JPS58141076A - Picture expanding method - Google Patents

Abstract

PURPOSE:To perform excellent expansion even at an oblique line and a right- angled part, by providing plural density distributions of original picture elements, selecting said density distributions according to the pattern of an original picture near picture elements to be converted, and finding the mean density of the original picture at the position of the picture elements to be converted. CONSTITUTION:An area is divided by combinations shown by X and Y. When there is the center of picture elements to be converted in, for example, a division area 8 in the figure (X), a decision on the expansion of a right-angled part is performed because original picture elements A, B, and C and surrounding picture elements L and I are black picture elements in a figure (a) to obtain the density of the picture element to be converted from IR=ID=IO. In figures (b), (c), and (d), IR=ID.(IA+IB+IC)+IA.IB.IC=1. In figures (e), (f), (g), and (h), respective logical equations are used. By using those equations, whether the right-angled part or ablique line part is to be expanded is judged and the density IR of the picture element to be converted is found by only easy judgement.

Description

DETAILED DESCRIPTION OF THE INVENTION According to the present invention, when an original image is projected onto a converted image plane, the average density of the original image located within one pixel of the converted image plane is calculated based on the average density of the original image located within one pixel of the converted image plane. This invention relates to an image enlargement method for determining density and obtaining a converted image.

For example, when allocating an image to a specific area in an intelligent copy with an editing function, it becomes necessary to enlarge the image at an arbitrary magnification. In addition, in image transmission systems, due to differences in scanning line density between input and output devices,
The size of the image being handled may vary from device to device, and to compensate for this it may be necessary to enlarge the image.

Conventionally, methods such as the SPC method, logical sum method, 9-division method, and projection method have been known as methods for enlarging images in such cases, and among these methods, the projection method is better than other methods. It is said that high image quality can be obtained. However, the conventional projection method has the drawback that the smoothness of the hatched area is lost when the magnification is large. The reason for this is that the density distribution of the pixels of the original image is assumed to be uniform (that is, the density is 1) and to be a square pattern having the same shape as the pixels.

The present invention has been made in view of this point, and includes preparing a plurality of types of density distributions of pixels of the original image, and adjusting the density of the pixels in the vicinity of one pixel on the converted image surface according to the pattern of the original image. By selecting one of the distributions, using the density distribution of the pixel to find the average density of the original image located within the -pixel of the converted image plane, and enlarging the image, the shaded area can be smoothed. This method realizes an image enlargement method that allows a converted image to be obtained without losing image quality.

Hereinafter, the present invention will be described in detail with reference to the drawings.

When pixels of the original image (hereinafter referred to as original pixels) have a density, various functions can be considered to represent the density distribution, but here we will consider a function that satisfies the following conditions. As an example, the influence of the density distribution on the image quality of the converted image will be described (note that the vertical and horizontal lengths of the original pixel are both 1).

(a) For one pixel of the original image, if its center is the origin, the horizontal axis is X, and the vertical axis is Y, then the point (X, Y
) The following equation holds true for the function f (X, Y) that indicates the concentration at

f (X, Y)d Xd Y=1-(1)(b)
At the center of the original pixel, the density is 1.(C) Pixels with density (hereinafter referred to as black pixels) are A, B, C, etc. in FIG.

In the case of adjacent pixels as in D, the density g (x, y) at the point (x, y) is the sum of the individual densities of each adjacent pixel, and as shown in Figure 1, (x, y) Taking the axis, it is expressed by the following formula.

g(x, V) = f(x-1/2.V-
1/2)+f (x-1/2.V +1/2)+f
(x +1 /2.1-1/2>+f (x +1
/2. V +1/2>...(3) This o(x, y) must satisfy the following formula.

0(X, Vl=1...(4) However, -1/2≦x, y≦1/2 As an example of a function that satisfies the above three conditions, the following three equations can be cited.

f s (X, Y) = rect(X) rect
(Y) ・(5) f2 (X, Y) = ... (6) fs (X, Y) = ... (7) By the way, the pixel of the converted image (hereinafter referred to as a converted pixel) R
and original pixels A, B, C, and D are in the positional relationship as shown in FIG. 2, and their centers are Ro (x, y),
Ao (-1/2.1/2>, So (-1/2.
-1/2>, Co (1/2°-1/2), Do
(1/2.1/2), when the magnification is sufficiently large, the average density of the original pixels A, B, C, and D within the converted pixel R is expressed by the following equation.

1” (x, y) = IA-f (x +1/2.V-'I/2>+ rB
-f (x +1/2.y +1/2)+■C・f
(x-1/2.V +1/2>+ID-f (x-1
/2. V-1/2)...(8) However, IA, IB, IC, and ID are the original pixel A, respectively.
B, C, D lr! 1 for black pixels;
In other cases, it is O.

In the projection method, the density of the converted pixel is set to 1 when F (x, y)≧1/2, and O when F (x, y)<1/2. Therefore, by determining the condition that F (x, y)≧12, a logical operation formula (conversion pixel calculation formula) for each density distribution can be determined. This logical operation formula can be determined for each divided area obtained by further dividing a square area connecting the centers AOBo, Go, and Do of the original pixels with straight lines. The aforementioned fl(X, Y>.

f 2 (X, Y), f s (X,, Y
) Nio<f Ru territory R division is shown in Figures 3 to 5,
In the logical sieve table, * indicates logical product, and + indicates logical sum.

