JPH02292954A - Binarizing method for gradation image - Google Patents

Abstract

PURPOSE:To eliminate a blackish highlight part which is produced with use of a low image resolution storage by increasing a valid output image part each dot or every plural dots in accordance with an input level in order to obtain a binarized pseudo gradation image of high quality corresponding to an input signal having the gradation. CONSTITUTION:When the threshold value matrices 11 are arranged vertically and horizontally with no space secured among them, a threshold value data pattern of an area having the same size as the matrix formed around the apex of said matrix 11 is formed so as to be equal to the matrix 11. An input image is binarized with use of the matrix 11 and its output image is turned into a mesh image having a screen angle of 45 deg. which is used for printing of the black/ white photos in a newspaper, etc. The mesh are least conspicuous at the 45 deg. screen angle. As a result, the picture quality is improved.

Description

[Detailed Description of the Invention] [Objective of the Invention] (Field of Industrial Application) This invention provides a gradation method suitable for expressing an input image having gradations in binary format, for example, black and white, especially in a printer device. This invention relates to an image binarization method.

(Prior art) For pseudo-gradation reproduction (pseudo-gradation expression) of an image with gradation (gradation image) read by a scanner etc. using a binary recording device such as a printer (by area modulation) As a binarization method, a dither method using a threshold data matrix called a dither matrix is conventionally known. This dither method is a method of expressing gradation in a pseudo manner by changing the threshold value according to the input level of each pixel of an input image having gradation. In Figure 4, 4 is used as a threshold to express 16 gradations.
An example of a dither method using an X4 dither matrix 4l will be shown. In this 4×4 dither matrix 4l, 16 pieces of threshold data representing gradations from 1 to 16 are arranged in a matrix.

In the dither method shown in FIG. 4, the input image 42 is divided into 4×4 matrix units, and for each divided image, the input level indicating the gradation of each pixel is compared with the level of the corresponding threshold data of the dither matrix 41. be done. As a result of this comparison,
When the input level≧threshold data level, the corresponding pixel is set to black (“1”), and when the input level is less than the threshold data level, the corresponding pixel is set to white (“0′).

With the dithering method described above, when gradation is reproduced using a high resolution recording device, a reproduced image of relatively good quality can be obtained. However, this dithering method can be applied to LBP with poor resolution.
When applied to low-resolution recording devices such as (laser beam printers), the black (black ink) dot diameter is too large, so the black is emphasized overall, making it difficult to obtain sufficient gradation. there were. This situation is shown in FIG. 5 in the form of a relationship between the input level and the density of the reproduced image (output image).

As is clear from the figure, the dithering method cannot properly reproduce the gradation of the highlight area, resulting in an overall dark image.

(Problems to be Solved by the Invention) As mentioned above, in the conventional gradation image binarization method using a dither matrix, the diameter of the black dots is too large in the case of low resolution, resulting in an entirely black image. Therefore, there was a problem that the gradation was insufficient and the image quality was poor.

Therefore, the problem to be solved by the present invention is to enable a low-resolution recording device or the like to easily obtain a high-quality binary pseudo-gradation image from an input image having gradations.

[Structure of the Invention] (Means for Solving the Problems) This invention provides each threshold data of a threshold data matrix for binarizing each pixel of an input image having gradation to give a pseudo gradation image. The value of is determined so that each effective part of the output image corresponding to the input image consisting of pixels of the same input novel is sequentially enlarged by one dot or several dots according to the input level, and each of the arbitrary input images is A binary pseudo gradation image corresponding to the arbitrary input image is obtained by sequentially comparing pixels with corresponding threshold data in the threshold data matrix and binarizing the pixels. be.

(Operation) According to the above configuration, the effective output image portion (black portion) is
Instead of fixedly increasing by 1 dot regardless of the input level as in the past, it increases by 1 dot or multiple dots depending on the input level.
It becomes possible to make the size of the effective output image part of the highlight part relatively small, and the inconvenience that the highlight part darkens when a low resolution recording device is used is eliminated.

(Example) Figure 1 shows threshold data matrix 1 applied in this invention.
1 is a matrix structure diagram showing an example of No. 1; FIG. The maximum value of the elements (threshold data) of the threshold data matrix 11 in FIG. 1 is 15. However, the size of the matrix 1l in Fig. 1 is 8x8, and the size of the matrix 1l in Fig. 4 is 8x8.
Please note that this is not an X4. That is, in this example,
By increasing the size of the matrix, the degree gradation of the highlighted area is brought out, as will be described later. The threshold value data matrix l1 in FIG. 1 is such that each effective part of the output image corresponding to the input image consisting of pixels of the same input level is sequentially enlarged by one dot or several dots according to the input level. It is made up of a threshold data pattern with a predetermined value. This threshold data pattern includes 2 pieces of threshold data each with a value (gradation) of 1 to 4 and 12 to 15, and 4 pieces of threshold chain data each with a value of 5.6.
, and eight pieces of threshold data each having a value of 7 to 11. According to the pattern of this matrix 11, when the same matrix 11 is arranged vertically and horizontally without gaps as shown in FIG. It will be expanded sequentially depending on the level. This effective portion will be approximately circular.

Now, when using the threshold value data matrix l1 of FIG. 1, in the highlight part where the input level is low (here, the area where the input level is 1 to 4), the effective part (black dot) is
Each time the input level increases by 1, it increases by 1 dot. This amount of increase is the same as in the case of dither matrix 4l in Fig. 4, but the effective part (black dots) for the matrix size
The relative size of matrix ll in FIG. 1 is smaller.

Therefore, the input image is divided into 8×8 matrix units, and for each divided image, the input level indicating the gradation of each pixel is compared with the level of the corresponding threshold data in the threshold data matrix 11 in FIG. Then, based on the comparison results, the input image is converted to 2fL in the same manner as before, and when recorded with a low resolution recording device such as an LBP (laser beam printer) with poor resolution, the dither matrix 4L shown in Fig. 4 is used. Since the relative dot diameter of black (black ink) is smaller than in the case of black (black ink), the gradation of the highlighted area is sufficiently expressed.

Next, in the area of intermediate input levels (area of levels 5 to 11), the density of the binarized image (output image) is conventionally
As shown in the figure, the gradation did not change much with respect to the input level, and the gradation was not expressed correctly. However, in this example, for input levels 5 and 6, 2 dots each, and for input levels 8 to 1
Since each dot is increased by 4 dots relative to 1, gradation expression corresponding to the input level can be performed. Furthermore, in areas with high input levels (areas with levels 12 to 15), the density increases due to the influence (synergistic effect) of (the density of) the output images of adjacent matrix-sized divided images, as is clear from Figure 2. Although there is a risk, in this embodiment, the increase in the number of dots is suppressed to the same extent as in the highlight area, so that it is possible to prevent the density from becoming abnormally high as in the conventional case (see FIG. 5). The above situation is shown in FIG. Note that the number of dots that increases in response to the input level varies depending on the recording device, so the actual aP
It is preferable to use the optimum value determined from the relationship between the input level and the density using No. 1.

Now, the threshold data matrix 11 in FIG. 1 is an area of the same size as the matrix 11 centered at the vertex of each matrix 1l, when the same matrix 11 is arranged vertically and horizontally without gaps as shown in FIG. 2. The threshold value data pattern of the matrix 11 is configured to be the same as that of the matrix 11. Therefore, when the input image is binarized using the threshold value data matrix 11 shown in FIG. 1, the output image will be a mesh image with a screen angle of 45'', unlike a dithered image.45
@Screen is used for printing black and white newspaper photos, etc., and the mesh is at the angle where it is least noticeable.
Image quality can be improved.

In the above embodiment, the case where the size of the threshold data matrix is 8x8 was explained, but each effective part of the output image corresponding to the input image consisting of pixels of the same input level is one dot or several dots depending on the input level. Any threshold data matrix may be used as long as it has a threshold data pattern in which the values of the threshold data are determined so as to be sequentially enlarged dot by dot, and the maximum value of the threshold data is not limited to 15.

[Effects of the Invention] As detailed above, according to the present invention, since the effective output image portion is increased by one dot or multiple dots depending on the input level, it is possible to cope with input images having gradations. It is possible to easily obtain a high-quality binary pseudogradation image, and the size of the effective output image part, especially the highlight part, can be made relatively small, so when using a low-resolution recording device, the highlight It is possible to eliminate the inconvenience of darkening of the skin.

[Brief explanation of drawings]

FIG. 1 is a matrix structure diagram showing an example of a threshold data matrix applied in the present invention, FIG. 2 is a diagram for explaining the characteristics of the threshold data pattern of the matrix in FIG. 1, and FIG. A diagram showing the density of an output image obtained using the matrix in Figure 1 in correspondence with the input level, and Figure 4 is for explaining a gradation image binarization method using a conventional threshold data matrix (dither matrix). and FIG. 5 are diagrams showing the density of an output image obtained using a dither matrix in correspondence with the input level. 1l... Threshold data matrix, 41... Dither matrix. Input page Figure 3

Claims (2)

[Claims] (1) A threshold data matrix for obtaining a pseudo gradation image by binarizing each pixel of an input image having gradation,
It has a threshold data pattern in which the value of the threshold data is determined so that each effective part of the output image corresponding to the input image consisting of pixels of the same input level is sequentially enlarged by one dot or several dots according to the input level. Using a threshold data matrix, each pixel of an arbitrary input image is sequentially compared with the corresponding threshold data in the threshold data matrix and binarized, thereby creating a binary pseudo gradation image corresponding to the above arbitrary input image. A gradation image binarization method characterized in that the gradation image is binarized. (2) The above threshold data matrix is configured such that when the same matrix is arranged vertically and horizontally without gaps, the threshold data pattern of the area of the same size centered on the vertex of each matrix is the same as that of the same matrix. The gradation image binarization method according to claim 1, characterized in that: