JPS622367A - Converting system for picture element number - Google Patents

Abstract

PURPOSE:To current picture element numbers by which an image of high quality which scarcely causes collapse and omission, by providing a processor for converting the number of picture elements by using dot data of one raster position of an image as one unit, and an image developing processor, etc. CONSTITUTION:When image data which has been encoded as an original image is inputted to an image developing processor 10, the processor 10 converts it to dot data, and accumulates it in a temporary storage device 12. Linear density information functioning as control information, and the number of vertical and horizontal dots, etc. are put to the hed of this encoded data, and given from the processor 10 to a picture element number converting processor 13. Dot data (dot data arranged in one horizontal line of an image) of one raster portion of an image are read out as one unit to the processor 13 from the device 12, and the picture element number conversion is executed. In this case, the processor 13 executes the picture element number conversion, based on the linear density information of the original image which has been given from the processor 10. In this way, the picture element number conversion of high quality, which scarcely causes collapse and omission can be executed.

Description

[Detailed Description of the Invention] [Field of Application of the Invention] The present invention relates to a pixel number conversion method for images such as characters and pictures [Background of the Invention] Devices that handle images include display devices, printers,
Many devices such as facsimiles are used. With these devices, images are produced at various linear densities, e.g. 120 D P
I (Dot Par Irt, ch), 160D
It is displayed as PI, 240 D PI, etc. When displaying an original image represented by a certain linear density on a device having a different linear density, it is necessary to convert the linear density, that is, convert the number of pixels. For example, if an image represented by rays of 52×52@ pixels is to be displayed using 24×24 pixels in total, a power to convert the number of pixels is required.

Conventionally, known conversion methods include the nearest-neighbor note (spc method), the logical sum method, the 9-division method, the projection method, and the distance inverse proportionality method. (19
82'), pp. 72-81, "High-speed uniformity density conversion method based on projection method" also introduces various methods.

When converting the number of pixels, it is desired that the converted image be closer to the original image.When reducing the number of pixels (reducing the number of pixels), for example, with nearest neighbor annotation, although there is no image collapse phenomenon, there are pixels that are not reflected in the converted image, and there may be gaps. This tends to occur, and while the disjunctive method and the 9-division method are infallible, they do have advantages and disadvantages, such as conspicuous collapse.

[Purpose of the invention]

SUMMARY OF THE INVENTION An object of the present invention is to provide a pixel number conversion method that saves time by obtaining high-quality converted images with less distortion and omission.

[Summary of the invention]

In order to convert an original image into a reduced converted image, the present invention has the following steps:
Letting the number of pixels of the original image be r, and the number of pixels of the converted image m, ΔX −n = Δram ΔX: Virtual distance between pixels of the original image ΔX: Virtual distance between both images determined by the virtual distance between pixels of the transformed image Based on the target pixel distance, all pixels of the original image are the pixels that determine the closest pixel of the converted image, and furthermore, one pixel of the converted image is determined by two or more pixels of the original image. When , these pixels are logically summed to form a pixel of the converted image.

[Embodiments of the invention]

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

FIG. 1 shows an embodiment of the present invention, and shows a system for converting the number of pixels of an input image and displaying the converted image on a display device.

Image data encoded as an original image is input to the image development processor 10. Image expansion processor 10 converts encoded image data into tod (pixel) data and stores it in temporary storage device 12 . At the beginning of the encoded image data, line density information (for example, 24
0 D P I), the number of vertical and horizontal dots, etc., and this control information is given from the image development processor 10 to the pixel number conversion processor 15. The pixel number conversion processor 15 reads dot data for one raster of the image (dot data arranged in rows and rows of the image) as one unit from the temporary storage device 12, and performs pixel number conversion.

The pixel number conversion processor 15 holds in advance the linear density information to be converted, and performs pixel number conversion based on the linear density information of the original image given from the image development processor 10. The storage device 14 is a working storage device used by the pixel number conversion processor 15. The dot data whose number of pixels has been converted is written into the screen buffer 15 and displayed on the display device 16. Further, the information may be printed out using a printer (not shown) or may be transferred by facsimile or the like.

