JPS6014573A - Method and apparatus for processing picture signal - Google Patents

Abstract

PURPOSE:To obtain a picture alike a sand surface by changing the 2nd random signal in the timing represented by a numeral in a sand surface particle distribution table selected sequentially in response to the value of the 1st random signal and binary-coding an input picture signal through the use of the value of the 2nd random signal as a threshold value arranged two-dimentionally on a picture screen. CONSTITUTION:The content of a subtraction circuit 33 is decremented by 1 at each clock pulse transmitted via a gate circuit 1, while the content of a random signal generating circuit 31 is updated. Then a storage device 32 outputs a numeral of the sand surface particle distrubution table according to the address respresented by the content of the random signal generating circuit 31. Further, when the content of the subtraction circuit 33 reaches 0 as the result of subtraction , a random signal generating circuit 34 is commanded to be updatd for the content. Further, the output data of the storage device 32 is fetched by the subtraction circuit 2 at the same time. Then the content of the random signal generating circuit 34 is outputted to latch circuits 18, 21 as the threshold value signal to binary-code the picture signal.

Description

[Detailed description of the invention] Industrial use 1! T The present invention scans the input image [elephant? ! The present invention relates to a 1iii'i antagonistic processing method and apparatus for producing a so-called halftone image based on a binary image signal (IJ'J!Jlk), which is expressed as a halftone image.

Conventional structure and its problems It is common to create a halftone image using a contact screen using a photo-optical method, but in recent years, the image data has been digitally processed to create halftones. Some of the methods have become more sophisticated. In principle, this is a type of method called a dither method.

Below, using the 11th Th', the dither method will be explained in 61
1.

The figure (8 indicates the quantization value of the original image signal, and the size of the number represents the density r of the image.
) shows the /ζ-th threshold value table for converting the image signal into 21 pixels, and corresponds one-to-one with each pixel of the image data. 4x4 in bold frame C
The numerical array in a matrix (hereinafter referred to as degree matrix) is repeatedly expanded into two dimensions. In the same figure (q represents a binarized image (elephant), and the image 1 elephant data in the figure () ~ which is smaller than the threshold 11σ in the figure (B) is white,
If they are larger or the same, they are black. For example, the same figure (A)
The 1-drop data a and b have quantization values of 1 and 2, and
t is compared and identified using the thresholds 1 and 9 of the corresponding coordinate points in the same figure (B), and is represented in black and white, respectively in the same figure (q). The number of blacks exceeding the threshold l1iI increases, and therefore the number of blacks increases.In this way, the number of blacks proportional to the level of the image No. 1 signal is generated for each degree matrix, and the intermediate density of the image is expressed on average. ing.

Now, the black and white pattern after binarization is 1. A4. This is possible by changing the numerical arrangement of the l1ilN table. If the black and white pattern has periodicity, it is sufficient to prepare one period as a unit matrix as a threshold value.

On the other hand, to create a black and white pattern with no periodicity, such as a grain pattern, there is a method of generating random numbers at each pixel position and using them as the darkness value of that pixel. However, in a simple method of generating random numbers, for example, by comparing a threshold value created with an M-series pseudo-random signal and a signal obtained by scanning one original image, it is difficult to generate a black and white pattern in the main scanning direction of the image. It has the disadvantage that it does not have a 57th pattern with no direction like grains created by an optical method using a contact screen.

OBJECTS OF THE INVENTION In view of the above-mentioned drawbacks, it is an object of the present invention to obtain a two-spot and one-spot image that has a grain-like feel during digital processing of both f-segment data.

Structure of the Invention The present invention consists of the first aspect. A second random signal is generated, and a temporal distribution value of the first random signal is selected from a grain distribution table representing a probability distribution of the grain size. The second random signal is changed according to the timing, and the value of the second random signal is calculated as a threshold value arranged two-dimensionally on the image plane. The above objective is achieved by binarizing the .

