JP2002185785A - Image processing method and device, recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing unit that can generate binary image data with a high resolution to generate an image with high image quality at a high-speed from multi-valued image data with a low resolution. SOLUTION: A quantization threshold generating section 300 generates 4 quantized thresholds according to a dot concentration type dither threshold matrix, and an error spread quantization processing section 100 uses the quantized thresholds to apply 5-value quantization to multi-valued image data with 600 dpi through error processing. A resolution conversion binary section 200 arranges dots whose number corresponds to a quantized value of a target pixel in the arrangement order depending on a corresponding position of the target pixel on the dither threshold matrix into a corresponding 2×2 pixel area so as to generate binary image data with 1200 dpi.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to the field of image processing, and more particularly to the field of devices for forming or displaying images.

[0002]

2. Description of the Related Art A dither method, a density pattern method, and an error diffusion method are generally used as a halftone reproducing method in an image forming apparatus such as a laser printer, a digital copying machine, and a facsimile machine.

In the dither method, the gradation is expressed by a plurality of pixels, and a color is expressed by a combination of the pixels in a color image. The dither method used for general printing is excellent in graininess and smoothly expresses a halftone image. In the so-called area gradation method represented by the dither method, the resolution deteriorates in order to obtain gradation. Further, for a print image such as a halftone dot, moiré is likely to occur in the dither method for generating a periodic image.

As a method of expressing gradation while maintaining resolution, there is an error diffusion method. The error diffusion method is suitable for reproducing character images because it can obtain resolution that is faithful to the original image.However, in halftone images such as photographic parts, isolated dots are arranged in a dispersed or irregularly connected manner. Therefore, the graininess is poor and a peculiar texture may be generated. In particular,
In an electrophotographic printer, since an image is formed with isolated dots, deterioration in graininess and banding due to uneven density are likely to occur.

In some cases, an image forming apparatus or the like wants to convert low-resolution multi-value image data into higher-resolution binary image data. In this case, if the error diffusion method is adopted, a method of increasing the resolution of the multi-valued image data before the error diffusion processing can be considered. However, error diffusion processing
This is a time-consuming process for performing a product-sum operation to diffuse the quantization error of the peripheral pixels, and the processing time increases almost in proportion to the number of pixels per unit area. For example, if the resolution is 60
When the resolution is changed from 0 dpi to 1200 dpi and 2400 dpi, the number of pixels per unit area increases to 4 times and 16 times, in proportion to the square of the resolution, so that the processing time greatly increases.

In order to suppress an increase in the processing time, it is effective to quantize the multi-valued image data by the error diffusion process while keeping the resolution low, and then convert the resolution. An example of such a prior art is found in JP-A-7-295527. In this technique, low-resolution (600 dpi) multi-value image data is multi-value quantized by an error diffusion process,
High resolution (120
0 dpi) is converted to binary image data.

This conventional technique aims at speeding up the processing and preventing an increase in the circuit scale, and at the same time, suppressing moiré and rosette patterns. However, since dots are arranged by a simple density pattern method or a dither method, the resolution is 1200 dpi or more. In a printer, especially in an electrophotographic printer in which dot reproducibility deteriorates as density increases, it is difficult to achieve improvement in image quality. Further, in the dot arrangement by the density pattern method or the dither method, an image has periodicity and may cause moire.

[0008]

SUMMARY OF THE INVENTION Accordingly, a main object of the present invention is to provide a high-resolution binary image for forming an image having excellent image quality such as graininess and resolution from low-resolution multi-value image data. It is an object of the present invention to provide a new method and apparatus for generating data at high speed.

[0009]

A main feature of the image processing apparatus according to the present invention for achieving the above object is that a plurality of quantization units corresponding to each pixel of multi-valued image data are provided. Quantizing threshold value generating means for generating a threshold value in accordance with a dither threshold value matrix, and quantizing data obtained by multi-quantizing the multi-value image data by error diffusion processing using a plurality of quantization threshold values generated by the quantizing threshold value generating means. And first processing means for converting the quantized data into binary image data having a higher resolution than the multi-valued image data. The second processing means comprises: The number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the pixel of interest of the multi-valued image data is determined according to the quantized data value of the pixel of interest, and The arrangement order of the number of the pixel region, is to control according to the corresponding position on the dither threshold matrix of the pixel of interest. Other features of this image processing apparatus are listed below. As described in claim 2, the arrangement order of the dot-on pixels is controlled so as to form a dot concentration type halftone dot. -Arranging threshold values so as to form dot concentration type halftone dots in the dither threshold value matrix as described in claim 3. As described in claim 4, arranging the lowest to fourth thresholds in the dither threshold matrix at different pixel positions. As described in claim 5, the difference between the fourth and fifth thresholds from the smallest in the dither threshold matrix is made larger than the step width of the dither threshold matrix. As described in claim 6, the main scanning is performed by shifting the dither threshold matrix by two or more basic dither threshold matrices in which thresholds are arranged so as to form dot concentration type halftone dots in the sub-scanning direction. Be connected in the direction. As described in claim 7, further comprising image feature extraction means for extracting image features of the multi-valued image data, wherein the quantization threshold value generation means calculates the amplitudes of the plurality of quantization threshold values by the image feature extraction. Control according to the feature quantity output from the means. The control of the amplitude of the quantization threshold is performed by switching a dither threshold matrix. As described in claim 9, the image feature extraction means outputs an edge amount of the multi-valued image data as a feature amount, and the quantization threshold value generation means has a large edge amount output from the image feature extraction means. Reducing the amplitudes of the plurality of quantization thresholds as much as possible. As described in claim 10, the image feature extracting means expands an edge amount of the multi-valued image data before outputting. As described in claim 11, the image feature extracting means averages and outputs the edge amount of the multi-valued image data. According to a twelfth aspect, the quantization threshold value generating means sets all of the plurality of quantization threshold values to the same fixed value when the edge amount output from the image feature extraction means is maximum. As described in claim 13, when the edge amount output from the image feature extracting means is the maximum, the quantization threshold value generating means responds to the value of the multi-valued image data as the plurality of quantization threshold values. To generate values that change. As described in claim 14, an average value of the target pixel and its neighboring pixels is used as the value of the multi-valued image data. As described in claim 15, the quantization threshold value generating means changes the value that changes in accordance with the value of the multi-valued image data with the first processing means as the value of the multi-valued image data increases. To reduce the number of quantizations by As described in claim 16, when the edge amount output from the image feature extracting means is the maximum, the second processing means sets the second value corresponding to the pixel of interest in the multi-valued image data.
Dot-on pixels are arranged in a predetermined arrangement order in a plurality of pixel areas of the value image data.