Therefore, by knowing in which divided region the center (x, y) of the converted pixel R is located, the density of the converted pixel [IR
can be found.

As is clear from FIGS. 3 to 5, various area divisions can be obtained by changing the density distribution of the image.

FIGS. 6 and 7 show transformed images obtained by the above-described area division. From these figures, fz(X,Y)
, fs (X, Y), the diagonal line part can be enlarged smoothly, but the right angle part can be enlarged smoothly using
It can be seen that, unlike area division using (X, Y), roundness occurs. For this reason, in the present invention, area division is used differently for the right angle portion and the diagonal line portion.

In order to easily configure a circuit that determines which divided region a point (X, y) belongs to, it is preferable to divide the region using straight lines. Figure 8 (
These are the combinations shown in a) and (b) (the same parts as in FIGS. 3 to 5 are given the same reference numerals). An example of how the logical operation expressions are used when the center of the converted pixel is located in the divided region 8, for example, in FIG. 9(b) will be described with reference to FIG. First, in the case of FIG. 9(a), to the original pixel, B, C
In addition, since the surrounding original pixel I is also a black pixel, it can be interpreted as an enlargement of the right angle part, so IR-ID=
Let it be Io.

However, for <a), if one or both of the original pixels I and I are not black pixels, that is, (b).

(0), (d>), calculate IR from the following equation.

IR-ID・(IA+IB+IC) +IA-IB-IC=1 Also, when only the original pixel is a black pixel, as in (e), IR=IO-0, and further, the original pixel is set as IR-ID.

If one or both of I and the original pixel are black pixels, that is, (
f ), (a), and (h) are calculated using the following formula.

I R= I D・(IA+IB+ IC) 10 IA-I
B・rc=.

Therefore, if the condition in FIG. 9 is expressed by one logical operation expression, it will be as follows.

IR=ID・(rA+IB+IO+IL-II)+IA
・IB-rc・(IL+rr) Similarly, original pixels E, F, around original pixels A, B, C, D,
When G, H, r, J, and K are taken, the logical operation formula becomes as shown in the following table.

By using this, it is possible to simultaneously determine whether an area is a right-angled area or an oblique area using a single calculation formula and a single area division. Furthermore, since the density IR of the converted pixel can be determined by only simple logical judgment, the complexity of calculation processing, which is a problem in the projection method, can be avoided.

As described above, according to the present invention, it is possible to realize an image enlarging method that can obtain a good converted image in both the diagonal line portion and the right angle portion.

[Brief explanation of the drawing]

Figure 1 is an illustration of original pixel density, Figure 2 is an illustration of the positional relationship between original pixels and converted pixels, Figures 3 to 5 are illustrations of area division, and Figures 6 and 7 are illustrations of area division. An explanatory diagram of an enlarged image, Fig. 8 is an explanatory diagram showing an example of area division in the present invention, Fig. 9 is an explanatory diagram of a logical operation formula in the present invention, and Fig. 10 is an explanatory diagram of an original pixel in the present invention. It is an explanatory diagram of a position. A to L...Original pixel R...Converted pixel AO to DO...Original pixels A to D Main patent applicant Roku Konishi Photo Industry Co., Ltd. Agent Patent attorney Ijima Fuji Yasuo 1 Figure 1 2 Figure 3 Figure Atsushi 5 Figure 6 (CI) Hara--I Kashik (b) f, Jiyo himself (C) f, f3 + 2 to Saikoji (b) Abusive figure 9 (b) (c) (e) (f) (CJ) (h)

Claims (3)

[Claims] (1) From the average density of the original image located within one pixel of the converted image plane when the original image is projected onto the converted image plane,
In an image enlarging method for obtaining a converted image by determining the density of the pixel on the converted image plane, a plurality of types of density distributions of the pixels of the original image are prepared, and the density distribution of the pixel in the vicinity of the one pixel on the converted image plane is Select one of the density distributions of the pixels according to the pattern of the original image, use the density distribution of the pixels to find the average density of the original image located within the -pixel of the converted image surface, and calculate the density distribution of the image. An image enlargement method characterized by performing enlargement. (2) By connecting the centers of each pixel of the original image,
A plurality of square areas are set on the converted image plane, and the area is equally divided into four square areas, and eight triangular areas are formed by dividing lines connecting the midpoints of each side of the area. A logical operation formula for determining the density of the one pixel of the converted image whose center is within the divided area is prepared in the divided WA area, and the right-angled portion is expanded. In the above, the pixel density of the converted image is calculated from the logical calculation formula based on the first area division, and in the enlargement of the diagonal area, the pixel density of the converted image is calculated from the logical calculation formula based on the second area division. 2. The image enlarging method according to claim 1, wherein (3) The first area division is made to serve as the second area division, and the logical operation expression based on the first area division is made to also serve as the logical operation expression based on the second area division. An image enlarging method according to claim 2, characterized in that: