JP2713764B2 - Halftone image enlargement method - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [Summary] The present invention relates to a halftone image enlarging method for enlarging a halftone image, with the object of emphasizing the edges of original pixels and allowing a density change to be expressed smoothly, in the vicinity of a converted pixel. Is sequentially input, and a temporary value of a virtual pixel is determined by weighting from the input density data and the position coordinates of the converted pixel, and the density of the original pixels around the converted pixel is determined. A correction value is obtained in accordance with the change state, and a virtual pixel value is determined by adding or subtracting the provisional value of the virtual pixel and the correction value in accordance with the density change state of surrounding original pixels, and a virtual pixel value is determined from the determined virtual pixel value. The conversion pixel value is obtained by a distance weighting operation.

The present invention relates to a halftone image enlarging method for enlarging a halftone image.

[Prior Art] FIG. 15 is an explanatory diagram of a conventional pixel density conversion method. In the figure, points A, B, C, and D are original pixels, and points R are converted pixels. A perpendicular line is drawn from the converted pixel point R to each point of the quadrangle formed by the original pixels A to D, and the intersections with each side AB, BC, CD, DA are M, N, O, and P, respectively. These points are indicated by ● as virtual pixel points.

The densities (pixel values) of these virtual pixel points M to P are determined by weighting the ratio of the distance from the two adjacent original pixels. The densities of the original pixels A to D are respectively Ia, Ib, Ic, Id
Then, the densities Im, In, Io, and Ip of the virtual pixels M to P are expressed by the following equations, respectively.

Im = AM × Ib + MB × Ia (1) In = BN × Ic + NC × Ib (2) Io = CO × Id + OD × Ic (3) Ip = DP × Ia + PA × Id (4) where AM, MB, etc. are line segments And these distances are normalized with the length of each side of the quadrilateral ABCD as 1, and expressed as a ratio to 1. After the density of each virtual pixel point is determined by the above equation, the density of the converted pixel point R is obtained by averaging the densities of these virtual pixel points. The concentration Ir at the point R is expressed by the following equation.

Ir = (Im + In + Io + Ip) / 4 (5) If the density is obtained by the equation (5), the density change can be made smooth even for the enlarged image.

[Problems to be Solved by the Invention] However, according to the above-described method, since the density of the converted pixel is determined by using the weighting operation and the averaging operation, the outline of the converted pixel is blurred without being clear. There is a defect. For example, assume that there is an original image as shown in FIG. In the figure, the vertical axis represents the density level, and the horizontal axis represents the position coordinates. The converted image after the density conversion of the original image is as shown in FIG. The converted image is on a line connecting the densities of the respective pixels of the original image as is apparent from the figure, and this gives a blurred feeling to the eyes. Therefore, it is preferable to have the characteristic shown in FIG. 3C than to such a converted image because the density change is smoother.

The present invention has been made in view of such a problem, and an object of the present invention is to provide a halftone image enlarging method capable of emphasizing edges of an original image and allowing a density change to be smoothly expressed. I have.

[Means for Solving the Problems] FIG. 1 is a flowchart showing the principle of the method of the present invention. According to the present invention, a virtual pixel is provided on each side of a quadrilateral formed by surrounding original pixels based on the position of the converted pixel projected on the plane on which the original pixels are arranged, and the converted pixel value is determined by weighting the distance. In the halftone image enlarging method, the density data of a predetermined original pixel group located in the vicinity of the conversion pixel is sequentially input (step 1), and the provisional value of the virtual pixel is determined from the input density data and the position coordinates of the conversion pixel. Determined by a weighting operation (step 2), a correction value is obtained in accordance with the density change state of the original pixels surrounding the converted pixel, and a provisional value of the virtual pixel and the correction value are determined in accordance with the density change state of the surrounding original pixels. Are calculated by adding and subtracting (step 3) to determine a virtual pixel value, and a converted pixel value is obtained from the determined virtual pixel value by a distance weighting operation (step 4).