Next, before explaining the specific system of the present invention, the principle of the present invention will be explained. Figure 2 shows the original image of 16 dots (pixels) shown by ■ to O, and the converted image of 15 dots shown by ■ to ■.
An example of conversion at a magnification of 3716 is shown. Here, for the sake of clarity, n indicates a one-dimensional dot array, but normally it is a two-dimensional dot array. During the conversion, the 16-dot original image and the 15-dot converted image are virtually made to have the same size, regardless of the actual displayed size. That is, 16-1X=1s-Δx Here, ΔX is the distance between virtual dots in the original image, and Δ
2 is the virtual inter-dot distance of the converted image. The correspondence between the dots in both images is determined by this ΔX and ΔX. Each dot in the original image is taken as the dot that determines the closest dot in the transformed image. For example, a dot ■ is determined to be a dot ■, a dot ■ is determined to be a dot ■, and a dot ■ and ■ are determined to be a dot ◎. Looking at dots ■ and ■, dot ■ is closer to dot ◎ when comparing dots ■ and ◎, and dot ■ is closer to dot ◎ when comparing dots ◎ and ■, so dot ■ is determined by dots ■ and ■. A dot in the transformed image determined by two or more dots in the original image is the logical sum of the dots in the original image. For example, dot◎ is dot■
It is the logical sum of and ■. The right side of FIG. 2 shows which dots in the original image determine each dot (1) to [phase] of the converted image.

According to this, all dots in the original image are reflected in the transformed image, and each dot in the original image is used to determine a unique dot in the transformed image, and never in more than one transformed image. Not used to determine dots.

Therefore, a high quality converted image with less distortion and omissions can be obtained.

FIG. 3 shows a conversion processing flow in the pixel number conversion processor 15 of FIG. 1, and includes processing flows for not only reduction but also enlargement and 1x.

FIGS. 4 and 3 illustrate in more detail the steps shown in FIG. 3 as CMP F ST and CMPSND. FIG. 6 shows a specific example to facilitate understanding of the processing flow shown in FIGS.

In FIG. 6, the first rask of the dot data of the original image in the temporary storage device 12 is read out to the conversion processor 15 and processed in units of rows (in units of rask). In Figure 6, the ``EO person'' part of the ``OA survey form'' is extracted and illustrated in a large size.

Each dot in the original image is labeled Pa, R...Pi from the left end, and each dot in the converted image is labeled Qa from the left end.

Ql...Qi is marked. The distance between each dot ΔX is that of the original image, and ΔX is that of the transformed image. L and L are distances from the left edge of the original image and the converted image, respectively. f3A will be described later.

Now, in FIG. 3, ΔX>ΔX is first checked (step 21). If ΔX is large, it is an enlargement process. The enlargement process will be described later. Next, if ΔX>ΔX,
j) (=Δ2 is checked (step 22). If they are equal, there is no need for pixel-to-number conversion, and all image data in the temporary storage device 12 is transferred to the screen buffer 15 and the process ends (step 25). If it is smaller than ΔX, the process proceeds to step 24, where a reduction process is performed.Step 24', designated CMPFST, is shown in detail in FIG.

In FIG. 4, the values shown in steps 240-242 are each initialized. In step 243, it is determined whether the difference between t and L is smaller than Δx/2. At the 0th dot, that is, Pa and Qo, the difference between t and L is 0, and the process proceeds to step 244K, where N is set to 1 and 5o=1 (step 245). Sm represents the number of dots in the original image obtained by performing a logical sum every time Qj changes, and is stored in the storage device 14 of FIG.

f3A is also used as <SR, which will be described later.

Next, ΔX is added to L (step 246). and P
i is ORed with Q7 (step 247) o Q
The initial value of / is 0. Now Pi, Qi is Pa
, Qa, and the 0th dot of the original image is the D-th dot of the converted image. Q, 4 are sequentially stored in the storage device 14. The content of Pa is 0 in the example of FIG. 6, and therefore Qo is 10. In step 248, t is set to 1, and the process proceeds to step 255. Since n, 1 raster has not been completed, the process returns to step 245.