Further, according to the present invention, the quantized data of the multi-level image data, which has been multi-level quantized by the error diffusion process using a plurality of quantization thresholds generated in accordance with the dither threshold matrix, is set to a higher level. An image processing apparatus for converting the image data into binary image data having a resolution is also provided. A main feature of this image processing device is that, as described in claim 17, the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data is determined by the target pixel. Means for determining the number of pixels determined by the first means in a plurality of pixel areas of the binary image data corresponding to the target pixel of the multi-valued image data. A second means for controlling the arrangement order of the dot-on pixels in accordance with the corresponding position of the pixel of interest on the dither threshold matrix. Another feature is that, as in claim 18, the arrangement order of the dot-on pixels in the plurality of pixel regions is controlled so as to form a dot concentration type halftone dot. The second means is provided with information indicating an edge region of the multi-valued image data, and a dot-on pixel is included in a plurality of pixel regions of the binary image data corresponding to pixels of the edge region of the multi-valued image data. Are arranged in a predetermined arrangement order.

According to another feature of the image processing apparatus of the present invention, in addition to the configuration of any one of claims 1 to 19, an image is formed according to the binary image data. Is to have a means. Another feature of the image processing apparatus according to the present invention is that, in addition to the configuration according to any one of claims 1 to 16, the multi-valued image is obtained by optically scanning a document. It has means for inputting image data and means for forming an image in accordance with the binary image data.

A main feature of the image processing method according to the present invention is that a plurality of quantization thresholds corresponding to each pixel of multi-valued image data are generated according to a dither threshold matrix. Generating, quantizing the multi-valued image data by error diffusion processing using a plurality of quantization thresholds generated in the quantization threshold value generating step to generate quantized data; and Converting the data into binary image data having a higher resolution than the multi-valued image data. In the converting step, a plurality of pixel areas of the binary image data corresponding to a target pixel of the multi-valued image data are included. The number of the dot-on pixels in the pixel is determined according to the quantized data value of the target pixel, and the arrangement order of the dot-on pixels in the plurality of pixel regions is determined. , It is to control according to the corresponding position on the dither threshold matrix of the pixel of interest. Other features of this image processing method are listed below. As described in claim 24, the arrangement order of the dot-on pixels is controlled so as to form a dot concentration type halftone dot. 26. The image processing apparatus according to claim 25, further comprising an image feature extracting step of extracting an image feature of the multi-valued image data, wherein the quantizing threshold generation step calculates the amplitude of the plurality of quantization thresholds by using the image feature extracting. Control according to the feature amount extracted in the step. As described in claim 26, the image feature extraction step extracts an edge amount of the multi-valued image data as a feature amount, and the quantization threshold generation step determines whether the edge amount extracted by the image feature extraction step is Decreasing the amplitudes of the plurality of quantization thresholds as they increase. As described in claim 27, in the image feature extracting step, an image obtained by expanding an edge amount of the multi-valued image data is extracted as a feature amount. As described in claim 28, in the image feature extracting step, an average of edge amounts of the multi-valued image data is extracted as a feature amount. As described in claim 29, in the quantization threshold generation step, when the edge amount extracted in the image feature extraction step is maximum, all of the plurality of quantization thresholds are set to the same fixed value. As described in claim 30, in the quantization threshold generation step, when the edge amount extracted in the image feature extraction step is a maximum, the plurality of quantization thresholds are determined according to a value of the multi-valued image data. To generate values that change. As described in claim 31, an average value of the target pixel and its neighboring pixels is used as the value of the multi-valued image data. As described in claim 32, the quantization threshold generation step includes a step of changing a value that changes according to a value of the multi-valued image data.
Changing the number of quantizations in the quantization step so as to decrease as the value of the multi-valued image data increases. As described in claim 33, when the edge amount extracted in the image feature extraction step is the maximum, the conversion step includes the step of: converting a plurality of pixels of the binary image data corresponding to a target pixel of the multi-valued image data. To arrange dot-on pixels in a predetermined arrangement order in an area.

Further, according to the present invention, the quantized data of the multi-level image data, which has been multi-level quantized by the error diffusion process using a plurality of quantization thresholds generated according to the dither threshold matrix, is set to a higher level. An image processing method for converting into binary image data of a resolution is provided. A main feature of this image processing method is that, as set forth in claim 34, the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data is determined by the target pixel A first step of determining in accordance with the quantized data value of the multi-valued image data, and a number of pixels determined in the first step in a plurality of pixel regions of the binary image data corresponding to a target pixel of the multi-valued image data. A second step of controlling the arrangement order of the dot-on pixels according to the corresponding position of the pixel of interest on the dither threshold matrix. Another feature is that, as in claim 35, the arrangement order of the dot-on pixels in a plurality of pixel regions is controlled so as to form a dot concentration type halftone dot, as in claim 36,
The second step is to arrange dot-on pixels in a predetermined arrangement order in a plurality of pixel regions of the binary image data corresponding to pixels in an edge region of the multi-valued image data.

The features of the present invention described above and the effects obtained thereby will be specifically described below.

[0015]

Embodiments of the present invention will be described below with reference to the accompanying drawings. To avoid repetition of the description, the same or similar reference numerals are used for the same or corresponding parts in a plurality of drawings in the accompanying drawings.

Embodiment 1 FIG. 1 shows a block configuration of an image processing apparatus according to Embodiment 1 of the present invention. This image processing apparatus uses image data of 256 gradations (8 bits / pixel) having a resolution of 600 dpi as input data.
It outputs binary image data (dot on / off) with a resolution of 1200 dpi.

This image processing apparatus, as shown in FIG.
An error diffusion quantization processing unit 100 that performs five-level quantization of input data by an error diffusion process, and outputs the quantized data to 1200d
resolution conversion binarization unit 2 for converting to pi binary image data
00, a quantization threshold value generating unit which generates a periodically changing quantization threshold value (Thr1, Thr2, Thr3, Thr4) corresponding to each pixel of the multi-valued image data and supplies it to the error diffusion quantization processing unit 100 And a position information generating unit 301 that supplies position information to the quantization threshold value generating unit 300 and the resolution conversion binarizing unit 200. In the case where the position information is provided from the source of the multi-valued image data, the position information generation unit 301 can be omitted. Further, the position information generation unit 30
Since 1 can be realized by a simple counter, it is possible to incorporate a unit corresponding to the position information generation unit 301 inside the quantization threshold generation unit 300 and the resolution conversion binarization unit 200.