The resulting invention is the first. random signal generating means for generating a second random signal; storage means for storing a grain particle distribution table representing a probability distribution of the size of grain particles; means for changing the second random signal at the temporal timing of the distribution value selected from the storage means in accordance with the above, and the values of the second random signal are arranged two-dimensionally on the image plane. The above objects are achieved by both signal processing apparatuses, which are equipped with, as a threshold value, a binarization means for binarizing a human-powered image signal obtained by decomposing an original image @full scan.

DESCRIPTION OF EMBODIMENTS An embodiment of the present invention will be described below with reference to the drawings.

FIG. 2 shows a block configuration of an image signal processing device in one embodiment of the present invention.

In Fig. 2, 1 is a gate circuit which has two terminals and receives all inputs of a 16 times pixel clock with the pixel clock frequency -
Ail ii outputs a 16 times pixel clock for one pixel section every 4 pixel clocks. Reference numerals 3 to 6 designate 16-bit offset registers which use the signal sent from the gate circuit 1 as a clock, and are initialized by a sub-scanning synchronizing signal sent through the terminal 7.

8 to 11 take binary information of ○ or -1.
In one pass, data is written to each of the shift registers 3 to 6 in response to one sub-scanning synchronization word sent through the terminal 7. 12 is a 2-bit counter that counts the sub-scanning signal sent through terminal 7; 13 is counter 1;
2 is a decoder that outputs a type 1 signal, 14 is a gate circuit that gates the outputs of the registers 3 to 6 as a gate signal for the decoder 13, and 15 (1 gate circuit). 16 is a 2-bit counter that operates on the output of counter 14 and is reset by the sub-scanning signal sent through terminal 7.
A decoder 17 outputs four types of signals according to the signal output from the gate circuit 1; 17 is a threshold direct data generator 1 means for outputting threshold data according to the signal sent from the gate circuit 1; 02 is a latch circuit that latches the threshold value data sent by the dark value data generation means 17.
2 is a 2-bit counter that counts the pixel clock sent through the terminal 23, and is reset by the sub-scanning synchronization signal sent out through the terminal 7'ff. 2
4 is a decoder that outputs all four types of signals according to the output of the counter 22;
26 is a latch circuit 1 according to the gate signal of the decoder 24.
Gate circuit that gates information 8 to 21, 26 is terminal 2
This is a latch circuit that outputs information sent from the gate circuit 26 to a terminal 27 in response to a pixel clock sent through the gate circuit 3. 28 is an ANp circuit which takes the logical sum of the signal sent out from the decoder 13 and the sub-scanning synchronization signal sent out through the terminal 7'f; 29 is the inversion circuit 30 (the signal sent out from the decoder 13 through 5 and the terminal 0, which is an AND circuit that takes the logical sum of the sub-scanning synchronization signals sent through 7. Hereinafter, the specific configuration of the dark value data generation means 17 will be explained with reference to FIG.

In FIG. 3, numeral 31 is a random signal generating circuit that generates a random signal every clock pulse sent from the gate path 1. 32 shows the grain particle distribution table's Tsuru value Kinki 1: ti 1. It's a device, uh/
U33, which outputs the number sent from the gate circuit 31 and set to L at address 6, subtracts the value by one for each clock pulse sent from the gate circuit 1 shown in FIG. The subtraction circuit inputs the numerical value sent from the storage device 32 when the subtraction result becomes 0, and also updates the contents of the random signal generation circuit 34. 1. '
A random signal is output as a 4+匝 signal.

In the above configuration, the operation of the threshold data generation means 17 shown in FIG. 3 will be explained first.

1. For each clock pulse sent out through the gate circuit 1, p1 is reduced in the interrogation path 33? ; is multiplied by $, and on the other hand, the contents of the random signal generation circuit 31 are updated. Next, the storage device 32 outputs all the numerical values of the "grained particle distribution tape" according to the address indicated by the contents of the random signal generating circuit 31.

