JPH0276367A - Density data processing unit - Google Patents

Abstract

PURPOSE:To reproduce a picture in an excellent way by generating an interpolation data from a contrast pattern of a picture data around the interpolation data. CONSTITUTION:A density pattern data generating means 2 consists of an address generating means 4 generating an address data, a vector digital signal processing section 5 to generate a data (pattern data) to predict an interpolation data and a vector memory 6 storing the pattern data. The output data generating means 3 consists of a digital signal process section 8 deciding a coefficient to calculate to a data around the recorded position between l and (l-1) lines of a picture pattern data from the vector memory 6 to generate an interpolation data and a selector 9 switching the data from line memories LM1-3 and the interpolation data to generate the output data.

Description

DETAILED DESCRIPTION OF THE INVENTION Field of the Invention The present invention relates to a density data processing device, and more particularly to a density data processing device that processes images line by line in facsimiles, printers, and the like.

Conventional Techniques As means for inscribing and thinning out images using density data processing devices such as facsimiles and printers, so-called binary recording of only black and white has been presented at various academic conferences. Furthermore, for recording information having gradation, a method such as that shown in Utility Model Application No. 106948/1983 has been used.

In the conventional method, for example, when performing inscribed interpolation on density data as shown in FIG. The result is as shown in Figure (b).

Problems to be Solved by the Invention However, when performing interpolation in conventional devices, interpolation was simply performed using average data between data t, so for example, the interpolated data circled with an O in FIG. The desired data may become lighter or darker, leading to deterioration in image quality. Further, in order to solve these problems, creating data to be interpolated from all density data in the vicinity has the problem of being impractical due to the increase in circuit scale, logic means, etc.

The present invention has been made in view of the above points, and it is an object of the present invention to provide an image signal processing method that causes less deterioration in image quality during interpolation and can be configured with a relatively simple circuit.

Means for Solving the Problems The present invention provides a gradation data processing device that expands input information by interpolating interpolation data between input density data. density pattern data generation means for generating density pattern data; and n from the density pattern data generation means.
(a) interpolation data generation means for generating the interpolation data according to the density pattern data;

When enlarging the action input information, when generating interpolated data to be interpolated between the input 11m data, the density pattern data generation means generates density pattern data according to the surrounding density pattern from the surrounding density data of the position to supplement the interpolated data. The interpolated data is generated by the interpolated data generating means according to the density pattern data.

Example FIG. 1 shows a block diagram of one embodiment of the invention.

In the figure, 1 is a storage means, 2 is a density pattern data generation means, and 3 is an interpolation data generation means.

The storage means 1 has memories LM1 to LM3 for three lines, and each memory LMI to 3 stores data on lines [-1, 2, e+1]. At this time, if the magnification is r) times, then = n + j (j = o → n
-1) Record the position corresponding to the relational expression. For example, when the magnification n is 2x, the pixel position in the recording line = 100 lines, the pixel position is 50 lines, and when k = 101 lines, 2 = 50 lines.
Or, the 51st line f-even corresponds.

The density pattern data generation means 2 includes an address generation means 4 for generating address data, a vector digital signal processing section 5 for generating data (pattern data) for predicting interpolated values, and a vector memory 6 for storing the pattern data. It becomes more. For example, if the image pattern data is 4 bits, there are 16 patterns, but they can be roughly divided into 4 types as shown in Figure 3.
It can be classified into two patterns. Of these, (d) is (C)
Since it has a symmetrical pattern, it can be roughly divided into (a) ~
It can be summarized into three (C).

For these patterns, for example, if we consider the case where the data as shown in FIG. 2 is doubled, the pattern data vectors Vi and L are defined as Vi, E=.

It is expressed as

Further, the output data generation means 3 uses the image pattern data from the vector memory 5 to generate interpolated data by determining coefficients to be extracted for peripheral data at the recording position between e and [-1 lines. 8 and a selector 9 that switches between data from the line memories LM1 to LM3 and interpolated data to create output data.

The line memories LM1 to LM3 are controlled by a signal process controller 10 in terms of loading and reading data.

Further, the signal process controller 10 supplies the selector 9 with a switching control signal for switching between interpolated data and data from the line memories 1-M1 to 1-M3.