An error diffusion quantization processing unit 100 is an adder 101 for adding the quantization error of the processed pixel to the multi-valued image data, and outputs the output data of the adder 101 to four quantization thresholds Thr1, Thr2, Thr3. , Thr4 (Thr1 <Thr2 <Thr3 <Thr4)
, A subtractor 103 for calculating a quantization error from the input and output of the quantizer 102, and a predetermined error from the quantization error calculated by the subtractor 103. The adder 101 calculates an error amount to be added to a pixel to be processed next according to the matrix.
, An error diffusion calculation unit 104 to be applied. Quantizer 10
The relationship between the input value and the output value of 2 is as follows: when input value <Thr1, output value = 0 Thr1 ≤ input value <Thr2, output value = 64 Thr2 ≤ input value <Thr3, output value = 128 Thr3 ≤ input When value <Thr4, output value = 192 When Thr4 ≦ input value, output value = 255.

The quantization threshold generation unit 300 is, for example, as shown in FIG.
The quantization thresholds Thr1 to Thr4 are generated according to a 6 × 6 (3 × 3 at 600 dpi) dither threshold matrix as shown in FIG. This dither threshold matrix is formed by spirally arranging thresholds increasing in a step width 3 from 74 to 179 in ascending order of values from 2 to 1 at 1200 dpi.
This is a dot concentration type which forms a halftone dot of the 00 line. In FIG. 2, four thresholds inside the solid grid are 60
It corresponds to one pixel of 0 dpi. As is clear from the observation of FIG. 2, the four thresholds (74, 77, 80, 83) from the smallest one in this dither threshold matrix
Are arranged at different pixel positions of 600 dpi.

The quantization threshold generator 300 outputs four thresholds corresponding to one pixel of 600 dpi in the dither threshold matrix as quantization thresholds corresponding to the pixels. For example, at the upper left pixel position, of the four thresholds (80, 107, 110, 113) at that position, the minimum 80 is the quantization threshold Thr1, the next smallest 107 is the child threshold Thr2, and the next smallest 110 is the quantum threshold Thr2. The quantization threshold Thr3 and the largest 113 are output as the quantization threshold Thr4.

The thresholds in the dither threshold matrix shown in FIG. 2 are rearranged at pixel positions of 600 dpi for each quantization threshold, to obtain a 3 × 3 dither threshold matrix shown in FIG. (A) of FIG. 3 corresponds to the quantization threshold Thr1,
(B) corresponds to the quantization threshold Thr2, and (c) is the quantization threshold.
(D) corresponds to the quantization threshold Thr4.

The position information generation unit 301 generates position information indicating which pixel position (600 dpi pixel position) on the dither threshold matrix shown in FIG. 2 corresponds to the pixel to be processed. It is composed of a counter that counts timing signals synchronized with the pixels of the value image data.

The quantization threshold generation unit 300 described above
Specifically, the dither threshold matrix shown in FIG.
Is stored in a memory such as a ROM, and by addressing this memory in accordance with the position information provided by the position information generating unit 301, a configuration in which four quantization thresholds are read can be obtained.

The resolution conversion binarization unit 200 converts the 600 dpi quantized data from the error diffusion quantization processing unit 100 into 1
It converts the image data into binary image data of 200 dpi. Conceptually, the dot number determination unit 201 and the dot output position determination unit 2
02.

The number-of-dots determination unit 201 determines the number of dot-on pixels (hereinafter, referred to as dots) in 2 × 2 pixels on 1200 dpi binary image data corresponding to each pixel on 600 dpi multi-valued image data. decide. Specifically, 0 when the quantized data value is 0, 1 when the quantized data value is 64, 2 when the quantized data value is 128, and 2 when the quantized data value is 192. Is output as the number of dots, and 3 when the quantized data value is 255.

The dot output position determination unit 202 includes information on the dither threshold matrix shown in FIG.
1, the number of dots determined by the dot number determination unit 201 is set to 1200 dp in accordance with the corresponding position of the pixel to be processed on the dither threshold matrix.
It is determined how to arrange in 2 × 2 pixels of i. More specifically, for example, when a pixel at the upper left pixel position in the dither threshold matrix shown in FIG. 2 is to be processed, a dot is first placed at a position having the smallest value among the 2 × 2 threshold values at the upper left. The second dot is arranged at the position of the next largest threshold, the third dot is arranged at the position of the next largest threshold, and the last dot is arranged at the position of the largest threshold. In this case, if the number of dots is 2, FIG.
As shown in (a), a dot is output at two pixel positions on the right side of 2 × 2 pixels (the two pixels are dot-on pixels). If the number of dots is 1, a dot is output at the lower right pixel position, and if the number of dots is 3, a dot is also output at the upper left pixel position. Similarly, when the right pixel is to be processed, if the number of dots is 2, dots are output at the lower two pixel positions as shown in FIG. 4B. By changing the output order (arrangement order) of dots according to the corresponding position of the pixel to be processed on the dither threshold matrix in this way, it becomes easier to form a halftone dot.

The number-of-dots determining section 201 can be realized, for example, as a lookup table storing the number of dots corresponding to the quantized data value. Similarly, the dot output position determination unit 202 can be a lookup table that outputs 2 × 2 pixel data using the number of dots and position information as input information. Also, the dot number determination unit 201
And the dot output position determination unit 202, for example, can be realized as a look-up table that outputs 2 × 2 pixel data using the quantized data value and the position information as input information.

The dot output position determination unit 202 sets 600
1200 dpi 2 × 2 pixel data for each pixel at dpi is created and buffered in output buffer 203. As the output buffer 203, for example, a line memory having a capacity of two or more lines is used. However, the output buffer 203 can be omitted.

As described above, in this image processing apparatus, a quantization threshold is generated according to a dither threshold matrix in which thresholds are arranged so as to form a dot-concentrated halftone of 200 lines as shown in FIG. And also 600dp
The arrangement order of the dots in the 2 × 2 pixel area of the binary image data corresponding to each pixel of the multi-valued image data of i is controlled according to the corresponding position of the pixel on the dither threshold matrix (as apparent from FIG. 2). The arrangement order is such that a dot concentration type halftone dot is formed). Therefore, it is possible to form a high-quality image with excellent stability and granularity. FIG. 5 shows an example of output data of the image processing apparatus. It can be seen that the formation of 200 halftone dots as in this example shows good stability and granularity. In addition, since the quantization process by the error diffusion method is performed on data of 600 dpi, much faster processing can be performed than when processing is performed after increasing the resolution to 1200 dpi.

As the dither threshold matrix used for generating the quantization threshold, a matrix other than the halftone type of 200 lines as shown in FIG. 2 can be used. At this time, as in the dither threshold matrix shown in FIG. 2, the four thresholds (74, 77, 80, and 83 in FIG. 2) are set to 600 dpi.
May be arranged at different pixel positions. In such a case, when one of the pixels corresponding to the four thresholds is turned on in the low-density part of the image, a negative error is diffused to the surrounding pixels. The pixels that are turned on are not concentrated, and the uniformity of dots in the low density portion is improved.

