JPS63311865A - Scanning line density conversion method - Google Patents

Abstract

PURPOSE:To quicken the conversion of a picture data at an optional magnifica tion simply and to reduce the distortion by interpolating an input picture signal of l-line/mm by an interpolation magnification S, converting the signal into a picture signal of S.l-line/mm, and converting the result into a picture signal of m-line/mo by the magnification data conversion. CONSTITUTION:The variable area with designated magnification is split into plural numbers and an interpolation magnification S (S<=m/l) represented by 2+ or -<n> (n is a positive integer) is set to each split area. Then the input picture signal of l-line/mm is converted into interpolation data S.l-line/mm and the designation multiple m-line/mm is designated by the magnification data conver sion. Thus, in the scanning line density with an optional magnification, the input picture signal is converted into 2+ or -<n> times of scanning line density data at a high speed with less distortion by a simple algorithm in response to the destination magnification. Thus, the scanning line density data designated by the simple algorithm is expanded further, then the high speed scanning line density conversion with less distortion is attained even with the entirely simple and wide variable area of the designation magnification.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a scanning line density conversion method when performing conversion such as enlargement or reduction of an image.

Prior Art Conventionally, there have been the following methods for scanning line density conversion. (Reference: Arai, Yasu 1) A study of facsimile line density conversion, Institute of Image Electronics Engineers, Vol. 7, No. 1, P, 11-18 (
1978) Method 1: Enlarge or reduce by repeating or thinning the input data Method 2: Enlarge or reduce by interpolating or averaging the input data Using Figures 7 (A) and (B), the above two methods To explain the scaling of 1. d13. Yus, d3+, d33
．． If you use d3s as input data and double the main and sub-scanning using method 1, do = d+z = d2t
= dzzd13 ” d14 = d23 : du
The same goes for the following. On the other hand, in method 2, d12 = (do + dos)/2 + d14=
(Mountain 3 + dos) / 2.

d3z=(dat+d32)/2°d34=(
da3+ d3s)/2゜d++ = (do + d
at )/2°d22= (d+2+d32) 72
The same goes for the following. When enlarging the figure three times in the same way in (B), in method 1, do = d+2
= d+3= ch1= d22=d23= dat
= d3z = seasonal, etc. However, in method 2, for example, to calculate d22 by interpolation, do, d14
．． Each of the input data of d41 and d43 is added with weighting inversely proportional to the spatial distance from d22, which complicates the calculation. In the case of reduction, the same figure (A) and (B
), when all the data is input data and the solid line circles are reduced data, in method 1 the dotted line circle data is thinned out, and in method 2, in the same figure (A), d
o = (do 10dlz + chl+d22)/
4 In the same figure (B), do = (do + d+z + d+3+ d2+
+ d22+ dz3+ dat +d32+d33)
The average is calculated as /9.

Problems to be Solved by the Invention However, in method 1, the arbitrary magnification conversion can be configured using a simple algorithm, so that the hardware arithmetic circuit can be made faster, but on the other hand, it has the disadvantage that the converted data is highly distorted. On the other hand, in method 2, the distortion of the converted data is small, and for the multiplication conversion of 2 times (n is a positive integer), it is only necessary to repeat 2 times or negative multiplication, and the calculation algorithm is simple. Compared to Method 1, which is complicated in terms of magnification, it is difficult to increase the speed of the arithmetic circuit.

The present invention solves the above-mentioned problems and provides a scanning line density conversion method that allows arbitrary magnification conversion of the scanning line density, which reduces distortion of converted data and enables high-speed processing.

Means for Solving the Problems The present invention divides a variable area of a specified magnification into a plurality of parts in order to convert an input image signal with a scanning line density of m lines/mm+ into an image signal with a scanning line density of m lines/mm. Then, for each divided area, an interpolation magnification S (S5m/l) that can be expressed as 2 (n is a positive integer) is set, and the input image signal of lines/m is multiplied by an interpolation magnification of 8 times. The above object is achieved by converting the interpolated data into S.l lines/ffIIT+ and then enlarging it at a specified magnification of m lines/mm by enlarging data conversion.

According to the method described above, the present invention converts an input image signal into scanning line density data of ±n 2 times with small distortion at high speed using a simple algorithm according to a specified magnification in scanning line density conversion at an arbitrary magnification. By further expanding the specified scanning line density data using a simple algorithm, it is possible to perform high-speed scanning line density conversion that is simple overall and has less distortion even when the variable area of the specified magnification is wide. This is what I did.

Example First, we will explain the reason for the small distortion as shown in Figure 6 (5) to (
D) will be explained using a one-dimensional signal for simplicity.

In the same figure (DJ) is an example of 3 times enlargement. When the analog image signal f in the figure (3) is quantized at the points S.about.S4, the quantized signal in the shaded area is obtained.This quantization The signal obtained by converting the signal to three times the scanning line density using Method 1 has a larger distortion (error) than the signal (dotted line) that was normally quantized at three times the scanning line density.Figure (B) is the case when the scanning line density is converted to 3 times the scanning line density using method 2, and the distortion is small, but since the conversion is not 2 degrees, the arithmetic circuit becomes complicated (especially the arithmetic circuit for two-dimensional signals), which makes processing difficult. As mentioned above, it takes time. Figures (0) to (DJ) are examples of the present invention, and Figure (C) shows a signal obtained by doubling the data interpolation using the above method. Figure (D) is a diagram in which the signal in Figure (C) has been enlarged 372 times using method 1, and the signal has been restated at equal intervals. As shown in the same figure (
This is obvious compared to 5).

Furthermore, it will be briefly explained that conversion processing with small distortion can be performed even when the specified magnification variable region is wide.

If the specified magnification variable area is wide, for example from 0.5x to 8x, in order to satisfy all the specified magnifications within this area, it is necessary to interpolate 0.5x and then expand to the specified magnification. It is possible to convert

However, after 0.5 times (2 times) interpolation, the method 1 up to 8 times
When the image is enlarged further, the distortion of the converted signal increases. This is also clear from the fact that method 1 is equivalent to 1x interpolation followed by 3x expansion. The distortion of this converted signal is
Depends on the number of consecutive iterations during interpolation. For example, in the case of triple conversion using method 1, the number of consecutive iterative processes is two (see Figure 6). On the other hand, in FIG. 6(D) which is an example of this method, the number of times is one. This difference is connected to the magnitude of distortion. Therefore, if the interpolation magnification is set as shown below according to the specified magnification so that the number of repetitions is reduced, a converted signal with less distortion can be obtained.

When 0.5 times ≦ M < 1.0 times, the interpolation magnification is 0.5 times (maximum number of consecutive repetitions is 1 time) When 1.0 times ≦ M < 2.0 times, the interpolation magnification is 1.0 times (maximum When 2.0 times ≦ M < 4.0 times, the interpolation magnification is 2.0 times (the maximum number of continuous repetitions is 1 time) When 4.0 times ≦ M < 8.0 times, the interpolation magnification is 4.0 times (maximum number of consecutive repetitions: 1 time) When M = 80 times, interpolation magnification is 80 times (maximum number of consecutive repetitions: 1 time) M is the specified magnification.

A specific embodiment of the present invention will be described below.

FIG. 1 is a block diagram of an apparatus for realizing a scanning line density conversion method in an embodiment of the present invention.

In this embodiment, repetition of method 1 in the conventional example is adopted as enlarged data conversion.

In Figure 1, 1 is an input terminal for a designated magnification in the main scanning direction, 2 is an input terminal for a designated magnification in the sub-scanning direction, 3 is an interpolation magnification setting section in the main scanning direction, and 4 is an interpolation magnification setting section in the sub-scanning direction. , 5 is an input terminal for the quantized input image signal, 6 is a sub-scanning data interpolation section, 7 is a main-scanning data interpolation section, 8 is a sub-scanning data repeating section, 9 is a main-scanning data repeating section, and 10 is an output image signal. The output terminal 11 is a timing signal generator.

Now, set the scanning line density of the input image signal to t lines/
The operation will be described below in the case where the scanning line density of the output image signal is set to m lines/mIT+ in both the main and sub-scanning, and the main and sub-scanning are expanded three times.

The designated magnifications of input terminals 1 and 2 are input to the main scanning interpolation magnification setting section 3 and the sub-scanning interpolation magnification setting section 4, respectively,
The interpolation magnification is set. In this embodiment, the relationship between the designated magnification M and the interpolation magnification S is as follows.

When ±n 2≦M<2 5=2 (n is a positive integer) Since the specified magnification M is now 3 times, the interpolation magnification is 2 times. Furthermore, the interpolation magnification setting sections 3 and 4 provide the repeating sections 8 and 9 with the following:
A repetition coefficient (3/2 in this example) is output.

The input image signal of the input terminal 5 is input to the sub-scanning data interpolation section 6, and the data in the sub-scanning direction is input to the sub-scanning interpolation magnification setting section 4.
The data is doubled according to the interpolation magnification from , and is input to the main scanning data interpolation section 7 . In the main scanning data interpolation section 7, the data in the main scanning direction is further doubled by the interpolation magnification from the main scanning interpolation magnification setting section 3. Thereafter, the data in the sub-scanning direction is expanded to 3/2 by the sub-scanning repeating section 8 using the repetition coefficient from the sub-scanning interpolation magnification setting section 4, and the main scanning repeating section 9
The data in the main scanning direction is enlarged to 3/2 by the repetition coefficient from the main scanning interpolation magnification setting section 3 and becomes an output image signal at the output terminal 10. Note that each of the units 6 to 9 is controlled by each timing signal from the timing signal generating unit 11. Hereinafter, more detailed examples of the above-mentioned parts 6 to 9 will be shown. The main scanning direction and the sub-scanning direction can basically have the same configuration, so to simplify the explanation, only the main scanning direction will be described. Also, for the input/output lines of each block, I thought that it would be good if I could understand the signal flow, so I added 1 for simplicity.
It is marked with book lines.