In step 245, in the example of FIG.
is not smaller than Δ:e/2, proceed to step 249,
The illustrated processing is performed up to step 252. In step 245, SA becomes Sl and 51=1. At step 247, the content of Pl becomes Ql.

Furthermore, in step 245 again, this time t-L is smaller than ΔV2, and the process proceeds to step 244. In step 244 N=
2, 81=2 in step 245, and QN+P2 is performed in step 247.

That is, as a result, Ql becomes the logical sum of Pl and P2. By repeating the above, conversion processing for one raster is performed from K. 6th
As can be seen from the figure, SA represents the number of dots in the original image obtained by calculating the logical sum corresponding to QJ.

That is, Sa is 1, QO reflects one dot in the original image, Sl is 2, and Ql is the logical sum of two dots in the original image.

Referring again to FIG. 3, when the first rask conversion process is completed, the process proceeds to step 26, and the S determined in step 24 is
It is determined whether a is 1 or not. So is now 1, which is the final first rask after conversion, and is transferred from the storage device 14 to the first rask in the screen buffer 15 in step 27. After setting R=1 in step 28, Q of the storage device 14 is
Clear the 7 area Step 29)
KSB, now Sl, is transferred (step 50). Next, the process proceeds to step 51 in which a second rask, denoted as lCCMPSND, is converted.

The processing after the second rask may be performed in the same manner as the first rask, and the processing content within one line is the same, and is performed using a simple method using the SA created in the first rask. Step 51 is shown in detail in FIG. Step 510
, 511 are initialized. At step 312, the register S &
Read C.

In this case, S is SO. And Q7 and Pl, that is, Q.

and Pa, in this case Qo is 0' and Po is Q
O (step 515). Step 514 is 5=
=8-1, and in step 315 it is determined whether S is o. Now S=o, the values shown are each set to 1 in steps 516-518, and 81 is read in S in step 519. Sl is 2 from FIG. Then, since one raster is not completed (step 52o), step 515 is executed again. Here, the content of Pl is assumed to be Ql. Step 5
14 and 515, and since 8=o is not true this time, the process proceeds to step 521 and sets the number to i+1, that is, 2. Returning again to step 515, Q1+P2 will be executed from now on, and as a result, Ql will become PI+P2, and the same process as the first rask will be performed. By repeating this process, the same conversion processing as that performed in the first rask is performed using the SA created during the first rask processing. Q/- of the second raster remains held in the storage device 14 and is not transferred to the screen buffer.

Returning to FIG. 3, when step 51 is completed, step 55
Proceed to step 1 and perform T-1. T=1.

In step 54, it is determined whether the value is 0 or not. T is now '1' and the process returns to step 51 again. In this step 51, the third
The processing shown in FIG. 3 is performed for the raster. This third
In the rask processing, in step 515, the converted image Q7 of the second rask is logically summed with the third rask Pi. .

Therefore, Ql of the second raster of the converted image is the logical sum of the four dots of Pl and P2 of the second raster and Pl and P2 of the third raster of the original image. As a result, in the same way that Pl and P2 of the original image are logically summed to obtain Ql for each raster, the second rask and the third rask are logically summed in the column direction to obtain the second rask of the converted image. Original image 3rd
When the rask is completed, the process proceeds to steps 55 and 54, and this time T=0, so the process proceeds to step 55. Step 54
Now convert column Q7 to screen buffer Image 9! , transferred as 2 rasters. In step 56, R+1 is performed and the process proceeds to step 57. If all rasters have not been completed, the process returns to step 29 and is repeated. Steps 58 and 59 are SO
The first rask and the second rask are processed in the same way as the second and third rask described above when is not 1.

This achieves the reduction processing that is a feature of the present invention.

Figure S also shows the enlargement process.

The principle of the enlargement process is shown in FIG.

FIG. 7 shows an example of converting a 13-dot original image to a 16-dot converted image. This means that the dot in the original image that is the closest in virtual distance is the dot in the converted image. For example, ■n, ■, and K are closest to dot ■, and dots ■ and ■ are closest to tod ■ and ■, respectively. Also, the dot ■ is closest to both the dots ◎ and ■, and the dot ■ is the dot C1■ in both cases. No logical sum is performed.

