JPS6179374A - Picture processing system - Google Patents

Abstract

PURPOSE:To apply multi-value in pseudo way to a binary-coded picture signal and to attain gradation reproduction faithful to an original by converting a binary-coded noticed picture element data into an intermediate level picture element data based on the binary-coded picture element data and plural binary- coded noticed picture element data adjacent to the said binary-coded noticed picture element data. CONSTITUTION:When the equipment is in the copying machine mode, an SW1 is thrown to a ternary-coding circuit 7 and the digital picture signal is subject to ternary-coding by two threshold values. Further, ternary-coded signals Sig1, Sig2 are inputted to a laser beam printer (LBP) by a changeover switch SW2 and an original picture is dot-reproduced in the 3 kinds of white/black/gray densities. Further, when the equipment is in the picture storage mode, the SW1 is thrown to the binary-coding circuit 2, its binary-coded output signal is coded by a coding circuit 3 and stored in a file equipment 4. Further, the coded picture data read from the file equipment 4 is converted into pseudo ternary-coded signals Sig1' and Sig2' similarly as the plural modes and the image is reproduced by the LBP8.

Description

Detailed Description of the Invention [Technical Field 1] The present invention relates to an image processing method applied to image processing devices such as digital copying machines and facsimile machines, and in particular to image processing that converts a binary image signal into multiple values based on it. It is related to the method.

[Prior Art] A conventional digital copying machine quantizes an image signal read by a solid-state image sensor (CCD) or the like into a 6-8 bit digital signal, and performs signal gradation processing according to the printer capability. That is, a printer that can only perform binary reproduction of white and black is subjected to the binarization process using one threshold value, and a printer that is capable of ternary reproduction of white, black, and gray is subjected to the binary conversion process using two threshold values.
Value processing is performed. By the way, even if you have a printer that is capable of reproducing fi files (as mentioned in 3), image data must be ternarized (2 Binarization (l-bit) handles less information than binary data, which suits the actual situation.Therefore, even though printers have the ability to reproduce 3-value data, once the data has been converted into 2-value data, it cannot be reproduced in 2-value data. The reality was that they were not able to obtain the desired results.

[Objective] The present invention has been made in view of the above-mentioned prior art, and its purpose is to propose a new method for pseudo-multi-leveling a binary image signal. The purpose is to enable gradation reproduction that is faithful to the original.

[Examples] Examples of the present invention will be described in detail below with reference to the accompanying drawings.

FIG. 1 is a block diagram of a printer apparatus to which an embodiment of the present invention is applied. In the figure, l is an image reading unit that A/D converts the image signal read by the CCD and outputs it, and 2 is the image reading unit that binarizes the digital image signal using one threshold value.
The digitizing circuit 3 converts the binarized signal into a modified Hofmann (
4 is a file device consisting of, for example, an optical disk; 5 is a decoding circuit that decodes MH encoded data read from the file device 4 into a binary code; 6 is a decoding circuit according to the present invention; Pseudo-ternarization of binary signal (2
7 is a pseudo ternarization circuit that converts the digital image signal into two threshold signals Sig and Sig2'.
-c 3i+t'i conversion (2 hits) Signal Sig I
.

A ternarization circuit that converts to Sig 2, 8 is a laser beam printer (LBP) with a 3-gradation recording method of white, black, and gray, for example.
It is.

Now, if the device is in copying machine mode, SWI is ternary circuit 7
The digital image signal is ternarized by two threshold values. Furthermore, the ternary signal Sig1. Sig
2 is input to (LBP) by the switch SW2, and the original image is reproduced in dots in three densities: white, black, and gray. Furthermore, when the device is in the image storage mode, SWl is connected to the binarization circuit 2 side, and its binarized output (1 bit) signal is encoded by the encoding circuit 3 and stored in the file device 4. . The encoded image data later read out from the file device 4 is decoded into a binary image signal by the decoding circuit 5, and pseudo-ternarized (binary) by the pseudo-ternarization circuit 6 in the same manner as in the copy mode described above. bit) signal Sig bit, Si
g 2 ′, and the image is reproduced by the LBP 8 via the SW 2 connected to the pseudo ternarization circuit 6 side.

FIG. 2 shows, as an example, a ternarized signal Sig1. Sig
2 is a diagram illustrating the relationship between laser beam dots and laser beam dots.

In the figure, Sig 1. Sig 2 indicates the output of the ternary circuit 7, and LBP 8 generates beam dots for forming three densities by combining these signals.

