JPS6387076A - Variable power method for digital picture - Google Patents

Abstract

PURPOSE:To obtain a variable power picture maintaining the gradation properties by turning the binary picture data into a master matrix of (m X m) and segmenting a submatrix of (n X n) out of the master matrix to multiplying the variable power picture by n/m. CONSTITUTION:The binary picture data is first fetched by a buffer 10 of 8 lines and then segmented as master matrices (alpha-I0), (alpha-II0)... (beta-I0), (beta-II0)... of (8 X 8) successively by a master matrix segmenting buffer 12. Then the desired parts bearing circle marks in a figure (I) are segmented as submatrices of (4 X4) out of those master matrices by a submatrix segmenting buffer 14. These segmented submatrices are evolved in a bit map memory 16 in such an array as shown in a figure (II). Then the contents of the memory 16 are outputted a semiconductor laser serving as a light source via an interface 28.

Description

[Detailed description of the invention] (Technical field) The present invention relates to a digital image scaling method.

(Conventional technology) Digital image recording methods are known.

FIG. 8 schematically shows only the main parts of an example of such an image recording type printer.

For example, when image data is input to the controller 30, which is a micro combinator, the controller 30
performs necessary data processing on the image data, outputs a write signal, and modulates the emission intensity of the semiconductor laser +-32.

The laser beam from the semiconductor laser 32 is collimated by the collimator lens 34, enters the rotating polygon mirror 36, is periodically deflected as the rotating polygon mirror 36 rotates, and passes through the fθ lens 38 to form a photoconductive beam. The light is focused on the photoreceptor 40 in the form of a spot, and the photoreceptor 40 is main-scanned. As the photoreceptor 40 rotates in the direction of the arrow, sub-scanning is performed on the photoreceptor 40, and an electrostatic latent image is formed on the photoreceptor 40. Note that, prior to each main scan, the laser beam enters the photosensor 44 via the mirror 42 to generate a synchronization signal. This synchronization signal is input to the controller 30, and the controller 30 controls the start timing of main scanning using the synchronization signal.

Now, the area on the photoreceptor 40 corresponding to the minimum unit of the write signal will be temporarily called a dot.

In a digital printer, the write signal is binarized into O or 1, so each dot on the photoreceptor 40 becomes a spot of laser light! Regardless of whether the IC is irradiated or not, the amount of irradiation light when irradiated is equal to each other. Therefore, in the printed recorded image, there may or may not be a dot-shaped pixel of a constant density in the portion corresponding to the above-mentioned dot. In this way, since a recorded image created by a digital image recording method is made up of a set of dot-like pixels of a constant density, in such a recorded image,
In order to express gradation, it is necessary to use a dither method or an area modulation method known as a density pattern method.

The area modulation method regards a set of multiple dots as the basic unit of density expression, and expresses density by changing the number of constant density dot-shaped pixels in this basic unit in the recorded image. It's a method.

FIGS. 7(1) to ω show four types of well-known density pattern matrices. These density pattern matrices are the basic unit of density expression, and are 8×8 6
The density patterns are made up of multiple sets of 4 dots, and therefore, by using these density patterns, 64 gradations of density from 0 to 64 can be expressed.

The image data is grouped into data corresponding to 64 dots of 8×8, for example, the average value of the 64 data is calculated, and it is determined which level of 64 levels the average value is located at. For example, if the level of the average value is at the j-th stage, the portion of the density pattern matrix that is assigned a number less than or equal to j is set as a dot-shaped pixel. .

For example, when using the density pattern matrix shown in FIG. 7(1), the density of the 32nd stage is
It is expressed as a set of dot-shaped pixels as shown in Figure (B).

Returning to FIG. 8, before the image data is applied to the controller 30, it is processed and converted into binary image data representing a pseudo-halftone image using the area modulation method.

Hereinafter, for the sake of concreteness, it is assumed that recording is performed in black and white, and the dot-shaped pixels are black dots.

Now, when there is binary image data obtained as described above, a recorded image obtained by using all of this binary image data in -degree increments is called a 1-size digital image. On the other hand, a digital image with a different size from the same size digital image is called a variable size digital image.

The magnification ratio between the same-size digital image and the variable-size digital image is not necessarily the ratio of the dimensions of the digital image as a recorded image. What determines the mutual magnification of digital images is the ratio of the number of binarized image data forming each digital image.

When the magnification of a variable magnification digital image is an enlargement magnification, some of the binarized binary image data for composing the same magnification digital image is used more than once,
At reduced magnification, only a portion of the binary image data is used to form a scaled digital image.

Here, as an example, let us consider a case where a variable magnification digital image is obtained with a reduction magnification of 1/2, that is, 1/4 in terms of area ratio.

Now, the binary image data for the same-size digital image is divided into M dots, that is, M data, in the main scanning direction, and N data in the sub-scanning direction.
In the case of dots, that is, N data, when these data are developed into MXN, for example, select only the odd-numbered data in both the main scanning direction and the sub-scanning direction while maintaining the arrangement order, and then If you combine them and output them as a write signal, you can use 1/4 of the original binarized data to create a scaled digital image with an area ratio of 1/4, that is, a scale ratio of 1/2. can be obtained.

In this way, data can be regularly thinned out from binary image data for a same-size digital image (in the case of scaling down), or data can be repeatedly output regularly (in the case of scaling up).
By doing so, a variable magnification digital image can be obtained.

However, the scaled digital image obtained in this manner generally has gradation that is different from that of the original, same-sized digital image. That is, when selecting binary image data for a scaled digital image, the black and white ratio in the basic unit of density expression changes.

(Purpose) This invention has been made in view of the above-mentioned circumstances, and its purpose is to maintain gradation by using binary image data given for full-scale digital images. At the same time, you can obtain variable magnification digital images. The purpose of the present invention is to provide a novel digital image scaling method.

(composition) This invention will be explained below.

The digital image scaling method of the present invention uses binary image data for a same-sized digital image, which has already been binarized by the dither method or the density pattern method, as representing a pseudo intermediate image. This is a method to obtain double digital images.

The binary image data is cut out as an mxm mother matrix. That is, the binary image data is M in the main scanning direction.
data, if there are N data in the sub-scanning direction, MX
N binary pixel data are sequentially extracted as an mxm mother matrix. m is an integer of 2 or more.

On the other hand, nxn submatrices are cut out from the mother matrix.

One submatrix is selected for one mother matrix, and by combining the selected submatrices according to the mother matrix arrangement, a scaled digital image that is approximately doubled in size is obtained.

The area magnification is n2/m2.

Note that the size of the mother matrix can be set independently of the sizes of the density pattern matrix and dither matrix.

Also, n is a positive integer that satisfies the conditions n to m,
n (If m, it will be reduced magnification, and if n>m, it will be enlarged magnification.

A detailed description will be given below with reference to the drawings.

Note that the binary image data for the same-size digital image is as follows.

It is assumed that the image is binarized according to the density pattern matrix shown in FIG. 7(1).

As mentioned above, the size of the mother matrix can be set independently of the size of the density pattern matrix, but here, for the sake of simplicity, we will assume that the size of the density pattern matrix and the size of the mother matrix are the same, that is, 8×8. Therefore, it is assumed that m=8, and that the 8×8 data constituting the mother matrix are binarized using the same density pattern matrix.

With reference to FIGS. 5 and 6, from the mother matrix,
The rules for cutting out a submatrix will be explained.

In FIG. 5(1), the solid square indicates an 8×8 mother matrix. This figure shows the case where a 4×4 submatrix is cut out from an 8×8 mother matrix. As is clear from the figure, from one mother matrix, four submatrices Slt r S12 *
S21 m S22 can be cut out.

The 64 pieces of data that make up the mother matrix are shown in Figure 7 (
The images are binarized according to the density pattern matrix of 1). Therefore, if the density expression pattern in the mother matrix is, for example, as shown in FIG. 7 (b), then in FIG. , the density representation pattern of the submatrix S12 is as shown in Q[) in FIG. - Figure 5 ρ shows the case where a 3×3 sub-matrix is cut out from an 8×8 mother matrix. In this case, 8 is not divisible by 3, so in order to cut out the submatrix, the same mother matrix must be used in both the horizontal (main scanning direction) and vertical (sub scanning direction)
The submatrix is cut out by assuming that the submatrix is repeated three times.

Therefore, in this case, the types of submatrices that can be extracted from the same mother matrix are Sll to 5ll.
There are 64 ways of a.

Assuming that the pattern of concentration expression in the conversion matrix is as shown in (b) in FIG. The pattern of expression is as shown in the top 2M (b) of FIG.

Figure 6 shows 12 x 12 from an 8 x 8 matrix.
The submatrix of So e S12 t 5zle
It shows the rules for cutting out S2□.

In this case, the same matrix is arranged in 3x3,
A submatrix can be cut out from this array. Again, if we take the pattern (a) in Figure 7 as an example of the density expression pattern in the matrix, we can use the pattern in Figure 6 (
For example, among the extracted submatrices in 1),
The pattern of the hail density expression of the submatrix S1t is as shown in FIG. 6(1). Therefore, in this case as well, four types of submatrices will be cut out from one transformation matrix.

In this way, from a single transformation matrix, it is not necessary to
Although it is possible to cut out multiple types of submatrices,
In order to construct a variable-magnification digital image, one submatrix is selected from a plurality of types of submatrices cut out from each conversion matrix.

Then, by combining the selected submatrices,
This is binary image data for variable-magnification digital images.

Next, how to select submatrices and how to combine them will be explained.

FIG. 3 shows the case where a 4×4 sub-matrix is cut out from an 8×8 matrix. In Fig. 3 (1), Io+I[o+I[o, ... and α,
Each combination of β, γ, . . . indicates a conversion matrix sequentially extracted from binary image data. From each matrix, there are 4 types of sub-matrices St1+St□.

S21 * S22 can be cut out.

Of the cut out sub-matrices, those that are completely enclosed are selected in the individualization matrix and shown in Figure 3 (I
They are combined as shown in t).

That is, as shown in FIG. 3 (1), in the row of the conversion matrix represented by α, the conversion matrix is 1o+Il
o, I[[o,..., the submatrix becomes Szt s SI2 e 5slt SI2
, and for example, for the sequence of the transformation matrix α
,β.

As γ... changes, the submatrix becomes S1
2゜S22 * 312 t 322 +... and S
12 * S22 is selected repeatedly.

The selected sub-matrices are then combined according to the arrangement order of the original matrix. Therefore, if a scaled digital image is formed using binary image data combined in this way, a scaled digital image with a magnification of 1/2 will be obtained.

If the matrix is 8x8 and the submatrix is 3x3, then with reference to Figure 5 (G),
First, in the first matrix, submatrix 8
11 is selected, and the sub-matrix S12 is selected from the transformation matrix that curves to the right in the main scanning direction, and the sub-matrix 31g is selected from the transformation matrix that curves to the right in the main scanning direction.

Therefore, the combination of submatrices is Szt St
z S13 S'14 Sts S16 S
17 5lit So S12 °°°”°.

821 Szz S23 S24 S25 S26 S
27 S21 S21 S22°°°°°゛S31 S
32... S81 S82... So Su S13... The individual submatrices that are combined in this manner are each cut out one by one from separate conversion matrices, and the arrangement of the submatrix combinations preserves the arrangement of the conversion matrices. The reduction magnification in this case is 3/8.

Similarly, if the matrix is 8 x 8 and the sub-matrix is 12 x 12, referring to Figure 6 (I), from the first mother matrix 1, sub-matrix SB is obtained, and from the mother matrix to the right of it, sub-matrix SB is obtained. submatrix 3
1z, and so on, and according to the arrangement of the mother matrix, So S12 So S12 Sll
S12...S21 S22 S21 S22
S21 S22°°0′°.

They may be combined as follows: So Stz S1t 812. In this way, an enlarged variable magnification digital image with a magnification of 12/8 is obtained.

In the example shown in FIG. 4, the mother matrix is 4×4 and the submatrix is 1×1. From each mother matrix, the data marked with a circle is cut out as a submatrix, selected, and the mother matrix are combined according to the sequence of In this way, the magnification is 1/4 (area magnification IA6
) can be obtained as a reduced scale digital image.

