JP2898836B2 - Image processing method - 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 method, and more particularly, to an image processing method suitable for use in an apparatus which needs to quantize multi-valued image data, such as a digital copying machine and a facsimile machine. .

[0002]

2. Description of the Related Art Conventionally, various methods, such as a dither method and an error diffusion method, have been proposed as methods of quantizing multi-valued image data. Was performed.

A specific example of quantization using the error diffusion method will be described. When the target pixel A is quantized with x, the error is expressed by the following equation. A− [x] = ε (1) Here, [x] is multi-valued data corresponding to the quantized data. For example, if the pixel data A is 8 bits and 0
When the quantization takes a binary value of 0, x = 0 or 1, and when x = 0, [x] = 0,
When x = 1, [x] = 255.

The above error ε is shown in FIG.
As shown, it is distributed to peripheral pixels B, C, D, and E in a predetermined manner, and is added to each. That is, the error is ε = ε
_B ε _C ε _{D D} ε _E , B = B + ε _B , C = C + ε
_C, and _{D = D + ε D, E} = E + ε E gutter Tsu form, is added to each of the surrounding pixels. Next, the above process is performed using the B pixel as the target pixel, and thereafter, the processes for the C and D pixels are performed in the same manner.

[0005]

However, according to the above-mentioned conventional method, since the target pixel is a single pixel unit, the dynamic range of the halftone to be handled is limited to a maximum of one pixel, and the dynamic range beyond that is limited. I couldn't have a range. For example, if the multi-valued data is 8-bit pixel data, it can have a dynamic range of 256 gradations, but conversely, it cannot be quantized with a larger range.

[0006]

In order to achieve the above object, the present invention provides (1) an image processing method for quantizing multi-valued image data, wherein an area to be quantized is set to N × M pixels, and N × M pixels are set. quantized on the basis of multi-value image data of each pixel in the area of M pixels to the value calculated by a predetermined formula, quantization
Based on the result (quantization pattern)
Setting data for each pixel, or (2) multi-valued image
In an image processing method for quantizing image data, a quantization pair
The elephant area is N × M pixels, and the area within the N × M pixels area
The multi-valued image data of each pixel is calculated based on a value calculated by a predetermined formula.
Then, based on the quantization result (quantization pattern), N × M data for each pixel can be determined in advance.
In accordance with the predetermined weighting .

[0007]

In the quantization of the image data, the image to be processed is not limited to one pixel, but is set to the N × M area pixels and the N × M area as one unit of the processing. The dynamic range of the halftone to be handled is extended to N × M times as compared with that of the unit of one pixel for processing. Here, N and M are integers, which means that N × M is 2 or more.

[0008]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention divides a pixel of interest (target) into N × M areas when quantizing image data, and treats each area as one unit.
When the image data before quantization is 8 bits and the dynamic range of the halftone is 256 gradations, the dynamic range is dramatically increased to 512 gradations by simply setting N = 1 and M = 2. And a significant improvement in image quality can be achieved. It should be noted that the effect of this image quality improvement is that processing is performed without changing the dynamic range of the original image data.
This is the effect when xM is set. Conversely, when NxM is set to 2,
Further, in order to set the dynamic range after quantization to 256 gradations, the original image data becomes 256/2, and 128
It means that the gradation, that is, the resolution of 7 bits is sufficient, and there is also an effect that the load on the hardware circuit such as A / D conversion can be reduced while maintaining the dynamic range of the halftone.

An embodiment in which N = 2 and M = 2 in N × M will be described below. Each pixel data is represented by 8 bits and has a dynamic range of 256 gradations per pixel. The quantization is binary, and the result is represented by 0 or 1, and the quantization threshold is 128, which is the median.

In FIG. 1, assuming that the target area corresponding to the N × M target area is four pixels within the broken line, the total pixel data within the broken line is A = Di, j + 2 + Di in the target pixel area A. , j + 3 + Di + 1, j + 2 + Di + 1, j + 3 (2) Similarly, if the target pixel areas around the target pixel area A are B, C, D, and E, as in A, B = Di, j + 4 + Di, j + 5 + Di + 1, j + 4 + Di + 1, j +5 (3) C = Di + 2, j + 4 + Di + 2, j + 5 + Di + 3, j + 4 + Di + 3, j + 5 (4) D = Di + 2, j + 2 + Di + 2, j + 3 + Di + 3, j + 2 + Di + 3, j + 3 (5) E = Di + 2, j + Di + 2, j + 1 + Di + 3, j + Di + 3, j + 1 (6) Here, according to the above formula, the possible range of each pixel area is 0 to 255 for each pixel.
It means that a value of X255 × 4 = 1020 can be taken.

[0011] The quantization is based on the four pixels in the area of interest .
The multi-valued data is converted into a predetermined formula, for example, as described above.
Is calculated (added) based on the equation, and based on the result,
The four pixels, as represented in binary (0 or 1)
And follow the steps below. Each pixel of the above noted area A
When the addition result of the multi-valued data of is 0 to 128, FIG.
As shown in FIG. 3, the result of quantization is defined as a quantization pattern 0.
Based on the result, as shown in FIG.
Data in which all four pixels are set to 0 is set. The quantization error at this time is A. These relationships are shown in FIG.
In FIG. 3, a table is shown as specific numerical data in FIG.