Embodiment 2 FIG. 6 shows a block configuration of an image processing apparatus according to Embodiment 2 of the present invention. This image processing apparatus has a configuration similar to that of the image processing apparatus shown in FIG.
00 dpi multi-valued image data as input data and 12
It outputs binary image data of 00 dpi.
An image feature extraction unit 350 is added to control the amplitude of the quantization threshold according to the image feature of the input data, and the quantization threshold generation unit 300A also includes a quantum control unit whose amplitude is controlled according to the output of the image feature extraction unit 350. It has been modified to generate the activation threshold. In addition, a delay element 302 is added before the error diffusion quantization processing unit 100 as needed to adjust the timing between the image feature quantity extraction unit 350 and the error diffusion quantization processing unit 100. Otherwise, the configuration is the same as that of the image processing apparatus in FIG.

The image feature extraction unit 350 includes an edge amount calculation unit 351, an edge amount quantization unit 352, and an edge amount expansion processing unit 353. The edge amount calculation unit 351 calculates an edge amount using, for example, a 5 × 5 filter shown in FIG. FIG. 7A is for extracting a vertical edge, FIG. 7B is for extracting a horizontal edge, and FIGS. 7C and 7D are for extracting an oblique edge. The total of the product of the input data of the 5 × 5 pixel area centered on the target pixel and the filter coefficient at the corresponding position is the edge amount of each filter, and the largest value among them is the edge amount of the target pixel as the edge amount. Calculation unit 35
1 is output. The edge amount is quantized by the edge amount quantization unit 352 into four values from the edge amount = 0 (non-edge area) to the edge amount = 3 (edge area) and input to the edge amount expansion processing unit 353. .

The edge amount expansion processing section 353 performs processing for expanding the edge area. More specifically, for example, the edge amount in a 5 × 5 pixel area centered on the target pixel is referred to, and the maximum edge amount among them is output as the edge amount of the target pixel. By performing such expansion processing, a non-edge area sandwiched between edge areas can be converted into an edge area, so that a halftone image or the inside of a thin character can be an edge area over the entire area. It becomes possible.

The quantization threshold value generating section 300A is provided with a matrix (referred to as a dither coefficient matrix) 600 shown in FIG.
A coefficient corresponding to the pixel position of dpi and a multiplier A determined by the edge amount (A = 3 when the edge amount = 0, the edge amount =
(1 = A = 2, A = 1 when the edge amount = 2, A = 0 when the edge amount = 3), and generates a quantization threshold according to a dither threshold matrix having a value obtained by adding 128 to the product. . For example, in the upper left of FIG.
When processing pixels corresponding to (7, -6, -5), the value calculated using the smallest coefficient (-17) is the quantization threshold Thr1, and the value calculated using the next smallest coefficient (-7) is used. The calculated value is the quantization threshold Thr2, the value calculated using the next smaller coefficient (−6) is the quantization threshold Thr3, and the value calculated using the largest coefficient (−5) is the quantization threshold Thr2. Thr4 will be output.

The coefficients in the dither coefficient matrix of FIG.
The matrix shown in FIG. 10 is obtained by rearranging the quantization thresholds Thr1, Thr2, Thr3, and Thr4 collectively at the pixel positions of 600 dpi. FIG. 10A shows the quantization threshold Thr1.
(B) corresponds to the quantization threshold Thr2, (c) corresponds to the quantization threshold Thr3, and (d) corresponds to the quantization threshold Thr4.
Corresponding to

FIG. 11 shows a matrix in which the dither threshold matrices created as described above are grouped for each quantization threshold and arranged at pixel positions of 600 dpi.
FIG. 12 and FIG. FIG. 11 shows an edge amount = 0 (non-edge area), that is, A = 3, and FIG. 12 shows an edge amount = 1.
In the case of (A = 2), FIG. 13 shows the case of the edge amount = 2 (A = 1). Since A = 0 in a region where the amount of edges is 3 (edge region), the same fixed value 128 is output as the four quantization thresholds at all pixel positions.

Here, the dither coefficient matrix at 1200 dpi shown in FIG. 9 is a basic 6 × 6 (60
In this configuration, two matrices (areas surrounded by bold lines) having a size of 0 dpi and 3 × 3 are shifted in the sub-scanning direction by one pixel of 600 dpi in the main scanning direction. Accordingly, the dither threshold matrix created using this dither coefficient matrix is also a matrix in which two basic matrices are shifted in the sub-scanning direction by one pixel of 600 dpi in the main scanning direction. As is clear from the dither coefficient matrix in FIG. 9, the dither threshold matrix created using the matrix is 1
The threshold value arrangement is such that a dot concentration type halftone dot of 200 lines is formed at 200 dpi. Further, the minimum four coefficients (−17, −17, −17, −
17) are arranged at different pixel positions of 600 dpi. That is, the thresholds from the smaller one to the fourth in the dither threshold matrix are arranged at different pixel positions of 600 dpi. Further, the difference between the fourth coefficient (−17) and the fifth coefficient (−14) from the smallest one is 3, which is larger than the difference 1 between the other coefficients. This means that, in the dither threshold matrix, the difference between the fourth and fifth thresholds from the smallest one is larger than the step width (when the thresholds are arranged in order of magnitude, the difference between adjacent thresholds).

As described above, the quantization threshold generator 30
0A is a dither threshold matrix created for each edge amount using the dither coefficient matrix of FIG. 9, or a dither threshold matrix as shown in FIGS. 11, 12, and 13 associated with each quantization threshold. A configuration in which the quantization threshold value is stored in a memory such as a ROM and the quantization threshold is read from the memory in accordance with the position information and the edge amount can be adopted. Alternatively, the dither coefficient matrix shown in FIG. 9 or the dither coefficient matrix shown in FIG. 10 associated with each quantization threshold is stored in a memory, and a coefficient A corresponding to an edge amount is added to a coefficient read from this memory according to position information. It is also possible to adopt a configuration in which a quantization threshold is generated by multiplying and multiplying the product by 128. However, the former configuration is generally more advantageous than the latter configuration because the product-sum operation is not required and the processing can be simplified and speeded up.