FIG. 1 is a block diagram showing an example of the controller portion of the apparatus shown in FIG. 8, excluding the system control system. To explain using the example explained in conjunction with FIG. 1o), (α-11
o)-(α-I[o)..., (β-1o), (β-1
1o)... are cut out.

The extracted mother matrix is then stored in the submatrix extraction buffer 14, where necessary portions (FIG. 3 (I)
) is cut out as a 4×4 submatrix, and the cutout submatrix is developed in the bitmap memory 16 in an arrangement as shown in Figure 3 (]I). The contents of this bitmap memory 28 are outputted to a semiconductor laser serving as a light source via an interface 28.

Therefore, if an image is recorded using the write signal processed in this way, the magnification ratio will be 1/2 (
A reduced scale digital image with an area magnification of 1/4) is obtained.

FIG. 2 is a block diagram showing another example of the portion of the controller other than the system control system.

The binary image data is applied to counters 20, 22 and data writing device 24.

Counters 20 and 22 are counters for counting the number of input data.

If the mother matrix is mxm and the submatrix is nxn, the counter 20 is a counter that counts m or nat-counts, is on until n pieces of data are input, is off until the next m pieces of data are input, and then turns off until the next m pieces of data are input. In response to n inputs, it is turned on, and so on, so that n and m outputs are alternately turned on and off. Counter 22&ζn×
(Number of data M for one line), that is, nM pieces, or m
It is a counter that counts x (number of data M for one line), that is, mm, and outputs an on state until nM data are input, and an off state until the next mm data is input.

The data writing device 24 writes input data to the bitmap memory 26 only while both counters 20 and 22 are on.

The contents of the bitmap memory 26 are then output via the interface 18. Note that the counter 20 is returned to its initial state when it receives input for one line.

When m=49=1, the data marked with circles in FIG. 4 can be selected, combined, and output, and a reduced scale digital image with a reduction ratio of 1/4 is shown as compared to the same scale digital image. This is the 9th to 9th figure, and the figure (1) is (9×9) for the original size, and (3×3) for the latter. Note that the binary image data is
The image was binarized using the density pattern matrix shown in Figure (I). All variable-magnification digital images preserve the gradation of the 1-scale digital image.

(Effects) As described above, according to the present invention, it is possible to change the magnification of a digital image while maintaining gradation like a new digital image.

As mentioned earlier, the size of the mother matrix does not have to be the same size as the density pattern matrix, but if both are made the same size, the shape and pitch of the halftone dot pattern can be changed even in a variable-magnification digital image. , it can be made the same as a full-size digital image.

Note that although binary image data has been explained using an example in which it is binarized using a density pattern matrix, the same applies to binary image data that is binarized using a dither matrix.
It goes without saying that the present invention can be applied in the same manner as described above.

[Brief explanation of the drawing]

FIG. 1 is a boot 2 diagram showing an example of a circuit for outputting data for variable-magnification digital images from binary image data;
FIG. 2 is a block diagram showing another example of a circuit for outputting data for a scaled digital image from binary image data;
FIGS. 3 to 7 are diagrams for explaining the present invention, and FIG. 8 is a view of the ice. 32... Semiconductor laser, 34... Collimator lens, 36... Rotating polygon mirror, 38... fθ lens,
40...Photoreceptor. Cleaning of the drawings (no changes to 83) Figure 7 Figure Z Figure 3 (L) (■n Figure 4 Figure 5 (I) (n) (m) Figure C (I) Figure 7 Diagram (I) l'II) Relationship with the case of the person making the amendment Name of patent applicant (674) Ricoh Co., Ltd. 4, Agent
Address: 4-5-4 Kyodo, Setagaya-ku, Tokyo I will submit the full specification and all 8 drawings as shown in the attached sheet.

Claims (1)

[Claims] As an expression of a pseudo-halftone image, a scaled digital image having a size different from that of the same-size digital image created from binary image data that has already been binarized. , the binary image data is cut out as an m×m (m≧2) mother matrix, and from this mother matrix, n×n (n≧1 and n≠m) is obtained.
By cutting out the submatrices of and a digital image scaling method, characterized in that a scaling digital image is obtained.