When the inner world of the subtraction circuit 33 reaches the conclusion of the subtraction, the random signal generation circuit 34 is instructed to update the contents. At the same time, the output data of the storage device 32 is taken into the subtraction circuit 2. Then, the random signal generation circuit 340 Uchitani uses @2 as a threshold signal for binarizing the image signal.
It is output to the launch circuits 18 to 21 shown in the figure. Therefore, since the 1 factor 1 direct error signal sent to the launch circuits 18 to 21 continues to have the same value by the size of the numerical value output from the storage device 32, the average continuation length of the same threshold value is
It can be considered that the number stored in
If there are 4 pieces of , 16 pieces of 2, 11 pieces of 4, and 2 pieces of 8, then (IX4+2X15+4X11+8X2)÷(4+15
+11+2''92.94, and the average continuation length is approximately 2.94.If this value is thicker, the grain grains can be made larger, and if this value is smaller, the grain grains can be made smaller. The operation of the apparatus shown in FIG. 2 will be explained with reference to FIGS. 4 and 6.

In addition, Figure 4 (: mouth 5 (j lli k) shows the outline when arranged two-dimensionally, and the operation below is shown in 4 x 4 according to Figure 3 -1 to 16. 16 (Ljbl I direct output data 11 sent from the data generation means)
shall be distributed and expanded over the entire screen.

1. FIG. 6 is a timing diagram of the main signals in the configuration of FIG. 2. 51st Evil (8 Genus, 116x pixel clock,
Fig. 4 (CB) is the pixel clock, Fig. 5 (q is the gate pulse of gate circuit 1, Fig. 5 (D) - (q - is the output pulse of the nft registers 3 to 6, respectively, and is an example above the valley nocils. The numbers are shown in Figure 4 ((2) 6 for the numbers 1 to 16 shown.
It is something to do.

First, for each sub-scanning synchronization signal marked 7Jt+ on the terminal 7, the data held in the respective data circuits 8 to 11 is put into the shift registers 3 to 6, and each shift register 3
The contents of ~6 are initialized. On the other hand, a one-pixel lock signal 16 times the pixel clock frequency is applied to the terminal 1 and sent to the gate circuit 1. One pixel section is sent to the shift register 3 to 6 degrees and the threshold data generation means 1 for every 1 to 4 pixel clocks to the gate circuit to which all 16 times pixel clocks are input. Each of the shift registers 3 to 6 uses the output of the gate circuit 1 as a shift clock (while looping the data, an output pulse train is generated at the timing shown in FIG. 6).

(Threshold value data obtained at the timing of the pulse of q is finally arranged as shown in FIG. 4 and becomes ε0.) Now, the 2-bit counter 12 is connected to the terminal. Counts the sub-scanning synchronization signal input from 7 and repeatedly outputs 00, 01, 10°11. Note that this counter 12 is cleared at the beginning of scanning and becomes oO. This output causes the decoder 13 to output four signals ' oo + Ho1rH1o, Hll (the subscript corresponds to the counter contents, and in that state it is a spear, and the others are 0).Then, the gate circuit 14 generates this H8° to H11.
The signal is used as a gate pulse, and each output signal of the shift registers 3 to 6 is gated. When 0 is 1, the output of the foot register 3 is output. (When 11 is 1, the output of the /ft register 6 becomes the output of the gate circuit 14.
counts the output of this gate circuit 14 and outputs the count information to the total decoder 16. and decoder 16Vi
The output of the counter 15 is decoded, and the decoded signals C0°, Co1
．． C4° and C11 (however, the subscript corresponds to 71 according to the contents of the counter, and is 1 in that state and ◯ in other cases). Then, using this decoded signal C6O''011C101C11 as a control signal, the dark value data sent from the threshold generation means 17 is selectively taken into the latch 1 channels 18 to 21. That is, the threshold 111 data is output from the decoder 16 as the output C60. is 1
When Co1 is 1, it is sent to latch circuit 1B, when Co1 is 1, it is sent to latch circuit 19, and when C1° is 1, it is sent to latch circuit 2o.
On the other hand, when C11 is 1, the latch circuit 21 takes in 1.

On the other hand, the decoder 24 outputs the decoded signal P according to the count value of the counter 22 which counts the pixel clocks displayed at the terminal 23. . , Pol, Pl. .

P (however, the subscript corresponds to the counter contents, and is 1 in the 1 state and 0 in other states) is output to the gate circuits 1 and 25. The gate circuit 31Q1 selectively inputs and outputs the information held by the latch circuits 18 to 21 using the decoded signals "0OIP01.plo and pll sent out by the decoder 24 as control signals. That is, the gate circuit 26
4 output decoded signal P. .

When Pol is 1, the contents of the latch circuit 18 are determined, and when Pol is 1, the contents of the latch circuit 19 are determined as Pl. When P11 is 1, the contents of the launch circuit 20 are output, and when P11 is 1, the contents of the launch circuit 21 are output. The latch circuit 26 takes in the output of the gate circuit 25 at the timing of the pixel clock applied to the terminal 23, and outputs it to the output terminal 27 as final threshold data. This final threshold data is used to binarize the image signal. Note that among the output signals of the decoder 24, P0° becomes the gate signal of the gate circuit 1. 1, decoder 1
Of the output signals of 3, H8° and the product 11 scanning synchronizing signal applied to the terminal 7 are sent to the AND circuit 28 (11-valve logic interest taking terminal 28a), and Ho. The scanning synchronization signal is logically summed through the AND circuit 37 and output to the terminal 29a.
The signal stores the contents of the random signal generation circuit 31, the contents of the subtraction circuit, and the contents of the register 34 shown in FIG. 3 in the memory 2 (not shown). On the other hand, terminal 2
The signal 9a is used as a timing signal for restoring the stored contents. This is to operate the circuit shown in the 31st line for each of the four sub-scanning lines with the same data content.

Effects of the Invention The present invention has the following features as described above. A second random signal is generated, and the second random signal is activated at the timing indicated by the numerical values in the grain particle distribution table that are successively selected in response to one turn of the first random signal. Italy, by binarizing the input image signal as darkness values that are two-dimensionally arranged on the image plane, the binarized image has no directionality. It is possible to obtain an image with a granular pattern that feels like sand grain, and the size of the grain particles is also a table constant 't'! In other words, it can be controlled very much, and its value is great.

[Brief explanation of the drawing]

1(A) to q are schematic diagrams for explaining the dither method, FIG. 2 is a block diagram of an image signal processing device in one embodiment of the present invention (9th stage), and FIG. 3 is a schematic diagram of the same device. Main part block connection diagram, Figure 4 i1', diagram in which one threshold is arranged two-dimensionally,
Figure 5 (8~(G) is a timing chart of the same device. 1.14.25...Gate circuit, 2,7,27°28a, 29a...Terminal, 3~ 6...Shift register, 8-11... Data circuit, 12, 15, 2
2゜・・・Counter, 13, 16.24...
Decoder, 14.25... Gate circuit, 17.
...Threshold data generation means, 18 to 21.26...
- Launch circuit, 31.34...Random signal generation circuit, 32...Storage device, 33...Subtraction circuit. , Name of agent: Patent attorney Toshio Nakao and one other person Figure 1

Claims (2)

[Claims] (1) First. A second random signal fj is generated, and the temporal timing of a distribution signal selected according to the value of the first random signal from a grain particle distribution table representing a probability distribution of the size of grain particles. The value of the second random signal is changed to an input image signal obtained by scanning and decomposing the original image as a threshold value arranged two-dimensionally on several planes. An image signal processing method that binarizes and obtains a grain-like image signal. (2) First. a random signal generating means for generating a second random signal; a storage means for storing a grain particle distribution table representing a probability distribution of the size of grain particles; means for changing the second random signal according to the temporal timing of the distribution value selected from the storage means; As a value, both input signals obtained by decomposing the original image @ full scan are binarized 2
Both signal processing devices are provided with value conversion means.