The operations of the error diffusion quantization processing section 100 and the resolution conversion binarization section 200 are the same as those of the image processing apparatus shown in FIG. The dot output position determination unit 202 in the resolution conversion binarization unit 200 outputs the dot according to the size of the coefficient on the dither coefficient matrix in FIG. 9 or according to the size of the threshold value on the corresponding dither threshold value matrix. Determine the position. However, each pixel in an area having an edge amount of 3 (edge area) is subjected to the error diffusion quantization processing unit 10.
A fixed value of 128 is used as a quantization threshold value of 0 or 2 to 0 or 255.
Therefore, all the pixels in the corresponding 2 × 2 pixel area in the output data of 1200 dpi are dot-on or dot-off.

With the above configuration, halftone dots of 200 lines are formed in the non-edge area of the input data, and an image having excellent stability and granularity can be obtained. The dither threshold matrix used in the quantization threshold generator 300A is:
As described above, since the phase is shifted in the sub-scanning direction, an effect that banding hardly occurs can be obtained. on the other hand,
In the edge region of the input data, 2
Since the binarization is performed, and the low-frequency screen image and the inside of the fine character are also processed as the edge area over the entire area by the expansion processing of the edge amount as described above, there is no moire in the halftone area. An image can be formed, and an image with good sharpness can be formed in characters and line drawings. In addition, since the amplitude of the quantization threshold is switched stepwise from the edge region to the non-edge region, the switching region between the edge region and the non-edge region while maintaining both graininess and stability in the edge region and sharpness in the edge region. Thus, it is possible to form a high-quality image with less discomfort. Further, as described with reference to FIG. 9, since the four smallest thresholds in the dither threshold matrix are arranged at different pixel positions of 600 dpi, it is possible to prevent excessive concentration of dots in a low density portion of an image. In addition, the granularity of the low concentration portion can be improved. Further, by making the difference between the fourth threshold value and the fifth threshold value from the smallest one larger than the step width, it is easy to dot on the pixel at the center of the halftone dot, and the halftone dot in the middle / high density part of the image is displayed. The collapse of points can be suppressed.

The edge extraction filter used in the edge amount calculation unit 351 may be other than the one shown in FIG. 7, for example, the one shown in FIG. 8 (a) and 8 (b) extract vertical and horizontal edges, and FIG.
It extracts 5 ° halftone dots and oblique lines. Since there is a difference between the number of 1 (or -1) of the filters of (a) and (b) and the number of 1 (or -1) of the filter of (c), (a)
And a difference occurs between the maximum edge amount calculated by the filter of (b) and the maximum edge amount calculated by the filter of (c). Therefore, for example, a value obtained by multiplying the edge amount calculated by the filters (a) and (b) by 1/6,
A value obtained by multiplying the edge amount calculated by the filter (c) by 1/16 may be used as each edge amount.

Embodiment 3 FIG. 14 shows a block configuration of an image processing apparatus according to Embodiment 3 of the present invention. In this image processing apparatus, the configuration of an image feature extraction unit 350A is different from that of the image feature extraction unit 350 of the image processing apparatus shown in FIG. Other configurations are the same as those of the image processing apparatus of FIG.

The image feature extraction unit 350A averages the edge amount calculated by the edge amount calculation unit 351 by the edge amount averaging unit 354. This averaging is performed, for example, by calculating the average value of the edge amounts of the respective pixels in the 5 × 5 pixel region centering on the target pixel. In the edge amount quantization unit 352, the averaged edge amount is converted from the edge amount = 0 (non-edge area) to the edge amount = 3 (edge area).
Are quantized into four values up to the above, and this is
0A.

When a pixel having a large edge amount locally exists in a region having a small edge amount, the edge area is expanded by performing the edge amount expansion processing as in the image feature extraction unit 350 in FIG. May cause sexual deterioration. On the other hand, in this image processing apparatus, the edge amount averaging unit 354 calculates the average value of the edge amounts. Therefore, even if there is a pixel having a large edge amount locally, the influence can be suppressed. . Further, since pixels having a large edge amount are continuously present in the edge portion of the image, the sharpness does not deteriorate even if the edge amounts are averaged.

Fourth Embodiment FIG. 15 shows a block configuration of an image processing apparatus according to a fourth embodiment of the present invention.
In this image processing apparatus, an image feature extraction unit 350 shown in FIG. 6 or an image feature extraction unit 350A shown in FIG. 14 is used for extracting image features.

Further, the quantization threshold value generating section 300 B in this image processing apparatus is provided with an image feature extracting section 350 or 350.
In the region where the edge amount output from A is 0, 1, or 2,
The operation is the same as that of the quantization threshold generator 300A in FIG.
In the region where the edge amount is 3 (edge region), the error diffusion quantization processing unit 1 does not generate a fixed value quantization threshold value but increases the value of the input data in a low density range.
A quantization threshold value is generated that changes so that the number of quantization by 00 becomes smaller. Therefore, the input data of 600 dpi is also provided to the quantization threshold generation unit 300B.

More specifically, in the region where the edge amount is 3, the quantization thresholds Thr1, Thr2, Thr3, and Thr4 output from the quantization threshold generator 300B are, for example, as shown in FIG. It changes according to. As can be seen from FIG. 16, when the input data is 0, Thr1 = 51 and Thr2 =
102, Thr3 = 153 and Thr4 = 204, but Thr1 increases and Thr2 decreases as the input data increases within a range smaller than α. And with input data = α
Thr1 = Thr2. That is, when the input data is from 0 to α, the error diffusion quantization processing unit 100 selects five values (0, 64, 12
8, 192, 255), but the probability that the quantized data becomes 64 decreases as the input data approaches α. And if the input data is in the range from α to β
Since Thr1 = Thr2, four values (0, 128, 192, 25
5), and the quantized data does not become 64. Similarly, in the range from β to γ, the input data is quantized to three values (0, 192, 255), and
In the above, quantization is performed to binary (0, 255). The values α, β, and γ of the input data at which the quantization number changes are, for example, α =
12, β = 24 and γ = 34 can be selected, but are not necessarily limited thereto.

Such a quantization threshold generation unit 300B
A means for generating a quantization threshold varying with an amplitude corresponding to the edge amount (0, 1, 2) in accordance with a dither threshold matrix similarly to the quantization threshold generation unit 300A of FIG. For example, means for generating a quantization threshold value that changes as shown in FIG. The latter means can be realized, for example, as a lookup table that uses the value of input data as search information.

The edge output information from the image feature extraction unit 350 or 350A is also input to the dot output position determination unit 202A of the resolution conversion binarization unit 200A of this image processing apparatus. When the edge amount is 0, 1, or 2, the operation of the dot output position determining unit 202A is the same as that of the dot output position determining unit 202 shown in FIG. However, when the edge amount is 3 (edge area), the dot output position determination unit 202A outputs dots in a predetermined arrangement order, for example, as shown in FIG. That is, in the edge area, if the number of dots is 1 (quantized value = 64), dots are output to the positions shown in FIG.
If (quantized value = 128), dots are output at the position shown in FIG. 17B, and if the number of dots is 3 (quantized value = 192), dots are output at the position shown in FIG. If the number of dots is 4 (quantized value = 255), dots are output at the positions shown in FIG.

