixel. (If error diffusion is performed without reflecting the error generated in the pixel adjacent to the pixel of interest in the processing pixel, the error diffusion peculiar to the constant density portion The image has a regular texture strongly appearing in the image, and the image quality deteriorates.) Therefore, the conventional error coefficient matrix is processed by weighting the previous line pixel and the adjacent pixel close to the target pixel.

However, in this conventional method, the next pixel cannot be processed unless the processing is completed and an error is calculated. Therefore, there is no other way but to perform the operation for each pixel in the main scanning direction. There was a drawback that it was difficult to achieve the conversion.

SUMMARY OF THE INVENTION The present invention has been made in view of the conventional problems, and generates an error for a pixel on the same line as a processing pixel by using a random number.
It is an object of the present invention to provide a pseudo-gradation processing device that can perform error diffusion processing on pixels on one main scanning line at a time and can speed up processing while maintaining image quality.

[0007]

In order to achieve the above object, the present invention is directed to a pseudo gradation processing for converting input M-value image data into N-value image data smaller than the M value. A pseudo-gradation processing device for performing an error diffusion process, wherein an error weighting operation unit for performing a weighting operation on an error up to a processed previous line, and a pseudo-error for a pixel preceding the pixel of interest. A random number generator, an input / output unit that compares an error weighted to an input value and data obtained by adding a random number to a threshold value, converts the data into N-values, and outputs the N-valued data.
And a memory unit for storing an error generated when the value is converted.

According to a second aspect of the present invention, in the first aspect of the present invention, a random number of a random number generator is used for an error of a pixel on a target pixel line.

According to a third aspect of the present invention, in the first aspect of the invention, the random number generated by the random number generator takes a value in an optimum range for the pseudo gray scale number N. .

[0010]

Embodiments of the present invention will be described below with reference to the accompanying drawings. FIG. 1 is a block diagram of an image processing unit of the digital copying machine. The image processing unit of the digital copying machine includes a scanner input unit 100, a filter processing unit (filter circuit) 101, a gamma processing unit 102, a pseudo gradation processing unit 103, and a printer output unit 104.

In such a configuration, the original image data taken in by the scanner input unit 100 is subjected to a process such as an MTF process or a smoothing process according to the type of the document in the filter processing unit 101. Thereafter, gamma conversion is performed by the gamma processing unit 102 so that, for example, if a document is a character document, the rise of low density is good, and if it is a photo document, the gradation of low density is good. The gamma conversion is performed by the conversion table in consideration of not only the type of the document but also the γ characteristics of the printer.

The converted data is further converted into N-valued image data by a pseudo gradation processing unit 103, and a pseudo gradation expression is performed to create an output image. The pseudo gradation processing performed by the pseudo gradation processing unit 103 includes 1
1-dot multi-value processing for converting dots to N-values by a certain threshold, dither processing for area gradation expression using a dither matrix, and input values generated when 1-dots are converted to N-values with a threshold A typical example is an error diffusion process in which the difference (error) between output values is stored and distributed to other surrounding pixels, thereby preserving the density energy of the entire document.

The present invention relates to an error diffusion process of the pseudo gradation process, which is a known technique.
This is briefly described here. FIG. 3 is an explanatory diagram of the error. When the dotted line is an arbitrary threshold and the input value is above the dotted line (a in FIG. 3), the output is a black circle in the figure and the error is -AE. If the input value is below the dotted line (b in FIG. 3), the output is a white circle in the figure. The error is BE. If the input value is the same as the output white circle as in FIG. 3C, no error occurs.

The error generated as described above is distributed to surrounding pixels at an arbitrary ratio. FIG. 4 shows this example. In FIG. 4, when an error generated in a target pixel is DE, in this example, the error is evenly distributed to four surrounding pixels. That is, the generated error is distributed to the four pixels in quarters. This is an example of a distribution, where any surrounding pixels,
The number of pixels and the ratio are arbitrary. In general, many errors are often distributed to pixels closer to the target pixel.

However, in the method of FIG. 4, the error (DE) generated at the target pixel is distributed to unprocessed pixels, so that the hardware configuration becomes complicated. The fact that the error coefficient matrix of FIG. 4 moves one pixel at a time in the main scanning direction to influence the error has the same meaning as receiving an error of a pixel that has already been processed.

Therefore, as shown in FIG. 5, an error generated in each pixel is stored in a memory, and when the pixel of interest is processed, the error is received and processed at an arbitrary ratio. Have been done. According to this method,
Although it is possible to perform simpler processing than distributing the error to unprocessed pixels, since an error generated in a pixel processed immediately before the processed pixel is also calculated, one pixel in the main scanning direction is calculated. Only one line can be processed at a time, and a method of processing one line at a time is impossible.

Therefore, as shown in FIG. 6, a pseudo error is generated by a random number in the error of the previous pixel of the target pixel (as an error), and the sum of the errors up to the previous line (ERR) is obtained.
SUM). Thereby, even when the error diffusion processing is performed on the target pixel, it is not necessary to wait for the processing result of the previous pixel, so that the batch processing of one line in the main scanning can be performed.

FIG. 2 is a block diagram of a pseudo gradation processing apparatus showing an embodiment of the present invention. This block shows the inside of the pseudo gradation processing unit 103 of FIG. 1, and includes a random number generator 110, an error weighting operation unit 111, and an adder 11
2, comparator 113, output unit 114, error memory 115,
A subtractor 116 is provided. Reference numeral 117 denotes input data,
Reference numeral 118 denotes a threshold.