Next, the operation of this embodiment will be explained with reference to FIG.

For example, when interpolating data as shown in FIG. 3(a), if we focus on interpolation position 10 shown by the mouth in FIG. It comes to show. The vector memory 6 has, for example, 4 bits, and the upper 2 bits are used as an index in the direction of arrow E, and the lower 2 bits are used as an index in the direction of arrow F. Also, among the top two bits,
The upper 1 bit is used as the upper part of the interpolation position 11, and the lower 1 bit is used as the lower part index.Of the lower 2 bits, the higher 1 bit is used as the left part index of the interpolation position 11, and the lower 1 bit is used as the right part index. The upper 2 bits are 1 when the density increases from left to right, 0 when the density increases from left to right, and the lower 2 bits are 0 when the density increases from bottom to top. 1. The number from top to bottom is 0. Therefore, the pattern data at interpolation position 10 shown in FIG. 2 is expressed as 1001, and it can be seen that the vector runs diagonally to the right.

This density pattern data creates interpolated data that is strongly influenced by the density data in the diagonal right direction. For example, if you weight density data in the right diagonal direction by 3/8 and weight data in the left diagonal direction by 1/8, the interpolated data a
is a = (1/8)Di-+, t+(3/8)Di-1, t+1
+(3/8)Di, t+(1/8)Di, t++=(1
/8)・16+ (3/8)・64-t (3/8)
・ 64+ (1/8) ・ 16-5
It is calculated as 4. Therefore, interpolation data a at interpolation position 11
becomes 52. This value is close to 64 of the density data in the right diagonal direction. Therefore, a better image can be reproduced when printed.

Further, when creating interpolated data for the position (B) in FIG. 2, it is necessary to determine it from the density pattern data of νi,L and Vi-1,t. For example, when trying to create interpolation data at interpolation position 11 using 'a degree data as shown in FIG. , t) = 26, but V! , t and Vi-+, creating interpolated data b by weighting according to t, b = (1/3) D i -1, t + (2/3)
D i , t-30,66→31. In this way, in part (A) of Figure 2, the small part (
B) By enlarging the area, the total density is dispersed, resulting in a clearer image.

An image is obtained. Furthermore, the data in part (C) of Figure 2 is also included (B
) can be created by weighting in the same way as the part.

Although this embodiment deals with double enlargement, it can be inferred that these methods can be applied when enlarging by an integral number of times.

Note that the accuracy of the contents of the vector memory 6 can be further improved by increasing the number of bits and setting the weight for each direction more precisely.

Note that pattern detection is not limited to the above example, and it is possible to have as many types as each data difference and the number of bits of the vector memory 6.

Furthermore, when data is to be thinned out, instead of being thinned out strictly, it can be assumed that data is thinned out by correcting it by detecting a pattern.

Effects of the Invention As described above, according to the present invention, since interpolated data is generated from the grayscale pattern of surrounding image data, it is possible to interpolate data close to ideal data to be interpolated between each image data. , it has the advantage of being able to reproduce images well.

[Brief explanation of the drawing]

FIG. 1 is a block diagram of one embodiment of the present invention, FIG.
4 and 5 are diagrams for explaining the operation of an embodiment of the present invention, and FIG. 6 is a diagram for explaining an example of the conventional operation. 1... Storage means, 2... Density pattern data generation means, 3... Interpolation data generation means. Patent applicant: Victor Japan Co., Ltd. Figure 2 Figure 3 [al fbl tcl
(diDi, l Di
-j, l-+ Di, l D black-1,
l◆I DiJ Di-1, l◆I
Di,l Di-1,1ku1
・ ・-( WJ 5 + 40) □ a6 diagram q1 0081664 0f 81664158 81; 1664
1680 12■ 0C f161 i −1 +. f16) i◆1 瀉b1 312 [Yes] 40◎40 Su40 @) 40 @) 12) 4o@12s t.

Claims (1)

[Scope of Claims] In a density data processing device that expands input information by interpolating interpolation data between input density data, density pattern data that generates a density pattern according to a density pattern of density data surrounding the interpolated data. A density data processing device comprising: a generation unit; and an interpolation data generation unit that generates the interpolation data according to the density pattern data from the density pattern data generation unit.