Such a dot output position determining unit 202A
Is, for example, the number of dots, position information, and edge amount information (one bit indicating whether the edge amount is 3 is sufficient)
Can be implemented as a look-up table using search information as search information. This lookup table can be integrated with the lookup table of the dot number determination unit 201.

In this image processing apparatus, as described above, in the edge area, a quantization threshold value that changes so as to reduce the quantization number in accordance with the input data is used. ) Or (b), small dots are output, so that images with good sharpness can be formed without noticeable line breaks in low-contrast characters and line drawings. Image can be formed. In the middle and high density portions of the non-edge area, the dots are concentrated because the dither threshold is used. Therefore, when a small dot is output as shown in FIG. 17 (a) or (b) in the middle / high density part of the edge area, a sense of incongruity tends to be conspicuous in a switching portion between the edge area and the non-edge area. However, in this image processing apparatus, the dots are concentrated and output in the high-density part in the edge area as in the high-density part in the non-edge area as shown in (c) or (d) of FIG. Dots are output), and the sense of incongruity is not noticeable at the switching portion between the edge region and the non-edge region. In addition, since large dots are output even in the halftone dots of the high density part in the edge area, stability and
An image with good gradation can be formed.

It has been described that the quantization threshold is changed in accordance with the value of the input data in the edge region, but this only means that the quantization threshold is necessarily changed in accordance with the input data value of the target pixel. Not necessarily, the target pixel and its neighboring pixels (for example, 3 × 3 pixels centered on the target pixel)
The quantization threshold value may be changed according to the average value of the input data, which is also included in the present invention. The method using the average value of the input data as described above has an advantage that the quantization number can be smoothly changed even in a region where the density of the input data changes rapidly.

The image processing apparatus of the present invention as described above, or the processing content thereof, can be realized by software using a general-purpose or special-purpose computer. In the latter case, a program for realizing the functions (processing steps) of each unit of the image processing apparatus on a computer is recorded on various recording media such as a magnetic disk, an optical disk, a magneto-optical disk, and a semiconductor storage element on which the program is recorded. From a computer or from an external computer via a network. Various recording media on which such a program is recorded are also included in the present invention.

The image processing apparatus or method of the present invention is not limited to a single apparatus, but can be applied to image forming apparatuses such as printers, digital copiers and facsimile apparatuses, display apparatuses, image scanners and the like. An example of such an embodiment will be described below.

Fifth Embodiment FIG. 18 is a schematic sectional view showing an example of the structure of a mechanism for reading and forming an image in a digital copying machine to which the present invention is applied. This digital copying machine includes a scanner unit 400 as an image reading unit.
And a laser printer unit 411 as an image forming unit, and a circuit unit 550 (not shown) (FIG. 19).

The scanner section 400 has a flat document table 403.
An original such as a bound original placed on the
And the reflected light image is reflected by mirrors 503, 504,
An image is formed on a reading sensor 507 such as a CCD via a lens 505 and a lens 506, and an illumination lamp 50 is formed.
2 and the mirrors 503, 504, and 505 are moved to sub-scan the original, thereby reading the image information of the original and converting it into an electric image signal. An image signal output from the reading sensor 507 is input to a circuit unit 550 (not shown) and processed. The image data output from the circuit unit 550 is transmitted to a laser printer unit 41 as an image forming unit.
Sent to 1. The reading resolution of the scanner unit 400 is 6
00 dpi.

In the laser printer unit 411, the writing optical unit 508 converts the image data input from the circuit unit 550 into an optical signal, and exposes an image carrier made of a photoconductor, for example, a photoconductor drum 509. As a result, an electrostatic latent image corresponding to the document image is formed. The writing optical unit 508 drives the semiconductor laser by the light emission drive control unit based on the image data to emit the intensity-modulated laser light, and deflects and scans this laser light by the rotary polygon mirror 510 to perform f / θ lens and reflection. Irradiate the photosensitive drum 509 via the mirror 511. In the standard mode, writing is performed with binary data, that is, one dot on or off at 1200 dpi in both the main scanning and sub-scanning directions. High speed modulation of writing laser light, 2 bits per dot
There is also a mode for writing data, that is, four values.
Further, a mode in which writing is performed at 2400 dpi in both the main scanning and sub-scanning modes;
A mode for writing at a pitch of pi vertically and horizontally is also provided.

The photosensitive drum 509 is driven to rotate by a driving unit and rotates clockwise as shown by an arrow.
After being uniformly charged by the writing optical unit 5
The exposure by 08 forms an electrostatic latent image. The electrostatic latent image on the photosensitive drum 509 is developed by the developing device 513 to become a toner image. Further, a plurality of paper feed units 514 to 5
18. Transfer paper (paper) is fed to the registration roller 520 from one of the manual paper feed units 519.

The registration roller 520 is connected to the photosensitive drum 50.
The transfer paper is sent out in synchronization with the toner image on the transfer sheet 9. The transfer bias is applied to the transfer belt 521 from a transfer power supply, and transfers the toner image on the photosensitive drum 509 to the transfer paper and conveys the transfer paper. The transfer paper on which the toner image has been transferred is conveyed to a fixing unit 522 by a transfer belt 521 to fix the toner image thereon, and then is transferred to a discharge tray 5.
It is ejected as a copy to 23. Also, the photosensitive drum 5
In step 09, the toner image is cleaned by the cleaning device 524 after the transfer of the toner image, and the electric charge is gradually decreased by the electric charger 525 to prepare for the next image forming operation.

FIG. 19 shows a circuit section 5 of this digital copying machine.
50 shows an example of a block configuration of 50. The input of the circuit unit 550 is an analog image signal read at 600 dpi by the reading sensor 507 of the scanner unit 400. The circuit unit 550 includes a part for scanner processing and a part for digital image processing for processing and correcting digital image data.

The reading sensor 507 uses 600 dp
The analog image signal read by the i.
After the level is adjusted by 1, the A / D converter 552 converts the data into 8-bit (256 gradations) digital data per pixel, and furthermore, the shading correction circuit 553.
Thereby, the variation of the sensitivity and the illuminance of each pixel of the reading sensor 507 is corrected.

The image data after the shading correction is sent to a filter processing circuit 554, where the image data is filtered.
Specifically, MTF correction for correcting the amplitude of an image generated by reading, and smoothing processing for smoothly expressing a halftone image are performed. The image data after the filter processing is subjected to scaling processing in the main scanning direction in accordance with the copy magnification by the main scanning scaling processing section 555, and then subjected to gamma correction for conversion to writing density by the gamma correction section 556. Will be applied. The image data after the gamma correction is sent to the halftone processing unit 557. The image processing apparatus of the present invention as shown in FIGS. 1, 6, 14 or 15 is provided in the halftone processing section 557. When an image processing apparatus having an image feature extraction unit as shown in FIG. 6, FIG. 14, or FIG. 15 is used, as shown by a broken line in FIG. Input to the image feature extractor will generally be preferred. The image data processed by the halftone processing unit 557 is sent to the light emission drive control unit of the semiconductor laser in the writing optical unit 508 as 1-bit or 2-bit data per dot. In addition, many processes such as background removal processing, flare removal processing, and image editing are often enabled in a digital copying machine, but the description thereof is omitted.

[0063]

As described in detail above, according to the present invention, (1) high-resolution binary image data for forming a high-quality image can be converted from low-resolution multi-value image data at a high speed. Can be generated. (2) An image having excellent graininess and stability can be formed. (3) It is possible to prevent excessive concentration of dots in the low-density part of the image and improve the granularity of the low-density part of the image. (4) Distortion of halftone dots in the middle and high density portions of the image can be reduced. (5)
It is possible to form an image that matches image characteristics. (6) Both graininess and stability in a non-edge region of an image and sharpness in an edge region of an image can be compatible. (7)
It is possible to prevent the occurrence of moire in the low frequency halftone dot portion. (8) Deterioration of graininess due to a locally large edge amount can be prevented. (9) The sharpness of the edge region can be improved, and the occurrence of moire can be prevented.
(10) The sharpness of low density characters and the graininess of halftone dots in low density portions can be improved. (11) It is possible to obtain many effects such as suppression of banding.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an example of an image processing apparatus according to the present invention.

FIG. 2 is a diagram illustrating an example of a dither threshold matrix for generating a quantization threshold.

FIG. 3 is a diagram showing thresholds in a dither threshold matrix of FIG. 2 for each quantization threshold;

FIG. 4 is a diagram for explaining a dot arrangement order.

FIG. 5 is a diagram illustrating an example of output data.

FIG. 6 is a block diagram illustrating another example of the image processing apparatus of the present invention.

FIG. 7 is a diagram illustrating an example of an edge extraction filter.

FIG. 8 is a diagram illustrating another example of the edge extraction filter.

FIG. 9 is a diagram illustrating an example of a dither coefficient matrix.

FIG. 10 is a diagram collectively showing coefficients in a dither coefficient matrix of FIG. 9 for each quantization threshold.

FIG. 11 is a diagram collectively showing thresholds in a dither threshold matrix used in a region where the edge amount = 0 (A = 3) for each quantization threshold.

FIG. 12 is a diagram showing thresholds in a dither threshold matrix used in an area of edge amount = 1 (A = 2) for each quantization threshold;

FIG. 13 is a diagram collectively showing thresholds in a dither threshold matrix used in a region where the edge amount = 2 (A = 1) for each quantization threshold.

FIG. 14 is a block diagram showing another example of the image processing apparatus according to the present invention.

FIG. 15 is a block diagram showing another example of the image processing apparatus according to the present invention.

FIG. 16 is a diagram illustrating a relationship between a quantization threshold generated in an edge region and an input data value.

FIG. 17 is a diagram for explaining a dot arrangement order in an edge area.

FIG. 18 is a schematic sectional view of a digital copying machine to which the present invention is applied.

19 is a block diagram of a circuit section of the digital copying machine shown in FIG.

[Explanation of symbols]

 REFERENCE SIGNS LIST 100 error diffusion quantization processing unit 101 adder 102 quantizer 103 subtractor 104 error diffusion calculation unit 200, 200A resolution conversion binarization unit 201 dot number determination unit 202, 202A dot output position determination unit 203 output buffer 300, 300A , 300B quantization threshold generator 301 position information generator 302 delay element 350, 350A image feature extractor 351 edge amount calculator 352 edge amount quantizer 353 edge amount expansion processor 354 edge amount averaging unit

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2C262 AA24 AA26 AA27 AB13 AB19 BB01 BB08 BB22 BB23 BB27 DA03 DA16 5B057 CA02 CA08 CA12 CA16 CB02 CB07 CB12 CC01 CE13 CH07 CH08 CH09 5C077 LL19 MP01 NN09 NN11 P47 PP48 PP48 PP48 AA06 DA01 FA06 GA02 GA17

Claims (36)