[Operation] A correction value is obtained in accordance with the density change state of the original pixels surrounding the converted pixel, and the provisional value of the virtual pixel and the correction value are subjected to an addition / subtraction operation in accordance with the density change state of the surrounding original pixels. The value is determined, and the pixel value of the converted pixel is determined by a distance weighting operation based on the determined virtual pixel value. Therefore,
According to the present invention, it is possible to obtain a density-converted image having characteristics as shown in FIG. 16 (c), and it is possible to emphasize the edges of the original image and express the density change smoothly.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings. The case where the virtual pixel N is determined from the four original pixels F, B, C, and I will be described with reference to FIG. In the figure, the vertical axis represents the density level, and the horizontal axis represents the position coordinates. In the conventional method, a value on a thick solid line in the figure is obtained by weighting the distance (see FIG. 16 (b)). In the method of the present invention, the original pixels F, B, C,
The value determined by the conventional method from the density change state of the original I pixel is corrected. That is, first, a provisional value is obtained by weighting the distance from the original pixels B and C on both sides of the virtual pixel N. Provisional value of virtual pixel N
In is given by In = BN × Ic × NC × Ib (6). Ib and Ic are the densities of the original pixels B and C, respectively.

Next, the density change amounts a, b, and c are obtained by the following equations.

 a = If−Ib (7) b = Ib−Ic (8) c = Ic−Ii (9) Further, values S1 and S2 defined by the following equations are obtained.

S1 = | b−a | (10) S2 = | c−a | (11) A point Q that is internally divided on the line segment BC at a ratio of S2: S1 is obtained. At this point Q, it is determined whether a correction value to be obtained later is added or subtracted. By determining the point Q from the ratio of the difference in the density change amount, a smooth density change can be obtained. However, the point Q may be the middle point of the line segment BC.

Here, the correction value for determining how much to change from the thick solid line is determined according to the degree of edge enhancement and S1, S2. For example, it is determined by the following formula: (size of ba) × (degree of edge enhancement) × 1 / (value determined by image density width) (12)

Next, it is determined whether the virtual pixel N is between BQ and QC.

The original pixels F, B, C, I correspond to (a), (b), (c),
It is determined which of the four types of density change state of the original pixel shown in (d) applies, and the magnitude relationship of the density change amounts a, b, and c corresponding to the change state is determined.

As a result of the determination, as shown in FIG. 4, the virtual pixel N is between BQ and if a> b, the correction value is added to the temporary value, and if a <b, the correction value is subtracted from the temporary value. Also, the virtual pixel N is between QC,
If c> b, the correction value is subtracted from the temporary value, and if c <b, the correction value is added to the temporary value.

By the above processing, the value (density) of the virtual pixel N can be obtained. Next, specific numbers will be described. It is assumed that the densities of the original pixels are F = 50, B = 14, C = 50, and I = 22.

First, the provisional value of the virtual pixel N is obtained by equation (6).

 Assuming that BN: NC = 1: 4, In = 14 × 4/5 + (50 × 1/5) = 22.2.

Next, the density change amounts a to c are a = 14−50 = −36 b = 50−14 = 36 c = 22−50 = −28 according to the equations (7) to (9). Further, S1 and S2 are S1 = | b−a | = 72 and S2 = | c−b | = 64 according to the equations (10) and (11).

A point Q which is internally divided on the line segment BC at a ratio of S2: S1 is obtained.

Since S2: S1 = 64: 72, if BC is internally divided at this ratio, the point Q comes at the position in FIG.

Next, a correction value is obtained by equation (12). The edge enhancement degree is, for example, 5, and the value determined by the image density width is, for example, 100.
Then, since ba is 72, 72 × 5/100 = 3.6. The provisional value of the virtual pixel N is 2
2.2.

Since the virtual pixel N is between the BQs, (provisional value−correction value) is obtained from the expression (b) in FIG. 4, and In = 22.2−3.6 = 18.6. Here, round to 1
9 is assumed.