The input data 117 that has been gamma-processed by the gamma processing unit 102 in FIG. 1 includes an error that has been weighted by the error weighting calculation unit 111 with respect to an error up to the processed line, and a processed pixel. The adder 112 adds a random number (a pseudo error of the previous pixel) generated by the random number generator 110 instead of the error of the previous pixel. Then, the sum is compared with a threshold value 118 in a comparator 113, and is output as an N value from an output unit 114.

The difference between the output value from the output unit 114 and the data before entering the comparator 113, that is, the output value of the adder 112 is calculated by the subtractor 116 and stored as an error in the error memory 115. Then, it is used when processing the pixel of the next line. FIG. 11 shows the timing chart.

FIG. 8 is an explanatory diagram relating to random numbers generated by a random number generator. In this example, for the input data,
Compared to the threshold value 128, data larger than 128
F, 128 or less outputs 0. The error D of the surrounding pixels is added to the actual data D of the target pixel, and is compared with the threshold value. When the error coefficient matrix used in this case is the example of FIG. 7, the breakdown of the error is obtained by multiplying the generated errors err1 to err4 of the previous line by 1/8 or 2/8, respectively, and a random number. Are the pseudo-errors R 'and R generated by.

In FIG. 7, err3 and R which are close to the target pixel are weighted by 2/8, and other errors are 1%.
An example is shown in which weighting is performed every / 8. Therefore, the breakdown of the error here is err1, err2, er
The sum of r4 multiplied by 1/8, err3 multiplied by 2/8, R 'corresponding to error weighting 1/8, and R equivalent to error weighting 2/8 is all added.

As shown in FIG. 8, since the binarization is performed using 128 as the threshold value here, possible values of the generated error are 0 to 128. That is, the possible range of the error occurring at the positions of R 'and R is between 0 and 128, and furthermore, it corresponds to 1/8 and 2/8, respectively. This is a range of possible values, and R is a range of possible values from 0 to 32. As described above, the pseudo error can be optimally processed by generating the pseudo error with a random number within the range of the error that can be taken when the N value is obtained.

FIG. 9 shows an example of generation of a pseudo error using a random number. By generating a random number and shifting it to the right by 3 bits, a 1/8 weighted random number can be obtained. FIG. 10 shows an example in which the error is switched between plus and minus depending on the data of the previous pixel.

As can be seen from FIG. 8, since data exceeding 128 outputs 255, a negative error occurs, and data below 128 becomes 0, and a positive error occurs. Therefore, a more optimal error can be generated by switching the sign of the error depending on the size of the data of the previous pixel.

The pseudo gradation processing apparatus according to the embodiment of the present invention includes a random number generator 110, an error weighting operation unit 111, an adder 112, a comparator 113, an output unit 114, an error memory 115, and a subtractor 116. I have. Then, a weighted error and a random number indicating a pseudo error of a pixel preceding the target pixel are added to the input data 117, and the comparator 113 compares the input data 117 with a threshold 118. This comparison result is converted into an N-value,
Output from the output unit 114. Also, the output value and the comparator 1
The difference from the data before entering 13 is stored in the error memory 115 as an error, and is used when processing the pixel of the next line. Therefore, batch processing of one line in the main scanning becomes possible.

[0027]

According to the first aspect of the present invention, it is possible to perform an error diffusion process by generating a pseudo error by using a random number, and to speed up the process.

According to the second aspect of the present invention, the error of the pixel on the target pixel line is not subjected to the arithmetic processing by using the random number of the random number generator (in the error coefficient matrix for weighting the error, Since the error coefficient of the matrix of the pixel line of interest is 0), one main scanning line can be subjected to the error diffusion processing at a time, and the processing speed can be increased.

According to the third aspect of the present invention, the random number generated by the random number generator takes a value in an optimum range for the pseudo gray scale number N. Therefore, in the error diffusion processing by the pseudo error generated by the random number, The conventional image quality can be maintained without deteriorating the image quality and speeding up the processing.

[Brief description of the drawings]

FIG. 1 is a block diagram of an image processing unit of a digital copying machine.

FIG. 2 is a block diagram of a pseudo gradation processing device according to an embodiment of the present invention.

FIG. 3 is an explanatory diagram of occurrence of an error.

FIG. 4 is a diagram illustrating an example of an error coefficient matrix.

FIG. 5 is a diagram illustrating another example of an error coefficient matrix.

FIG. 6 is a diagram illustrating an example of an error coefficient matrix according to the present invention.

FIG. 7 is a diagram showing another example of the error coefficient matrix of the present invention.

FIG. 8 is an explanatory diagram relating to random numbers generated by a random number generator according to the present invention.

FIG. 9 is an explanatory diagram showing an example of pseudo error generation in the present invention.

FIG. 10 is an explanatory diagram showing another example of pseudo error generation according to the present invention.

FIG. 11 is a timing chart of the pseudo gradation processing device of the present invention.

[Explanation of symbols]

 Reference Signs List 110 random number generator 111 error weighting operation unit 112 adder 113 comparator 114 output unit 115 error memory 116 subtractor 117 input data 118 threshold

Claims (3)

[Claims] 1. A pseudo-gradation processing device for converting input M-value image data into N-value image data smaller than the M value, wherein the pseudo-gradation processing device performs an error diffusion process. An error weighting operation unit that performs a weighting operation on the error up to the line, a random number generator for generating a pseudo error in a pixel preceding the target pixel, and data obtained by adding a weighted error and a random number to an input value. A pseudo-gradation processing device, comprising: an input / output unit that outputs an N-value in comparison with a threshold value; and a memory unit that stores an error generated when the N-value is converted. 2. The pseudo gradation processing device according to claim 1, wherein a random number of a random number generator is used for an error of a pixel on a target pixel line. 3. The pseudo-gradation processing device according to claim 1, wherein the random number generated by the random number generator takes a value in a range optimal for the pseudo-gradation number N.