JP2001036745A - Device and method for generating image processing matrix - Google Patents

Abstract

PROBLEM TO BE SOLVED: To simplify a binarization processing method, to accelerate binarization processing rate and to improve picture quality by deciding the mutual threshold of both small matrices adjacent to each other by means of plural small matrices obtained by dividing the size of a matrix that is to be generated. SOLUTION: The size of a large matrix to produced is decided in consideration of the resolution and the number of scales of an image, and the storing area of the large matrix is set in a threshold storage means 3. For instance, the size of the large matrix is set at (256×256) and then the large matrix is divided into 256 small matrices of each (16×16) size. The threshold of scale 0 is set for every small matrix and then the threshold is calculated for the scales 1-256 respectively. For instance, the threshold of scale 1 is calculated at an optimum position within (48×48) matrices in all including 8 small matrices which are adjacent vertically and horizontally to each other and 9 small matrices whose threshold should be decided. The calculation is carried out for every small matrix.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing matrix creating apparatus and method for use in image processing of recording devices such as color and monochrome printers and plotters.

[0002]

2. Description of the Related Art When printing with a digital color or monochrome printer, when printing on a binary output device such as an ink jet system, a laser system, or a fusion type thermal transfer system, multi-level data such as 256 gradations is used. It is necessary to convert a full-color image or a grayscale image into binary data. This is binarized (halftone processing)
That. At the time of this conversion, a method for preventing the resolution and gradation of the image from deteriorating is important.

[0003] One of the binarization methods is a dither method. The dither method arranges the threshold values of the matrix regularly.
Broadly, there are a dispersion type and a centralized type. In the dispersion type, the dot arrangement is formed by one dot as much as possible, so that the resolution becomes high. This method is often used for a fusion-type thermal transfer system or the like. In addition, since the centralized type prints by arranging dots so that the dots become thicker in order with the center as a nucleus, it is often used for a laser method or the like in which one dot is not stable.
In general, the dither method has good gradation but low resolution. For example, if the matrix size is 16 × 16 at 300 dpi, 256 gradations can be obtained, but the resolution is 300/16 = 18.7.
The value is as low as 5 dpi.

As another method of the binarization processing, there is an error diffusion method. In this method, error data for each pixel generated at the time of the binarization processing is obtained by assigning a certain weighting factor to some nearby pixels. It is a method of spreading and processing. In this method, since the error of each pixel data is diffused to the neighboring dots as much as possible, the image is an image close to the original image with high gradation and high resolution. This method is often used for an ink-jet method having good stability for each dot.
However, there is a disadvantage that the processing speed is slow because calculation such as multiplication is required.

As a method of solving these drawbacks, an FM screen using a blue noise mask has recently attracted attention. This method can realize the resolution and the gradation like the error diffusion method while using a matrix like the dither method. Blue noise is obtained by removing low frequency components of white noise, and is a matrix of only high frequency components.

In order to generate this blue noise mask, a considerable amount of calculation is currently required using Fourier transform, inverse Fourier transform, variance, and the like.

[0007]

As described above,
Binary processing using a blue noise mask can obtain an image quality similar to the error diffusion method, but because a matrix is generated by calculation such as Fourier transform, inverse Fourier transform or variance, it is necessary to generate a blue noise mask. The algorithm is complicated and the amount of calculation is large. Therefore, there is a problem that the image quality is high but the image processing speed is low.

Accordingly, an object of the present invention is to simplify the method of generating a matrix of an FM screen, which is one method of the binarization processing, and to generate an image which does not deteriorate the resolution and gradation. To improve the binarization processing speed and image quality.

[0009]

In order to solve the above problems, according to the present invention, a required matrix size to be created is set, the set matrix is divided into a plurality of small matrices, and a threshold value is calculated. A threshold value calculating means, a threshold value storing means for storing the calculated threshold value, a processing procedure storing means for storing the generating method and the processing procedure, and a control means for controlling these and controlling the output to the comparing means. Configure the device.

In this configuration, first, the size of a matrix to be created is divided into a plurality of small matrices, and an operation for determining a threshold value between adjacent small matrices is performed using the small matrices. According to the above configuration and method, the calculation is omitted for the distant pixels having relatively little influence in determining the threshold value, and the calculation is performed using only the matrix of the adjacent neighborhood having a high influence. , It is possible to determine the optimum threshold value, that is, to arrange the dots with a small amount of calculation without reducing. Therefore,
Improves the binarization processing speed while maintaining high resolution and high gradation.

[0011]

Embodiments of the present invention will be described below. FIG. 1 is a block diagram showing an embodiment of the present invention. The comparison means 2 compares the threshold value of the matrix output from the matrix creation device 1 of the present invention with the image data,
The input image multi-value data is converted into binary data. The comparison means 2 is not particularly limited as long as it has the same function as the comparison circuit used in the halftone processing so far.

First, the configuration of the matrix creation device 1 of the present invention will be described. The threshold value storage means 3 stores a matrix of a required size to be created (hereinafter, referred to as a large matrix).
It stores a threshold value calculated and calculated, and is not particularly limited as long as it can store data such as a memory device such as a RAM or a storage device such as a magnetic disk. The threshold value calculating means 4 includes a plurality of small matrices (hereinafter, referred to as a small matrix) generated by dividing a large matrix.
The threshold value is calculated for the small matrix, and finally the threshold value in the large matrix is calculated. Processing procedure storage means 5
Stores a procedure for calculating and determining a threshold value, such as a large matrix division method and size, and a small matrix calculation method and order. This is not particularly limited as long as it can store program data such as a memory device such as a programmable ROM or a storage device such as a magnetic disk. The control means 6 includes a threshold storage means 3 and a threshold calculation means 4 based on the procedure stored in the processing procedure storage means 5.
And the output to the comparing means 2 is controlled. In FIG. 1, the threshold value calculating means 4 and the control means 6 are separately described, but may be configured by a processor including both functions such as a processor (CPU).

Next, a method of creating a matrix in the matrix creating apparatus 1 configured as described above will be described. FIG. 2 shows a flowchart of the matrix creation method. First, the size (number of pixels: L × M) of a large matrix to be created is determined in consideration of the image resolution and the number of gradations, and a storage area in the threshold storage unit 3 is set. This is step S1. For easy understanding, the size of the large matrix (number of pixels: L × M) is 256 × 256, and the size of the small matrix to be divided (number of pixels: S × K) is 16 × 16 in consideration of the number of gradations. This will be described below.

In step S1, after setting the size of the large matrix as 256 × 256, step S2
In, the large matrix is divided into 256 small matrices of 16 × 16 size. This allows one
A threshold of 0 to 255 can be set for a 6 × 16 small matrix. Thus, from 0 to 25 for each small matrix
By setting the threshold value of 5, the threshold values for each gradation can be evenly arranged in the 256 × 256 matrix. FIG. 3 shows the concept of a matrix divided into 256 large matrices and 256 small matrices described above.

Next, in step S3, as shown in FIG. 4, first, 256 thresholds of 0 gradation are set one by one in a small matrix of 16 × 16. In this setting method, at least 8 pixels or more and 24 pixels or less for each 16 × 16 small matrix adjacent to each other vertically and horizontally and obliquely in order to take into account the dispersibility of the dot arrangement.
A threshold value is set at a position distant from a pixel or less, and this is used as an initial matrix. This means that the distance between the pixels of adjacent small matrices is a minimum of 1 pixel and a maximum of 32 pixels (about 3 pixels in the oblique direction).
(5 pixels), the dispersibility can be maintained by setting a threshold value at a position at least 8 pixels or more and 24 pixels or less in consideration of the continuity of the small matrix.

Next, in step S4, from the first gradation, 2
Calculate the threshold up to 55 gradations. First, as shown in FIG. 5, the method of determining the threshold value of the first gradation is as follows: eight 16 × 16 small matrices adjacent to each other vertically and horizontally and diagonally, and a 16 × 16 small matrix in which the threshold is to be determined. Nine pieces total 48 × 4
The calculation is performed so that the threshold value is determined at an optimum position in the matrix of No. 8. In this case, the calculation method is to arrange the pixels where the threshold value has been determined so far in a place having good dispersibility. As a method for this, a threshold value is determined next for a position where a threshold value has been set in a 48 × 48 matrix.
The variance value of the distance from the pixels in the × 16 small matrix is calculated. A threshold value is set at a position where the variance value is minimum.
This is because the threshold value of a pixel located at a short distance has a large influence on the calculation, and the threshold value of a distant pixel has a small influence.

The above operation is performed for each small matrix (256)
The threshold is determined for each of the small matrices. Next, the threshold for the second gradation is determined for each small matrix in the same manner as the method for determining the threshold for one gradation, and the threshold is determined. Hereinafter, the above is repeated to calculate the threshold up to 255 gradations. If the top, bottom, left, right, and diagonally adjacent 16x
In the case where there are no 16 small matrices, as shown in FIG. 6, in consideration of the repetition of the matrix, when determining the threshold value in the small matrix number 1, the numbers 256, 241 adjacent to each other in the up, down, left, and right directions are used. , 242, 16, 2, 32, 17, 1
The calculation is performed using eight small matrices as small matrices used for the calculation. In the description so far, the calculation is performed in consideration of the eight small matrices in addition to the small matrix for determining the threshold, but the calculation is not limited to eight, and the calculation is performed by additionally referring to any other small matrix. Thereby, the matrix can be optimized.

In step S5, one or a plurality of large matrices created by the above calculations are integrated, and the creation of the matrix is completed. The created matrix is stored in step S6.
Is output to the comparison means 2 by the control of the control means 6 at 2
Perform a digitized image. As described above, by performing an operation using the divided small matrices, compared to a method of calculating a variance value for each pixel in a large matrix of 256 × 256 at the maximum, the amount of operation in a small matrix of 48 × 48 is small. Therefore, the amount of calculation is significantly reduced, and the time required to create a matrix can be reduced.

[0019]

As described above, the matrix to be created is divided into a plurality of small matrices composed of a plurality of pixels, and the threshold value is calculated. In determining, a method is employed in which arithmetic processing is performed by focusing on a matrix in the vicinity of an adjacent area having a high degree of influence. As a result, in a binarization process using a matrix, the amount of calculation and the amount of memory are significantly reduced, and a matrix can be created quickly and easily. Also, by using this matrix, high image quality and high speed can be achieved.

[Brief description of the drawings]

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

FIG. 2 is a flowchart illustrating a matrix creation method according to the present invention.

FIG. 3 is an explanatory diagram showing a matrix concept of the present invention.

FIG. 4 is an explanatory diagram showing a threshold value arrangement of a 0th gradation in the present invention.

FIG. 5 is an explanatory diagram showing the arrangement of small matrices used in the arithmetic processing of the present invention.

FIG. 6 is an explanatory diagram showing the arrangement of small matrices used in the arithmetic processing of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Matrix creation apparatus for image processing 2 Comparison means 3 Threshold storage means 4 Threshold calculation means 5 Processing procedure storage means 6 Control means

Claims (6)

[Claims] An image processing matrix creating apparatus for creating a matrix used to convert multi-valued image data into binary data, sets a required matrix size to be created, and sets the set matrix. Threshold calculating means for dividing the matrix into small matrices composed of a plurality of pixels and calculating a threshold; threshold storing means for storing the calculated threshold; processing procedure storing means for storing a creating method and a processing procedure; An image processing matrix creation device, comprising: a threshold value storage unit; and a control unit that controls the processing procedure storage unit and controls output to a comparison unit. 2. The image processing matrix according to claim 1, wherein said threshold value calculating means calculates the threshold value by using small matrices adjacent to each other as calculation targets when calculating the threshold value to be determined. Creating device. 3. The threshold value calculating means, when calculating a threshold value to be determined, performs the threshold value calculation on small matrices in adjacent neighborhoods and on arbitrary small matrices other than adjacent neighbors. The image processing matrix creating apparatus according to claim 1, wherein 4. A method for creating a matrix used to convert multi-valued image data into binary data, wherein a required matrix size to be created is set, and the set matrix is composed of a plurality of pixels. A matrix that is divided into small matrices, and a threshold value is calculated and determined for each of the small matrices. 5. The matrix according to claim 4, wherein when calculating the threshold value of the small matrix, a calculation process is performed between a small matrix to be determined and small matrices adjacent to the small matrix to determine the threshold value. How to make. 6. The method according to claim 4, wherein when the threshold value of the small matrix is calculated, a calculation process is performed between the small matrix to be determined, the small matrix adjacent to the small matrix, and any small matrix other than the adjacent neighbors. And a threshold value.