Assume that the densities of the four virtual pixels M, N, O, P (see FIG. 15) are, for example, Im = 30, Io = 40, and Ip = 24. Further, it is assumed that MR: NR: OR: PR = 1: 2: 3: 4. The density Ir of the fluctuating pixel R is obtained by calculating a converted pixel value from the determined virtual pixel value by weighting the distance. The formula is as follows:

Ir = 30 × (4/10) + 19 × (3/10) + 40 × (2/10) + 24 × (1/10) = 28 (rounding 28.1) Fig. 5 shows the pixel value of virtual pixel N 6 is a flowchart showing a sequence for obtaining the following. The flowchart shown in the figure,
The above sequence is summarized in a flow. First, the provisional value of the virtual pixel N is determined by weighting the distance from the original pixels B and C (S1), and the calculations represented by the equations (7) to (11) are performed (S2). When a to c and S1, S2 are obtained by the calculation, the coordinates of a point Q which is internally divided by S2: S1 between the original pixels BC are obtained (S3), and correction is performed according to the degree of edge enhancement and the size of S1, S2. Determine the value (S4).

Next, it is determined whether the virtual pixel N is between BQ or QC (S5). If it exists between BQs, it is checked whether a> b (S6). If so, the pixel value is determined by In = temporary value + correction value (S7). Otherwise, the pixel value is determined by In = provisional value−correction value (S8). If the virtual pixel exists between the QCs, it is checked whether c> b (S9). If so, the pixel value is determined by In = provisional value−correction value (S10). Otherwise, the pixel value is determined by In = provisional value + correction value (S11).

FIG. 6 is a diagram showing an example of a locus of density. f1 indicates the concentration locus of the conventional method, and f2 indicates the concentration locus of the method of the present invention. The vertical axis is the density level, and the horizontal axis is the position coordinates. The method of the present invention
It can be seen that the shape is close to that shown in FIG.

FIG. 7 is a principle block diagram of an apparatus for implementing the method of the present invention. Reference numeral 1 denotes original pixel input means for inputting original pixel values, for example, the pixel values of A, B, C, and D in FIG. 15 are input. The output of the original pixel value input means 1 enters a plurality of virtual pixel value determination means 2, and the virtual pixel value determination means 2 determines a virtual pixel value. Assuming that the virtual pixel values are, for example, M, N, O, and P in FIG. 15, the pixel values of these virtual pixels are obtained by the above-described sequence. The obtained virtual pixel value enters the subsequent conversion pixel determination means 3. The coordinate value of the converted pixel is input to the converted pixel determining means 3, and the converted pixel determining means 3 obtains the converted pixel value by a distance weighting operation.

FIG. 8 is a diagram showing a specific configuration example of the virtual pixel value determining means 2. The virtual pixel provisional value determining means 21 receives the position coordinates of the transformed pixel with respect to the original pixel value, and calculates the provisional value by equation (6). The correction value determining means 22 calculates the correction value for the original pixel by inputting the edge emphasis degree according to the equation (12). Receiving the original pixel value, the original pixel state determining means 23 obtains the density change state of the surrounding original pixels by the equations (7) to (11). The internally dividing point determining means 24 obtains a point internally divided in S2: S1. The virtual pixel value determining means 25 receives the output of each of the elements 21 to 24 and determines a virtual pixel value.

FIG. 9 is a block diagram showing an embodiment of an apparatus for implementing the method of the present invention. This circuit is assumed to be a circuit for calculating the pixel values of the converted pixels R having the positional relationship shown in FIG. 9, the same components as those in FIGS. 7 and 8 are denoted by the same reference numerals. The position coordinate calculation circuit 30 obtains the position coordinates of the conversion pixel R by inputting the conversion magnification, and gives it to the weight calculation circuit 2 as the conversion pixel value determination means 3. Next, the weighting calculation circuit 1 as the virtual pixel provisional value determination means 25 receives the output of the position coordinate calculation circuit 30 and the pixel values Ia to Id of the surrounding A, B, C, and D, and outputs four virtual pixels. The provisional values of M, N, O, and P are obtained and given to the addition / subtraction circuit 25 as the virtual pixel value determination means 25.

Next, the subtraction circuit 31 inputs the pixel values Ia to Ii of the original pixels A to I, calculates the density change amounts represented by the equations (7) to (11), and stores the density change amount in the register 32 which is the a to c register 1. , S1 and S2 are stored in the register 33 which is the register 2. The comparison circuit 1 as the original pixel state determination means 23 receives the density change amounts a to c, determines the density change state of the original pixel, and sends the result to the addition / subtraction circuit 25.