To simplify the explanation, the variable area of the specified magnification M is 1≦
M<8.

FIG. 2 is a detailed configuration diagram of the main scanning data interpolation section 7 shown in FIG. 1, and shows a case where input image data can be supplied sufficiently quickly.

Moreover, FIG. 3 shows the timing.

In FIG. 2, 20 is a signal line of an input image signal array;
21, 22, and 26 are latch circuits, 23 to 25 are averaging circuits, 27.28 are selector circuits, 29 are output image signal string signal lines, and 30 to 33 are PI, p, , respectively.
, p, , pa are signal lines for timing signals. On the other hand, (1) to (3) in Fig. 3 are each designated magnification M
is 4≦M<8. When 2≦M<4.1≦M<2, the input image signal sequence is 0, (4) is P, signal, (5) is P22
multiplier, (6) is P, signal, and (7) to @ are the output of latch 21 and output of latch 22 when the specified magnification M is 4≦M<8.2≦M<4゜1≦M<2, respectively. , α葎~Q! 9 is a string of output image signals when the specified magnification M is 4≦M<8°2≦M<4.1≦M<2.

It represents the timing of.

The operation of the above configuration will be explained below.

The selector 27 receives the timing signal P of the signal lines 30 to 32.
1, P3, and P4, a selector circuit selects P4 when 1≦M<2, P1.4 when 2≦M<4, and P1 when 4≦M<8 according to the specified magnification M, and outputs it as a signal line 34. It is switched by the interpolation magnification signal P supplied from the main scanning interpolation magnification setting section 3 in FIG. 1 on the signal line 33. Input image signal sequence D+ of the signal line 20 d+, dl, ds,...
) is set in the latch 21, and the output of the latch 21 is set in the latch 22 in synchronization with the rise of the timing signal of the signal a34 ((7) to (2) in FIG. 3). At this time, the data width of the input image signal sequence varies depending on the designated magnification ((1) to (3) in FIG. 3), but this is to make the processing faster. The averaging circuits 23 to 25 are 2
This is a circuit that calculates and outputs the arithmetic average of two input image data.
The averaging circuit 24 averages the outputs of the latch 22 and the averaging circuit 23, and the averaging circuit 25 averages the outputs of the latch 21 and the averaging circuit 23. Now, assuming that the output of the latch 21 is city and the output of the latch 22 is d, then the averaging circuits 23 to 25
The output of is as follows.

The outputs of the averaging circuits 23 to 25 and the output of the latch 22 are set in the latch 26 in synchronization with the rise of the timing signal on the signal line 34. The selector 28 is a selector circuit that switches the output of the latch 26 in accordance with the timing signal P4 of the signal line 32, so that the output shown in FIG. can get. That is, when the specified magnification M is 1≦M<2, only the output of the latch 22 is selected, when 2≦M<4, the output of the latch 22 and the output of the averaging circuit 23 are selected, and 4≦
When M<8, the output of the latch 22 and the averaging circuit 23~
When 1≦M<2, all outputs of 25 are selected, and with the period of the timing signal P4 of the signal line 32, i d,, d,, d,...
) When 2≦M<4, (a,, (dH 1 dt)/2 + city, (d, + a,)/2,...1. When 4≦M<8, t d+, (3d+-+-d,)/
4. (d+ + dl)/2゜(4t+3d*)/
4. Outputs a signal string in the order of a, . . . 1. In this embodiment, the designated magnification M is 1≦M<8,
The number of stages of the averaging circuit is two stages (the first stage is the averaging circuit 23).
．． The second stage is averaging circuits 24 and 25), but if you make it a 3-stage configuration and add four averaging circuits to the third stage, it will be 16.
It is clear that it is possible to reduce the number of times to less than twice, and if the n-stage configuration is used in the same manner, it becomes possible to make the number of times less than twice as much.

With the above configuration, the period of the output image signal sequence D1 is constant regardless of the specified magnification, and if the interpolation magnification is small, the period of the input image signal sequence can be shortened, so that interpolation processing can be performed at high speed. can. In this embodiment, the interpolation magnification is set as follows with respect to the designated magnification M, and the effect thereof is obvious as shown in FIG.

When the specified magnification M is 1≦M<2, the interpolation magnification is 1x 2≦M<
41 72 times l 4 times Now, FIG. 4 shows an operation flowchart of the main scanning repeating section 9 of FIG. 1.

The enlarging process in the repeating section is the same as that of method 1 of the conventional example, but will be explained in detail using the operational flowchart of FIG. 4.

In FIG. 4, R is an arithmetic register, A, l! :B is a constant register, and the expansion rate of the repeat part is A/B (A≧B
). The operation flowchart will be explained in order below. Reference numeral 41 indicates the starting point of image processing for one line. 42 is an initial setting section which clears the register R to O and performs initial setting. Reference numeral 43 denotes a subtraction unit that subtracts the contents of register B from the contents of register B, and the comparison unit in the box judges whether or not the result is negative. In the case of this embodiment, since B>0, the output of the subtraction unit 43 is always negative in the first process. If the comparison member determines that it is negative, it goes to the comparison section 45,
It is determined whether the processing of the l line is completed. If the comparison unit 45 has completed the determination, the image processing calculation end point 49 for one line is reached.
If the process is not completed, the process goes to 46 and inputs data for one pixel. In the first process, the data for the first pixel is input here, and thereafter, the data for the second pixel and the third pixel are input in this order every time the line passes through this point.

After inputting the data, processing goes to the adder 47 and register R
After adding the contents of register A to the contents of , go to 48,
Outputs one pixel data. On the other hand, if the comparison member determines that it is not negative, the process goes to 48 and outputs the data of one pixel. That is, if it is determined that the value is not negative, the input data is not updated by inputting one pixel data, so the input data of the previous process is output as is. After outputting one pixel data in step 48, the process returns to the subtraction unit 43. The processing from the subtraction unit 43 to the 1-pixel data output 48 is a unit of processing for one pixel, and this unit of processing is similarly repeated until the comparison unit 45 determines that the image processing operation for one line has ended. .

Through the above arithmetic processing, the input data is updated only when it is determined to be negative by the comparison member, and at other times, the data input in the previous pixel process is not repeatedly output, but is enlarged repeatedly. be able to. Note that the addition/subtraction processing used here is the commonly used DDA
(Cigital differenti-al a
It embodies the concept of "nalyzer".

FIG. 5 shows an example of repeated enlargement using the operation flow shown in FIG. FIG. 5(5) is an example when the magnification rate is 3/2, and FIG. 5(B) is an example when the magnification rate is 7/4.

In FIG. 5, the comparison contents indicate the values of the registers input to the comparison unit 44 in FIG. It shows. As is clear from Fig. 5, when the comparison content of R is negative, the input data string is not changed to the output data string, and when it is not negative, the same data as the output data of one pixel before is repeatedly output to the output data string. Ru.

The operation of the sub-scanning repeating section 8 is basically the same as that of the main-scanning repeating section 9, so a description thereof will be omitted.

Effects of the Invention As described above, the present invention provides an interpolation magnification S (according to the specified magnification m/L) when converting an input image signal with a scanning line density of lines/mm to an image signal with a scanning line density of m lines/mm. However, S≦
m/l) and interpolates the input image signal of books/plates at the interpolation magnification of 8 times to convert S into an image signal of books/m, and then By converting the image signal into enlarged data and converting it into the image signal of m lines/lTl lines/mm, arbitrary magnification conversion of image data can be performed easily and quickly, and distortion can be reduced even over a wide range of magnifications. , the effect is great.

[Brief explanation of drawings]

FIG. 1 is a block wiring diagram of a device that implements a scanning line density conversion method in an embodiment of the present invention.FIGS. 2 and 3 are block wiring diagrams and timing diagrams of a main scanning data interpolation section in the same device. , FIG. 4 is an operation flowchart of the main scanning repetition section in the same device, FIG. It is. 3... Main scanning interpolation magnification setting section, 4... Sub-scanning interpolation magnification setting section, 6... Sub-scanning data interpolation section, 7... Main-scanning data interpolation section, 8... Sub-scanning repeating section , 9... Main scanning repeat section. Name of agent: Patent attorney Satoshi Nakao and one other person Figure 2 (Sot) M-ke designated rower Figure 4 Figure 6 - Master! Direction So St 52 59 54
So 51 52 bz -wx Fig. 7 (B

Claims (2)

[Claims] (1) When converting an input image signal with a scanning line density of l lines/mm to an image signal with a scanning line density of m lines/mm, the specified magnification is m/mm.
The interpolation magnification S (however, S≦m/l) is set according to l, and the input image signal of l lines/mm is interpolated to the interpolation magnification S times to produce an image signal of S·l lines/mm. After conversion, the image signal of S·l lines/mm is converted into enlarged data,
A scanning line density conversion method for converting the m lines/mm image signal. (2) Divide the variable area with the specified magnification into multiple parts and calculate 2^±^n for each divided area (where n is a positive integer)
2. The scanning line density conversion method according to claim 1, further comprising setting an interpolation magnification S that can be expressed as .