FIG. 8 shows a view similar to FIG. 6 when enlarged.

SA corresponds to Pi and indicates the number of reflected dots.

The enlargement process shown in FIG. 3 enters step 41 labeled EXPF8T. Step 41 is shown in detail in FIG.
Extend the first rask.

When the enlargement process of the first rask is completed, in step 42, Qj
Currently, the Qo is stored in the screen buffer 15 for the SO raster, that is, 1
Transfer raster parts. In steps 45 and 44, R is set to 1, T[SR, that is, 5l (=2) is transferred, and the process proceeds to step 45, which is indicated as BXPSND. The details of step 45 are shown in FIG. 10, and the enlargement process for the second rask and subsequent steps of the original image is performed.

The enlargement/enlargement processing after the second rask is also easily performed using the SA created in the first rask, similarly to the reduction processing. When the enlargement process of the second rask is completed, in step 46, the Q
7 for T rasters (currently 2 rasters) is sent to the screen buffer.In step 47, RIKl is added, and in step 48 it is determined whether all rasters have been completed.

When dealing with a color image, the above embodiment can be applied since the logical sum should be an intermediate color between the two colors.

〔Effect of the invention〕

According to the present invention, high-quality pixel reduction conversion with less distortion and omission can be achieved. Further, according to the present invention, the conversion magnification can be set arbitrarily, and a highly usable system as an image editing function can be obtained.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing the system configuration of an embodiment of the present invention, FIG. 2 is a diagram explaining the present invention in detail, FIG. 3 is a diagram showing a processing flow according to an embodiment of the present invention, and FIG. Figure 3 is a diagram showing some of the steps in Figure 3 in detail, Figure 6 is a diagram for explaining the reduction process, Figure 7 is a diagram showing the principle of the enlargement process, and Figure 8 is an illustration for explaining the enlargement process. FIGS. 9 and 10, which are diagrams for explaining the process, are diagrams showing some of the steps in FIG. 3 in detail. 10... Image expansion processor, 12... Temporary storage device 8.15... Pixel conversion processor, 14... Storage device, 15... Screen buffer, 16... Display device. Figure 2 Figure 4 Figure 3 Figure 6 Figure 7 Ward 8 Figure 9

Claims (1)

[Claims] 1. In a method of converting an original image into a converted image having a smaller number of pixels than the original image, where n is the number of pixels in the original image and m is the number of pixels in the converted image, ΔX・n=Δx・m ΔX : Virtual pixel distance of the original image Δx: Based on the virtual pixel distance of both images determined by the virtual pixel distance of the transformed image, all pixels of the original image are converted to the one and only one with the closest distance. When a pixel of an image is a determining pixel, and two or more pixels of the original image are determined as a pixel of the converted image, the pixels are logically summed and converted. A pixel number conversion method characterized by converting images into pixels. 2. In a method of converting a two-dimensional original image into a two-dimensional converted image with a smaller number of pixels than the original image, where n is the number of pixels in the row direction of the original image and m is the number of pixels in the converted image, ΔX・n=Δx・m ΔX: Virtual pixel distance of the original image Δx: Based on the virtual pixel distance of both images determined by the virtual pixel distance of the converted image, all pixels of the original image are When only one pixel of the converted image at a close distance is the determining pixel, and one pixel of the converted image includes two or more pixels of the original image, those pixels are logically summed as pixels of the converted image, and further, based on the above row-by-row conversion, the converted pixels of two or more rows corresponding to the row direction position of the pixel subjected to the above logical sum are calculated. A pixel number conversion method characterized by performing logical sum to form a converted image in one row. 3. The correspondence relationship between each pixel of the original image and the converted image 6 resulting from the conversion in the first line is stored, and the conversions from the second line onwards are performed based on the stored correspondence relationship. A pixel number conversion method according to claim 2. 4. Based on the stored correspondence relationship, the converted pixels of the two or more rows are logically summed to form a converted image of one row, or Pixel number conversion method described in Section 3.