Further, the coordinate calculation circuit 24, which is the internally dividing point determining means 24, calculates S1, S
The coordinates of the internal dividing point are obtained based on the number 2 and the result is given to the comparison circuit 34 which is the comparison circuit 2. The comparison circuit 34 compares the position coordinates of the converted pixel with the coordinates of the internal dividing point, and compares the result with the addition / subtraction circuit.
Send to 25. On the other hand, the correction value determining circuit 22, which is the correction value determining means 22, receives the edge enhancement degree data and the S1 and S2 data and determines the correction value based on the equation (12). Then, the result is sent to the addition / subtraction circuit 25.

The addition / subtraction circuit 25 determines whether the correction value is added to or subtracted from the provisional value of the virtual pixel based on signals sent from the comparison circuits 1 and 2, and determines the pixel value of the virtual pixel. Next, the weighting operation circuit 2 performs a weighting operation according to the output from the addition / subtraction circuit 25 to determine the pixel value of the virtual pixel.

In the above description, the case where the correction value is determined based on the expression (12) is taken as an example, but the present invention is not limited to this. For example, as shown in FIGS. 11 and 12, the same effect can be obtained by making the distance proportional to the distance from the original pixel. For example, when the correction values shown in FIG. 11 are used, the density characteristics based on the correction values are as shown in FIG. f1 indicates the conventional method, and f2 indicates the method of the present invention. Also in this method, the density characteristic is similar to the shape shown in FIG.
Good results are obtained. That is, the edge can be emphasized and the density change can be made smooth. FIG. 14 shows density characteristics when the correction values shown in FIG. 12 are used. The effect is the same as in FIG.

[Effects of the Invention] As described above in detail, according to the present invention, a correction value is obtained according to the density change state of the original pixels around the converted pixel, and the correction value is calculated according to the density change state of the surrounding original pixels. A virtual pixel value is determined by adding / subtracting the provisional value of the pixel and the correction value,
The converted pixel value is obtained by weighting the distance between these virtual pixel values, whereby a converted pixel in which the edge is emphasized and the density change is smooth can be obtained.

[Brief description of the drawings]

FIG. 1 is a diagram showing the principle of the method of the present invention, FIG. 2 is a diagram showing how to determine a virtual pixel N, FIG. 3 is a diagram showing a pattern of a density change state, and FIG. FIG. 5 is a diagram showing a determination method, FIG. 5 is a diagram showing a sequence for obtaining a pixel value of a virtual pixel N, FIG. 6 is a diagram showing an example of a locus of density, FIG. 7 is a principle block diagram of an apparatus for implementing the method of the present invention FIG. 8 is a diagram showing a specific configuration example of a virtual pixel value determining means, FIG. 9 is a configuration block diagram showing an embodiment of an apparatus for implementing the method of the present invention, and FIG. FIGS. 11 and 12 are diagrams showing another example of the correction value between the original pixels BC. FIGS. 13 and 14 are diagrams showing density characteristics according to another correction method of the method of the present invention. FIG. 15 is an explanatory diagram of a conventional pixel density conversion method, and FIG. 16 is a diagram showing a conventional density change and a smooth density change. 7 and 8, 1 is an original pixel input means, 2 is a virtual pixel value determining means, 3 is a converted pixel value determining means, 21 is a virtual pixel temporary value determining means, 22 is a correction value determining means, 23 Is an original pixel state determining means, 24 is an internally dividing point determining means, and 25 is a virtual pixel value determining means.

Claims (1)

(57) [Claims] A virtual pixel is provided on each side of a quadrilateral formed by surrounding original pixels based on the position of the converted pixel projected on a plane on which the original pixels are arranged, and a converted pixel value is calculated by a distance weighting operation. In the halftone image enlarging method for determining the halftone image, the density data of a predetermined original pixel group located in the vicinity of the conversion pixel is sequentially input (step 1). The value is determined by a weighting operation (step 2), a correction value is determined according to the density change state of the original pixels surrounding the converted pixel, and the provisional value of the virtual pixel and the correction value are determined according to the density change state of the surrounding original pixels. A virtual pixel value is determined by adding or subtracting the correction value (step 3), and a converted pixel value is obtained from the determined virtual pixel value by a distance weighting operation (step 4). Toning image enlargement method .