[Claims] 1. A quantization threshold generation means for generating a plurality of quantization thresholds corresponding to each pixel of multi-valued image data in accordance with a dither threshold matrix, and a plurality of quantization thresholds generated by the quantization threshold generation means. First processing means for multi-value quantization of the multi-valued image data by error diffusion processing and outputting quantized data; and converting the quantized data into binary image data having a higher resolution than the multi-valued image data. A second processing unit, wherein the second processing unit determines the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data, And the arrangement order of the dot-on pixels in the plurality of pixel regions is controlled in accordance with the corresponding position of the pixel of interest on the dither threshold matrix. An image processing apparatus characterized by the above-mentioned. 2. The image processing apparatus according to claim 1, wherein the arrangement order of the dot-on pixels is controlled so as to form a dot concentration type halftone dot. 3. The image processing apparatus according to claim 2, wherein threshold values are arranged in the dither threshold value matrix so as to form dot concentration type halftone dots. 4. The image processing apparatus according to claim 3, wherein the lowest to fourth thresholds in the dither threshold matrix are arranged at different pixel positions. 5. The image processing apparatus according to claim 4, wherein a difference between a fourth threshold value and a fifth threshold value in the dither threshold matrix is smaller than a step width of the dither threshold matrix. 6. The dither threshold matrix is obtained by connecting two or more basic dither threshold matrices in which thresholds are arranged so as to form dot concentration type halftone dots in the main scanning direction while being shifted in the sub-scanning direction. 4. The image processing apparatus according to claim 3, wherein 7. An image feature extraction unit for extracting an image feature of the multi-valued image data, wherein the quantization threshold generation unit outputs the amplitudes of the plurality of quantization thresholds from the image feature extraction unit. The image processing apparatus according to claim 1, wherein the control is performed in accordance with the feature amount. 8. The image processing apparatus according to claim 7, wherein the control of the amplitude of the quantization threshold is performed by switching a dither threshold matrix used for generating the quantization threshold. 9. The image feature extraction unit outputs an edge amount of the multi-valued image data as a feature amount, and the quantization threshold generation unit outputs the plurality of image data as the edge amount output from the image feature extraction unit increases. The image processing apparatus according to claim 7, wherein the amplitude of the quantization threshold is reduced. 10. The image processing apparatus according to claim 9, wherein said image feature extracting means expands an edge amount of said multi-valued image data before outputting. 11. The image processing apparatus according to claim 9, wherein said image feature extraction means averages and outputs an edge amount of said multi-valued image data. 12. The quantization threshold generation means, wherein the plurality of quantization thresholds are all set to the same fixed value when the edge amount output from the image feature extraction means is maximum. 12. The image processing apparatus according to 9, 10, or 11. 13. The quantization threshold value generating means, when the edge amount output from the image feature extracting means is maximum,
12. The image processing apparatus according to claim 9, wherein a value that changes according to a value of the multilevel image data is generated as the plurality of quantization thresholds. 14. The image processing apparatus according to claim 13, wherein an average value of a target pixel and its neighboring pixels is used as a value of the multi-valued image data. 15. The quantization threshold value generating means changes a value that changes in accordance with the value of the multi-valued image data with the number of quantizations performed by the first processing means as the value of the multi-valued image data increases. The image processing apparatus according to claim 13, wherein the image processing apparatus is changed so as to decrease the value. 16. When the edge amount output from the image feature extraction unit is the maximum, the second processing unit performs a plurality of pixel regions of the binary image data corresponding to a target pixel of the multi-valued image data. 16. The image processing apparatus according to claim 15, wherein dot-on pixels are arranged in a predetermined arrangement order. 17. Quantized data of multi-valued image data, which has been multi-valued quantized by error diffusion processing using a plurality of quantization thresholds generated according to a dither threshold value matrix, is converted into binary image data of higher resolution. An apparatus for converting the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data according to the quantized data value of the target pixel First means for determining, and the arrangement order of the number of dot-on pixels determined by the first means in a plurality of pixel areas of the binary image data corresponding to the target pixel of the multi-valued image data, Second means for controlling the target pixel in accordance with a corresponding position on the dither threshold matrix. 18. The image processing apparatus according to claim 17, wherein the arrangement order of the dot-on pixels in a plurality of pixel regions is controlled so as to form a dot concentration type halftone dot. 19. The method according to claim 19, wherein the second means is provided with information indicating an edge area of the multi-valued image data, and stores information in a plurality of pixel areas of the binary image data corresponding to pixels in the edge area of the multi-valued image data 18. The image processing apparatus according to claim 17, wherein dot-on pixels are arranged in a predetermined arrangement order. 20. The image processing apparatus according to claim 1, further comprising means for forming an image in accordance with said binary image data.
The image processing apparatus according to claim 1. 21. The apparatus according to claim 1, further comprising means for inputting said multi-valued image data by optically scanning a document, and means for forming an image in accordance with said binary image data. The image processing device according to claim 1. 22. A recording medium readable by a computer, wherein a program for realizing each unit of the image processing apparatus according to claim 1 by a computer is recorded. Recording medium. 23. A quantization threshold generation step of generating a plurality of quantization thresholds corresponding to each pixel of the multi-valued image data according to a dither threshold matrix, using a plurality of quantization thresholds generated by the quantization threshold generation step. A quantization step of generating quantized data by performing multi-level quantization on the multi-level image data by an error diffusion process; and a converting step of converting the quantized data into binary image data having a higher resolution than the multi-level image data In the converting step, the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data is determined according to the quantized data value of the target pixel. And the arrangement order of the dot-on pixels in the plurality of pixel regions is determined according to the corresponding position of the pixel of interest on the dither threshold matrix. An image processing method characterized by controlling. 24. The image processing method according to claim 23, wherein the arrangement order of the dot-on pixels is controlled so as to form a dot concentration type halftone dot. 25. An image feature extracting step for extracting an image feature of the multi-valued image data, wherein the quantization threshold generation step extracts the amplitudes of the plurality of quantization thresholds by the image feature extraction step. The image processing method according to claim 23, wherein the control is performed in accordance with the feature amount. 26. The image feature extraction step extracts an edge amount of the multi-valued image data as a feature amount, and the quantization threshold generation step includes a step of generating a plurality of the plurality of image data as the edge amount extracted by the image feature extraction step increases. The image processing method according to claim 25, wherein the amplitude of the quantization threshold is reduced. 27. The image processing method according to claim 26, wherein, in the image feature extracting step, a result obtained by expanding an edge amount of the multi-valued image data is extracted as a feature amount. 28. The image processing method according to claim 26, wherein the image feature extracting step extracts an averaged edge amount of the multi-valued image data as a feature amount. 29. The quantization threshold generation step, wherein the plurality of quantization thresholds are all set to the same fixed value when the edge amount extracted in the image feature extraction step is maximum. 29. The image processing method according to 26, 27 or 28. 30. The quantization threshold generation step, wherein when the edge amount extracted in the image feature extraction step is a maximum, the plurality of quantization thresholds change according to the value of the multi-valued image data. 29. The image processing method according to claim 26, 27 or 28, wherein 31. The image processing method according to claim 30, wherein an average value of a pixel of interest and its neighboring pixels is used as the value of the multi-valued image data. 32. The quantizing threshold generation step, wherein a value that changes in accordance with the value of the multi-valued image data is reduced as the value of the multi-valued image data increases. 32. The image processing method according to claim 30, wherein the image processing method is changed so as to perform the processing. 33. When the edge amount extracted by the image feature extraction step is the maximum, the conversion step includes: setting a dot in a plurality of pixel areas of the binary image data corresponding to a target pixel of the multi-valued image data. The image processing method according to claim 32, wherein the ON pixels are arranged in a predetermined arrangement order. 34. Quantized data of multi-valued image data multi-valued quantized by error diffusion processing using a plurality of quantization thresholds generated according to a dither threshold value matrix, and binary image data of higher resolution A method for converting the number of dot-on pixels in a plurality of pixel regions of the binary image data corresponding to the target pixel of the multi-valued image data according to the quantized data value of the target pixel A first step of determining, and the arrangement order of the number of dot-on pixels determined in the first step in a plurality of pixel regions of the binary image data corresponding to a target pixel of the multi-valued image data, A second step of controlling the target pixel in accordance with a corresponding position on the dither threshold matrix. 35. The image processing method according to claim 34, wherein the arrangement order of the dot-on pixels in the plurality of pixel regions is controlled so as to form a dot-concentrated halftone dot. 36. The method according to claim 36, wherein the second step arranges dot-on pixels in a predetermined arrangement order in a plurality of pixel areas of the binary image data corresponding to pixels in an edge area of the multilevel image data. The image processing method according to claim 34, wherein