In the case of pixel l, both signals Sig 1. Since both Sig 2 are at HIGH level, the laser beam lights up (ON) L for one pixel, reproducing a completely black L dot pixel. In the case of pixel 2, both signals are Sig 1. 51g2 both LO
Since it is at W level, the laser beam is turned off by one pixel (OF
F) Regenerate one dot pixel within L, complete. In the case of pixel 3, signal Sig I is at HIGH level and Sig
Since dot 2 is at LOW level, the laser lights only the first half of one dot. Therefore, compared to the case of pixel 1, only half the light energy is irradiated onto the drum, and a substantially gray dot pixel is reproduced. It's like 2, LBP
8 is a beam control corresponding to one dot pixel, the first half of which is lit at the HIGH level of the signal Sigl, and the second half of which is lit by the signal S.
By turning on the ig 2 at HIGH level, you can switch between three different types of energy and apply it to the drum.

FIG. 3 is a diagram showing the operational truth values of the pseudo ternarization circuit.

Now, the pixel data of interest to be ternarized is P(0), the pixel data immediately before it is P(, -1), and the pixel data immediately after it is P(, -1).
(1), the pixel data of interest P(0) is placed in eight different states. The pseudo ternarization circuit 6 uses these as input conditions to generate a pseudo ternarization output signal S of the target pixel data P(0).
ig bi, 51g2' (in the diagram, white is O1, black is 1, gray is 1/
2) is output. Now, the input binary data array H(P(1), P'(0)).

Regarding P(-1)), if black is expressed as 1 in the figure and white is expressed as O in the figure, when the pixel data of interest P(0) is a stable change point from white to black or from black to white. is H(1,1,0
), H(0,1,1), H(0,0,l), H(1,0
, 0), but in this embodiment, the pixel data of interest P
(0) is black, and from white to black (or from black to the next white)
Pseudo ternary output 51g1. 5
Let us convert 1g2' to gray level (1/2).

FIG. 4 is a circuit diagram of an embodiment for realizing the pseudo ternarization operation of FIG. 3. In the figure, the latest binarized data P(1
) is the data input of the D-type flip-flop (F/F) 10, the output Q of FF10 is the immediately preceding target pixel data P(0).

The output Q of FFII is the previous data P (-1)
is held. Now, the above array H(1, l, 0)
When this occurs, all the inputs of the NAND gate 15 become HIGH level due to the inverter 14, and the output becomes L.
Obtain OW level.

On the other hand, when the array H (0, 1, l) occurs, the output of the NAND gate 17 is similarly LOW by the inverter 16.
get level. Therefore, the output of the AND gate 18 becomes a LOW level only when two consecutive arrays H (0, 1, 1) and H (1, l, 0) occur, and this is sent to the F/F 13.
By connecting to the clear terminal of , the output Q can be forcibly reset. Here, the pixel data of interest P
(0) to the data input terminals of F/F12 and F/F13, the next clock timing (not shown)
The output Q of F/F12 becomes HIGH level, and the F/F
The output Q of No. 13 becomes LOW level. Of the 2-bit data thus formed, the output Q of the F/F 12 corresponds to the above-mentioned Sig 1', and the output Q of the F/F 13 is the array H(0, 1, 1), H(, 1, 1, 0 51g whose output is forcibly kept at LOW level only when ) occurs.
2', and as a result, instead of forming a black pixel, a gray level state can be generated.

The conversion circuit section surrounded by the dashed line in the figure is a very simple circuit, and in this embodiment, this section is
(Programmable Logic Array)
It consists of The PLA can easily be applied to algorithms other than those shown in this embodiment.

FIG. 5 is a circuit diagram of another embodiment for realizing the pseudo ternarization operation of FIG. 3. The difference in the configuration in this figure from that in Figure 4 is that the output of NAND gate 17 is connected to the clear terminal of F/F 12, and the output of NAND gate 15 is connected to F/F1.
This is because they were connected to the clear terminals of No. 3, respectively. In this way, when array H is (0°1.1), 51g1' will be HIGH
Then, 51g2' becomes LOW, and if we look at FIG. 2 in which the beam scanning direction is set to the right, even with the same gray color, the first half of one pixel becomes black and the second half becomes white. Therefore, the immediately following pixel data P(-1
), black continuity is maintained. Also, array H is (1,
1, O), 51g1' is LOW, 51g2' is HIGH, and in the same way, the first half of the gray pixel becomes white and the second half becomes black. Therefore, black continuity is maintained in the immediately preceding pixel data P (1). One object of the present invention is to give softness to pseudo-ternary-reproduced characters, in particular, to reproduce the edges of black line parts having a width of several pixels in gray. Therefore, as a modification of the above algorithm, 4 before and after the pixel data of interest P(0)
It is also possible to determine gray based on pixel data.

Now, define this array H as (P(2).

P (1), P (0), P (-1),
P (-2)), and the array of pseudo ternary output signals is M(S(2), S(1), S(o)
, S (-1) , 5(-2)) then H(1
, 1, 1, 0, 0), then M(1, 1, 1/2.0.
O), or M in the case of H(0, 0, 1, l, 1)
(0, O.

]/2, 1.1). If you do this, 1~
A two-pixel wide black line has the effect of clearly emphasizing its edge portion without ternarizing it, and ternarizing it only when it is wider than that.

In addition, in this embodiment described above, edge portions are determined only in one-dimensional direction, but by using a similar circuit and a line memory for two lines, processing in two-dimensional V direction is also possible, which makes it easier to process the original. It is clear that faithful ternarization is possible.

Further, in this embodiment, the binary data source is shown as an output example of the file device 4, but the present invention can be similarly applied to, for example, the receiving mode of a facsimile machine to which the present invention is applied.

FIG. 6 is a diagram showing a circuit of an embodiment for pseudo-four-value conversion.

If the 4 values are 1 for black and O for white, and the 273 level and 1/3 level between them, then from the array H of 5 pixels including the pixel of interest P (0), according to the following algorithm, the intermediate density, Can generate 1/3.2/3 level.

H (0, 0, 0, 1, 1) → M (0, 0, 1/3, 1
．． 1) H (0, 0, 1, 1, attack) → M (0, 0, 2/
31. *) H (1, 1, 0, 0, 0) → M (1, 1,
1/3, 0.0) H (wood, l, 1, 0. O) → M (*
, 1.2/3, 0.0) * means that any value of 1.0 is acceptable. In the figure, F/F21 to F24 are F/F2
1, and NAND gates 30 to 33 are shown in the figure, respectively.
- Configuring etch detection gates in four ways. The result A
The output of the ND gate 34 is LOW level when the pixel data of interest P(0) is O and is to be converted into 4-value (173 level), and similarly, the output of the AND gate 35 is LOW when the pixel data of interest P(0) is O. 1, and four values (2
It becomes LOW level only when it should be converted to 73 level). Now, if an array H to be converted to 1/3 level is generated, the outputs of AND gates 34 and 36 become LOW level, clearing F/F38 and F/F39, and the outputs 51g2" and Sig 3" are at LOW level. On the other hand, since the F/F 37 is preset, its output 51g1" becomes HIGH level. Therefore, the laser beam 21 outputs Sig for the target pixel data P(0).
With a signal of only B′, that is, with an energy of I/3,
The dots are reproduced, and as a result, image reproduction at 1/3 level becomes possible. Similarly, when converting to 273 level, since the pixel data of interest P (0) is 1, Sig Bi',
Sig 2” becomes HIGH level, 51g3” becomes A
ND gate 35 output and AND gate 36 output are LOW
level, F/F39 is cleared and as a result L
It becomes OW. In other words, the laser beam is Sig 1 . S
ig 2 ″ enables 2/3 level image reproduction.

[Effects] As explained above, according to the present invention, binary data is pseudo-predicted to reproduce an image, which leads to an increase in the amount of information and processing time due to handling data with more than 3 values. You can get high image quality without any trouble. Furthermore, since the hardware configuration is simple, it can be said to be a very effective method suitable for cost and high-speed processing.

[Brief explanation of drawings]

FIG. 1 is a block diagram of a league printer device to which an embodiment of the present invention is applied, and FIG. 2 shows an example of ternary signals Sig I, Sig
2 and a laser beam dot, FIG. 3 is a diagram showing the operation truth value of the pseudo ternarization circuit, and FIG. 4 is an example of realizing the pseudo ternarization circuit shown in FIG. 3. FIG. 5 is a circuit diagram of another embodiment that implements the pseudo-ternarization operation shown in FIG. 3, and FIG. 6 is a diagram showing a circuit of an embodiment that performs pseudo-four-value conversion. Here, 1...reading section, 2...binarization circuit, 3...
- encoding circuit, 4... file device, 5... decoding circuit, 6... pseudo ternarization circuit, 7... ternarization circuit,
8... Laser beam printer (LBP).

Claims (1)

[Claims] An image characterized in that the binarized pixel data of interest is converted into intermediate level pixel data based on the binarized pixel data of interest and a plurality of binarized pixel data adjacent to the binarized pixel data of interest. Processing method.