Next, when the value of A exceeds 128 and is 383, one of the four pixels is set to 1, the remaining three pixels are set to 0 (FIG. 4B), and the quantization pattern is set to 1. ,
The quantization error at this time is A-255 (see FIGS. 2 and 3).

Similarly, when A exceeds 383 and is between 638, two out of four pixels are set to 1 and the remaining two pixels are set to 0 (FIG. 4 (c)), and the quantization pattern is set to 2. , And the error is A-510 (see FIGS. 2 and 3). When A exceeds 638 and is between 893, three of the four pixels are set to 1 and the remaining one is set to 0 (FIG. 4).
(D)), assuming that the quantization pattern is 3, the error is A-7.
65, and when A exceeds 893 and is between 1020,
Assuming that all four pixels are 1 (FIG. 4 (e)) and the quantization pattern is 4, the error is A-1020 (FIG. 2, FIG.
(See FIG. 3).

Next, a specific method of corresponding binarized data to each pixel based on the quantization pattern will be described. FIG.
Indicates 2 × 2 pixels in the attention area A, which indicates the priority of pixels to be printed. That is, the priority of printing is assigned to pixels having a small numerical value. 1 is Di, j + 3 in FIG. 1, 2 is Di + 1, j + 3 in FIG. 1, 3 is Di, j + 2 in FIG. 1, and 4 is Di, j + 2 in FIG. +1 and j + 2 pixels are shown. Accordingly, in the quantization patterns 0, 1, 2, 3, and 4 shown in FIG. 3, the pixel data of the target area after the quantization is (a) (b) (c) (d) in FIG.
(E).

As described above, the multi-value of each pixel in the area of interest A
As described in the procedure for performing quantization based on data,
Based on the result of quantization, the data of each pixel is
4 are set. At this time, FIG.
FIG. 4 (e) is dark and the one in between is binary data.
Data can be reproduced effectively even if quantization is applied to the data.
And Therefore, the error generated as a result of quantization is
As described above, the peripheral pixels, that is, the pixel areas B,
C, D, and E, respectively. As for the distribution method, it is possible to perform proportional distribution similarly to the conventional error diffusion, or to change the weight of the distribution by using a random number. Here, the method does not matter. In any case, all of the error is distributed to peripheral pixels.

With the above, the processing for the area around the center A is completed. Then, the target area is shifted by two pixels in the main scanning direction. Quantization is performed in the same procedure as in the above method A. When the main scanning direction ends, the target pixel is shifted by two pixels in the sub-scanning direction, so that the next line is quantized.

As described above, all the pixels in the main / sub-scan are quantized. In this quantization,
By setting four target pixels, 4 × 255 = 1020
This means that the gradation density has been preserved. In terms of area, despite having 255 gradations per pixel, it has a table capability of 1020 gradations, and it is possible to execute dense and expressive image processing on halftone images. .

FIG. 6A shows the priorities of quantization when 4 × 4 pixels are set as a target area, which are ranked concentrically around a specific pixel. The weighting also makes it possible to process halftones using a halftone dot representation technique. Also, as shown in FIG. 6B, it is possible to express a halftone streak by weighting in a specific scanning direction. Therefore, if a variety of such weighted conversion tables are prepared and these are switched arbitrarily by software, they can be used for a wide variety of image processing, and significant advantages can be obtained. .

[0019]

As is apparent from the above description, according to the present invention, the pixel area to be processed is expanded to a plurality of N × M pixels when quantizing the image data, so that the halftone image is processed. The dynamic range can be increased by N × M times, and further, in N × M pixels,
N × M data for each pixel is set based on the quantization result.
Since the pixel data is determined or set in advance in accordance with a predetermined weighting, it is possible to perform a dense and expressive image processing on the halftone image.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining an embodiment of the present invention, in which target pixel areas (A, B,... E) are represented by N (= 2), M (=
It is a figure which shows the example at the time of expanding 4 times of 2).

FIG. 2 is a graph showing an example of a quantization calculation method.

FIG. 3 is a table showing the quantization calculation method shown in FIG. 2 as numerical data.

FIG. 4 is a diagram showing pixel data of an area of interest after quantization of quantization patterns 0, 1, 2, 3, and 4;

FIG. 5 is a diagram showing pixels (2 × 2) in a pixel area of interest.

FIG. 6 is a diagram illustrating an example of weighting for quantization.

FIG. 7 is a diagram for explaining an example of an error diffusion method.

Claims (2)

(57) [Claims] 1. An image processing method for quantizing multi-valued image data, wherein a quantization target area is N × M pixels, and N × M pixels
Multi-valued image data of each pixel in the area of M pixels is quantized based on a value calculated by a predetermined formula , and the result of quantization is obtained.
(Quantization pattern) based on each of N × M pixels
An image processing method characterized by setting data (where N and M are integers and N × M ≧ 2). 2. An image processing method for quantizing multi-valued image data.
In the method, the quantization target area is N × M pixels, and N × M pixels
The multivalued image data of each pixel in the area of M pixels is
Quantization is performed based on the value calculated by the equation, and based on the quantization result (quantization pattern), data for each of N × M pixels is set according to a predetermined weight.
Image processing method according to claim Rukoto (however, N, M is an integer, a N × M ≧ 2).